JP2006293987A - Character input device, document creation device, character input method, character input program, and computer-readable recording medium recording character input program - Google Patents

Abstract

To efficiently perform character conversion.
A character input by a character input key is accepted by a character input device (S141), and an arbitrary character string is an input character string for an input character string composed of one or more accepted characters. Based on an erroneous input evaluation function that evaluates the possibility of being erroneously input to the character string, a character string that satisfies an evaluation value of the erroneous input evaluation function as a predetermined threshold is recognized as an erroneous input character string (S51 to 59), and accepted. The input character string thus recognized and the recognized erroneously input character string are converted to Kanji, a Kanji conversion character string candidate is created (S63), a list of the created Kanji conversion character string candidates is displayed, and the displayed Kanji conversion character is displayed. The selection of the kanji conversion character string candidate intended by the user is accepted from the column candidates (S64).
[Selection] Figure 37

Description

  The present invention relates to a character input device, a document creation device, a character input method, a character input program, and a computer-readable recording medium on which a character input program is recorded, and in particular, a character input device suitable for kana-kanji conversion and document creation The present invention relates to an apparatus, a character input method, a character input program, and a computer-readable recording medium on which the character input program is recorded.

  Conventionally, as a method of converting a kana character string input by a user into a kana / kanji character, a conversion candidate for the input kana character string is displayed, and a character string including a kanji character as a conversion result intended by the user is selected from the conversion candidates. There was a conversion method. In such a conversion method, when the user erroneously inputs the kana character string, the user himself must correct the erroneously input kana character string.

  Some kana-kanji conversion systems of recent document creation devices automatically perform corrections such as consonant excess correction, vowel excess / deficiency correction, and irregular romaji correction in a romaji input mode using a keyboard.

  Further, in Patent Document 1, when a user has pressed a key next to a key that the user originally intended to press due to a key input mistake on the keyboard, the key is replaced not only with the pressed key but also with the adjacent key. There has been disclosed a character input device that displays a kana-kanji converted character string generated from a keystroke string as a homophone word candidate.

Patent Document 2 discloses a character input device that has a plurality of keys, a plurality of characters are assigned to each key, and an input character is determined by the number of times the key is pressed. This character input device is physically close to the key to which the erroneously input character is assigned as the correct input character candidate when the erroneously input character of the character string input by the key is corrected, and the number of presses is the same. Select what to do.
JP-A-9-319739 JP 2004-185373 A

  In the conventional document creation apparatus as described above, in order to completely correct a user's key input mistake, the user's own correction process is required. This is very troublesome for the user and leads to wasted time.

  The present invention has been made to solve the above-described problems, and one of the objects of the present invention is to input a character that can efficiently replace the input character string with a preferable character string. An apparatus, a document creation device, a character input method, a character input program, and a computer-readable recording medium on which the character input program is recorded.

  Another object of the present invention is to provide a character input device, a document creation device, a character input method, a character input program, and a computer-readable recording medium on which the character input program is recorded that can efficiently perform kana-kanji conversion. Is to provide.

  In order to achieve the above object, according to one aspect of the present invention, a character input device includes a character input unit that receives a character input by a character input key, and one or more characters received by the character input unit. Evaluation using the replacement evaluation function as an index for replacing the input character string with the replacement candidate character string, and calculating the evaluation value of the replacement candidate character string for the input character string based on the replacement evaluation function A calculation unit, and a replacement character string recognition unit that recognizes a replacement candidate character string whose evaluation value calculated by the evaluation calculation unit satisfies a predetermined condition as a replacement character string for replacing the input character string.

  According to this invention, a character input by a character input key is accepted by a character input device, and an input character string made up of one or more accepted characters is targeted for a replacement candidate character string. A replacement evaluation function is used as an index for replacement. Based on the replacement evaluation function, an evaluation value of the replacement candidate character string for the input character string is calculated, and a replacement candidate character string that satisfies the predetermined evaluation condition satisfies the predetermined evaluation value. It is recognized as a replacement character string that replaces the input character string. For this reason, the replacement candidate character string is a character string whose evaluation value, which is an index for replacing the input character string with the replacement candidate character string, satisfies a predetermined condition, and therefore, a preferable character string may be recognized as a replacement character string. Becomes higher. As a result, it is possible to provide a character input device that can efficiently replace an input character string with a preferable character string.

  Preferably, the replacement evaluation function is a function for evaluating the possibility that the input character string is an erroneous input with respect to the replacement candidate character string.

  According to this invention, the replacement candidate character string is a character string whose evaluation value that evaluates the possibility that the input character string is erroneously input to the replacement candidate character string satisfies the predetermined condition. The possibility of being recognized as a replacement character string increases. As a result, it is possible to provide a character input device that can efficiently replace an input character string with a correct character string.

  Preferably, the replacement evaluation function is a function that gives a higher evaluation value to the replacement candidate character string as the input character string is more likely to be erroneously input. The replacement character string recognition unit recognizes a replacement candidate character string whose evaluation value exceeds a predetermined threshold as a replacement character string for replacing the input character string.

  According to the present invention, since the replacement candidate character string whose evaluation value exceeds the predetermined threshold is a character string that is highly likely to be erroneously input to the input character string, the correct character string is recognized as the replacement character string. Is even more likely. As a result, it is possible to provide a character input device that can more efficiently replace an input character string with a correct character string.

  Preferably, the replacement evaluation function is a function that gives a higher evaluation value to the replacement candidate character string as the input character string is more likely to be erroneously input. The replacement character string recognition unit recognizes a predetermined number of replacement candidate character strings as replacement character strings for replacing the input character strings in descending order of evaluation values.

  According to this invention, since the predetermined number of replacement candidate character strings are likely to be erroneously input to the input character string in descending order of evaluation value, the correct character string is recognized as the replacement character string. Is even more likely. As a result, it is possible to provide a character input device that can more efficiently replace an input character string with a correct character string.

  Preferably, when the input character string and the replacement candidate character string recognized as the replacement character string are Japanese, the input character string and the replacement character string are converted from kana to kanji and converted character string candidates. Is further provided.

  According to the present invention, there is a high possibility that a preferable character string is converted from kana to kanji. As a result, it is possible to provide a character input device that can efficiently convert an input character string into a preferable character string.

  Preferably, if the input character string and the replacement candidate character string recognized as the replacement character string are in Korean, the input character string and the replacement character string are converted to Korean characters by converting Hangul to Kanji. Is further provided.

  According to the present invention, there is a high possibility that a preferable character string is converted to Korean characters. As a result, it is possible to provide a character input device that can efficiently convert an input character string into a preferable character string.

  Preferably, when the input character string and the replacement candidate character string recognized as the replacement character string are in Chinese, the input character string and the replacement character string are converted into Pinyin vs. Kanji and converted character string candidates Is further provided.

  According to the present invention, there is a high possibility that a preferred character string is converted to Pinyin vs. Kanji. As a result, it is possible to provide a character input device that can efficiently convert an input character string into a preferable character string using Pinyin vs. Kanji.

  Preferably, the character input device includes a conversion character string candidate display processing unit that displays a list of conversion character string candidates created by the character conversion unit, and a conversion character string candidate displayed by the conversion character string candidate display processing unit. And a conversion character string selection receiving unit that receives selection of the conversion character string.

  According to the present invention, the character input device displays a list of the created conversion character string candidates and accepts selection of the conversion character string from the displayed conversion character string candidates. For this reason, a user can be made to select a conversion character string efficiently.

  Preferably, the replacement evaluation function inputs a character input key that is keyed to input a character in the character input unit and a character corresponding to the character in the replacement candidate character string for the input character string including the character. It is a function which gives the evaluation value according to the physical distance with the character input key for.

  According to the present invention, a character input key that is keyed to input a character by the character input device and a character corresponding to the character in the replacement candidate character string for the input character string including the character are input. An evaluation value corresponding to the physical distance from the character input key is given. There is a high possibility that an erroneous input is performed by a character input key at a distance close to the correct character input key. For this reason, a character string whose evaluation value corresponding to the physical distance of the character input key satisfies a predetermined condition is a character string that is more likely to be erroneously input to the input character string. As a result, the input character string can be replaced with a more preferable character string.

  Preferably, the replacement evaluation function inputs a character input key that is keyed to input a character in the character input unit and a character corresponding to the character in the replacement candidate character string for the input character string including the character. It is a function which gives the evaluation value according to the direction with the character input key.

  According to the present invention, a character input key that is keyed to input a character by the character input device and a character corresponding to the character in the replacement candidate character string for the input character string including the character are input. An evaluation value corresponding to the direction of the character input key is given. There is a correlation between the direction of the erroneously input character input key from the correct character input key and the probability of the erroneous input. For this reason, a character string whose evaluation value corresponding to the direction of the character input key satisfies a predetermined condition is a character string that is more likely to be erroneously input to the input character string. As a result, the input character string can be replaced with a more preferable character string.

  Preferably, the character input unit accepts input of one character from a plurality of characters assigned to the character input key according to the number of times the character input key is pressed, and the replacement evaluation function inputs the character in the character input unit. Therefore, an evaluation value corresponding to the difference between the number of times of keystroke of the character input key that has been keyed and the number of times of keystroke for inputting the character corresponding to the character in the replacement candidate character string for the input character string including the character This is the function to give.

  According to the present invention, the number of times of the character input key that is keyed to input a character by the character input device and the character corresponding to this character in the replacement candidate character string for the input character string including this character are obtained. An evaluation value corresponding to the difference from the number of keystrokes for input is given. When one character is input from a plurality of characters assigned to the character input key according to the number of times the character input key is pressed, there is a high possibility that an erroneous input is made based on the number of key inputs. For this reason, a character string whose evaluation value according to the difference in the number of keystrokes satisfies a predetermined condition is a character string that is more likely to be erroneously input to the input character string. As a result, the input character string can be replaced with a more preferable character string.

  Preferably, the replacement evaluation function is capable of adding to the character the character received in the character input unit and the character corresponding to the character in the replacement candidate character string for the input character string including the character. It is a function that gives an evaluation value according to the difference of symbols.

  According to the present invention, the symbol that can be added to the character by the character input device and the character corresponding to the character in the replacement candidate character string for the input character string including the character. An evaluation value corresponding to the difference is given. Since symbols that can be added to characters tend to be relatively small and overlooked, there is a high possibility that symbols that can be added to characters will not be entered or input incorrectly. For this reason, a character string in which an evaluation value corresponding to a difference in symbols that can be added to characters satisfies a predetermined condition is a character string that is more likely to be erroneously input to the input character string. As a result, the input character string can be replaced with a more preferable character string.

  Preferably, when the character received in the character input unit is Japanese, the replacement evaluation function calculates the character received by the character input unit and the character in the replacement candidate character string for the input character string including the character. This is a function that gives an evaluation value according to the difference between a muddy note or a semi-voiced note from the corresponding character.

  According to the present invention, when the received character is Japanese by the character input device, the received character and the character corresponding to this character in the replacement candidate character string for the input character string including this character. An evaluation value corresponding to the difference between a muddy note or a semi-voiced note is given. Since muddy notes or semi-muddy notes tend to be relatively small and overlooked, there is a high possibility that muddy notes or semi-muddy notes will not be entered or entered incorrectly. For this reason, a character string in which an evaluation value corresponding to a difference between a muddy note or a semi-voiced note satisfies a predetermined condition is a character string that is more likely to be erroneously input to the input character string. As a result, the input character string can be replaced with a more preferable character string.

  Preferably, the replacement evaluation function inputs a character input key that is keyed to input a character in the character input unit and a character corresponding to the character in the replacement candidate character string for the input character string including the character. An evaluation value according to a physical distance from the character input key for the character input, a character input key keyed to input a character in the character input unit, and the replacement candidate character string in the replacement character string for the input character string including the character An evaluation value corresponding to the direction of the character input key for inputting the character corresponding to the character, the number of times the character input key was pressed to input the character in the character input unit, and the input character including the character An evaluation value corresponding to the difference in the number of keystrokes for inputting a character corresponding to the character in the replacement candidate character string for the character string, a character received in the character input unit, and an input including the character An evaluation value calculated on the basis of at least two evaluation values among evaluation values according to the difference in symbols that can be added to the character corresponding to the character in the replacement candidate character string for the character string This is the function to give.

  According to the present invention, a character input key that is keyed to input a character by the character input device and a character corresponding to the character in the replacement candidate character string for the input character string including the character are input. Evaluation value according to the physical distance from the character input key, the character input key that was typed to input the character, and the character corresponding to this character in the replacement character string for the input character string that contains this character The evaluation value according to the direction of the character input key for inputting the character, the number of times the character input key is keyed to input the character, and the replacement candidate character string for the input character string containing this character The evaluation value corresponding to the difference in the number of keystrokes for inputting the character corresponding to the character, the received character, and the character in the replacement candidate character string for the input character string including this character Evaluation value calculated based on at least two evaluation value among the evaluation values corresponding to the symbols of the differences that can be added to the character of the character is given. As a result, the input character string can be replaced with a more preferable character string.

  Preferably, the erroneous input evaluation function inputs a character input key that is keyed to input a character in the character input unit and a character corresponding to the character in the replacement candidate character string for the input character string including the character. An evaluation value corresponding to a physical distance from the character input key for the character input, a character input key keyed to input a character in the character input unit, and a replacement candidate character string for the input character string including the character Evaluation value corresponding to the direction of the character input key for inputting the character corresponding to the character, the number of times the character input key is pressed to input the character in the character input unit, and the input including the character The evaluation value corresponding to the difference from the number of keystrokes for inputting the character corresponding to the character in the replacement candidate character string for the character string, the character received in the character input unit, and the character are included A function that gives an evaluation value calculated by weighting each of the evaluation values according to the difference in symbols that can be added to the character corresponding to the character in the replacement candidate character string for the force character string is there.

  According to the present invention, a character input key that is keyed to input a character by the character input device and a character corresponding to the character in the replacement candidate character string for the input character string including the character are input. An evaluation value corresponding to the physical distance from the character input key, a character input key keyed to input a character, and a character corresponding to this character in the replacement candidate character string for the input character string including the character The evaluation value according to the direction of the character input key for inputting the character, the number of times the character input key is keyed to input the character, and the replacement candidate character string for the input character string containing this character Corresponds to this character in the replacement candidate character string for the received character and the input character string that includes this character, and the evaluation value according to the difference in the number of keystrokes for inputting the character corresponding to the character That evaluation value calculated by weight the respective evaluation values corresponding to the symbols of the differences that can be added to the character of the character is given. As a result, the incorrect input due to the physical distance of the character input key, the incorrect input of the direction of the character input key, the incorrect input due to the number of keystrokes, and the erroneous input of the symbol that can be added to the character are evaluated. The influence can be adjusted according to the user's habit or the like.

  Preferably, the character input device updates the weight according to a difference between the character string before conversion to the conversion character string candidate received by the conversion character string selection reception unit and the input character string received by the character input unit. A weight updating unit for performing the processing.

  According to this invention, the weight is updated by the character input device according to the difference between the input character string and the character string before conversion to the selected conversion character string candidate. For this reason, since the weight is dynamically updated according to the user's habit or the like, the conversion suitable for the user can be performed every time the weight is updated.

  Preferably, the conversion character string candidate display processing unit displays the conversion character string candidates in a list in the descending order of evaluation value of the replacement evaluation function. For this reason, a user can select a conversion character string candidate more efficiently.

  Preferably, the conversion character string candidate display processing unit includes the conversion character string candidate converted from the input character string by the character conversion unit, and the replacement character string recognized by the replacement character string recognition unit in the conversion character string candidate, A list of conversion character string candidates is displayed, and when a conversion character string selection is accepted by the conversion character string selection receiving unit, a list of conversion character string candidates converted from the replacement character string by the character conversion unit is displayed.

  According to this invention, the conversion character string candidate converted from the input character string and the recognized replacement character string are included in the conversion character string candidate, the conversion character string candidates are listed, and the conversion character string can be selected. If accepted, a list of conversion character string candidates converted from the replacement character string is displayed. For this reason, the conversion character string candidate converted from the input character string and the conversion character string candidate converted from the replacement character string are displayed step by step, so that the user can select the conversion character string candidate more efficiently. Can be made.

  Preferably, a replacement character string candidate display processing unit that displays a list of replacement character string candidates recognized by the replacement character string recognition unit, and a replacement character string from among the replacement character string candidates displayed by the replacement character string candidate display processing unit A replacement character string selection receiving unit that receives a column selection is further provided.

Preferably, the document creation device includes a character input device.
Preferably, the character creation device includes a character input device whose character input unit is a remote control. According to the present invention, it is possible to provide a character creation device capable of efficiently replacing an inputted character string with a preferable character string.

  Preferably, the character input method replaces an input character string with a replacement candidate character string for a step of receiving a character input by a character input key and an input character string including one or more accepted characters. Using a replacement evaluation function as an index, and calculating a replacement candidate character string evaluation value for the input character string based on the replacement evaluation function, and a replacement candidate character string for which the calculated evaluation value satisfies a predetermined condition as an input character And certifying a replacement character string to be substituted for the column.

  According to the present invention, it is possible to provide a character input method capable of efficiently replacing an input character string with a preferable character string.

  Preferably, the character input program replaces the input character string with a candidate replacement character string for the step of accepting the character input with the character input key and the input character string composed of the accepted one or more characters. Using a replacement evaluation function as an index, and calculating a replacement candidate character string evaluation value for the input character string based on the replacement evaluation function, and a replacement candidate character string for which the calculated evaluation value satisfies a predetermined condition as an input character And causing the computer to execute a step of recognizing the replacement character string to be replaced for the column.

  Preferably, the character input program is recorded on a computer-readable recording medium. According to the present invention, it is possible to provide a character input program capable of efficiently replacing an input character string with a preferable character string and a computer-readable recording medium on which the character input program is recorded.

  Embodiments of the present invention will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

  First, an outline of the embodiment of the present invention will be described. Although there are various character input devices such as keyboards for personal computers and character input for mobile phones, an input error can be said to be a simple careless error.

  In character input of a cellular phone, since there are few input keys, a plurality of characters are assigned to the same key, and a target character can be input by pressing a plurality of times. For example, on the number “5”, when a character in the N line is assigned and “NO” is entered, the “5” key must be pressed five times. However, if a large number of input keys are accidentally typed or a small number of keys are typed, characters different from the intended input characters will be entered.

  In addition, in character input of a character input device in general including a keyboard, there is a case where a careless mistake that a user intends to press an input key and accidentally presses a nearby key is committed.

  In this way, more or less anyone can commit a careless mistake, which is a mistake in typing a key. Therefore, in the document creation apparatus of the present invention, an algorithm for automatically correcting a careless mistake of a user in the character input apparatus is incorporated in the conversion rule for kana-kanji conversion.

[First Embodiment]
FIG. 1 is a block diagram showing an outline of the configuration of a document creation apparatus 100 according to an embodiment of the present invention. Referring to FIG. 1, document creation apparatus 100 is configured by a general-purpose portable terminal such as a mobile phone or a PDA (Personal Digital Assistance).

  The document creation apparatus 100 includes a control unit 101 for controlling the entire document creation apparatus 100, a storage unit 120 for storing predetermined information, and an input unit for inputting predetermined information to the document creation apparatus 100. 130, an output unit 140 for outputting predetermined information from the document creation device 100, a wireless communication unit 170 for wirelessly communicating with an external server or portable terminal, and voice input to the document creation device 100 A voice input / output unit 180 that outputs voice from the document creation apparatus 100.

The control unit 101 includes an MPU (Micro Processing Unit) and an auxiliary circuit of the MPU, and controls the storage unit 120, the input unit 130, the output unit 140, the wireless communication unit 170, and the voice input / output unit 180, and the storage unit 120. A predetermined process is executed in accordance with the program and data stored in the data, the data input from the input unit 130, the wireless communication unit 170, or the voice input / output unit 180 is processed, and the processed data is stored in the storage unit 120. Or output to the output unit 140, the wireless communication unit 170, or the voice input / output unit 180.

  The storage unit 120 includes a RAM (Random Access Memory) used as a work area necessary for the control unit 101 to execute a program, and a ROM (Read Only Memory) for storing a program to be executed by the control unit 101. Including. In addition, a program and data for executing predetermined processing are read from the wireless communication unit 170 and the input unit 130 and stored in the RAM. Furthermore, an external storage device using a memory card as a recording medium may be used as an auxiliary storage device for assisting the storage area of the RAM.

  The input unit 130 is an interface for inputting a signal from an operation key, and can input necessary information to the document creation apparatus 100.

  The output unit 140 is an interface for outputting a signal to a display such as a liquid crystal display device (LCD, Liquid Crystal Display), and can output necessary information from the document creation device 100.

  The wireless communication unit 170 communicates with other mobile terminals and servers wirelessly via the antenna 171. The wireless communication unit 170 is controlled by the control unit 101 to transmit predetermined information to another mobile terminal or server, or receive information from another mobile terminal or server.

  The voice input / output unit 180 inputs the voice input from the microphone of the portable terminal to the control unit 101 and outputs the voice signal from the control unit 101 from the speaker of the portable terminal.

  FIG. 2 is a functional block diagram of the document creation device 100 according to the embodiment of the present invention. Referring to FIG. 2, the control unit 101 of the document creation apparatus 100 includes a character input unit 102, an erroneous input candidate recognition unit 103, a kana / kanji conversion unit 104, and a kana / kanji conversion determination unit 105.

  Hereinafter, each component in the document creation apparatus 100 will be described in detail. The document creation apparatus 100 is fully equipped with some character input device connected to the input unit 130. This character input device inputs characters by pressing a general key, or a screen such as a touch panel. Any character can be used as long as the character is assigned to the key regardless of the form, such as inputting a character by touching the key.

  The character input unit 102 acquires the kana character string input by the user through the character input device as the input kana character string when the user prompts the conversion instruction. The opportunity for the character input unit 102 to acquire the input kana character string is not limited to the time when the user prompts the conversion instruction, but may be the time when the user inputs a predetermined number of characters, or every predetermined period. There may be.

  Next, the erroneous input candidate recognition unit 103 is based on an erroneous input evaluation function that evaluates the possibility that an arbitrary input character string is erroneously input into the input input character string for the acquired input input character string. Thus, a new kana character string whose evaluation value of the erroneous input evaluation function satisfies a predetermined threshold is created, and the kana character string is stored in the erroneous input kana character string candidate list as an erroneous input kana character string.

  Here, referring to Table 1, there is an erroneous input kana character string candidate list used in the erroneous input candidate recognition unit 103 of the document creation apparatus 100 in the embodiment of the present invention.

  The erroneous input kana character string candidate list is for holding one or more erroneous input kana character string candidates as described above. Each erroneously input kana character string holds two elements of “evaluated input kana character string information”: an evaluation value evaluated based on the kana character string and an incorrect input evaluation function.

  Further, the input kana character string acquired by the character input unit 102 is included in the erroneous input kana character string candidate list as one candidate of the erroneous input kana character string.

  Next, the kana-kanji conversion unit 104 performs kana-kanji conversion processing for each kana character string held in the acquired erroneous input kana character string candidate list, and creates a kana-kanji conversion candidate list to be presented to the user.

  Next, the kana-kanji conversion determination unit 105 displays a list of kana-kanji conversion candidates created by the kana-kanji conversion unit 104 based on the evaluation value of the erroneous input evaluation function, and the character string selected by the user is displayed. Adopted as the conversion result and moves to the next document creation.

  FIG. 3 is a functional block diagram showing details of the erroneous input candidate recognition unit 103 of the document creation apparatus 100 according to the embodiment of the present invention.

  Referring to FIG. 3, the erroneous input candidate recognition unit 103 includes a processing control unit 301, a keystroke location erroneous input processing unit 302, a keystroke number erroneous input processing unit 303, a keyless non-input processing unit 304, and an erroneous input kana character string candidate. A storage processing unit 305 is provided. Formula 1 is an example of a mathematical formula showing an erroneous input evaluation function used by the erroneous input candidate recognition unit in the embodiment of the present invention.

  Here, with reference to Equation 1, the erroneous input evaluation function F evaluates the kana character string obj with respect to the input kana character string in, and evaluates the evaluation function POS evaluated based on the keystroke location erroneous input process. This is a function based on the three elements of the evaluation function NUM evaluated based on the erroneous input process and the evaluation function SYM evaluated based on the keyless non-input process. The functions POS, NUM, and SYM that evaluate the kana character string obj with respect to the input kana character string in are respectively the n-th input kana character in [ n] is composed of functions pos, num, and sym that evaluate kana characters obj [n].

  Hereinafter, each configuration and processing form of the erroneous input candidate recognition unit 103 will be described in detail. Based on the input kana character string acquired by the character input unit 102 and the erroneous input kana character string candidate list, the process control unit 301 determines a control shift to the next process to be performed. The contents to be controlled are the determination of the kana character string to be processed and the control shift to the keystroke location error input processing unit 302, the determination of the kana character string to be processed and the control shift to the keystroke number error input processing unit 303, These are the determination of the kana character string to be processed and the control shift to the non-key-entry input processing unit 304, the determination of the end of the process in the erroneous input candidate recognition unit 103, and the control shift to the kana-kanji conversion unit 104.

  When the control shifts to the key input location error input processing unit 302, the key input location error input processing unit 302 erroneously inputs the kana character string to be processed based on the evaluation function POS having the key input location error input process as an element. Search for kana character string candidates.

  FIG. 4 is a flowchart showing the flow of an erroneous keystroke location input process executed by configuring the keystroke location erroneous input processing unit 302 in the control unit 101 of the document creation apparatus 100 according to the embodiment of the present invention.

  With reference to FIG. 4, first, the control unit 101 acquires a kana character string “in” to be processed (step S <b> 11). The acquired kana character string in is set as the target kana character string in, and the kana character obj [when the user makes a mistake in the keystroke position for each kana character in [n] constituting the kana character string in [ n].

  The control unit 101 focuses on the first kana character in [1] of the target kana character string in (step S12). If there is a kana character in [n] to be focused here (YES in step S13), the control unit 101 searches for a candidate obj [n] of an erroneously input kana character and searches for obj [ An evaluation value pos (in [n], obj [n]) of the evaluation function pos for n] is calculated (step S14).

  Then, the control unit 101 creates a kana character string obj by substituting the searched kana character in [n] with the searched kana character obj [n] in the target kana character string in (step S15). Next, the control unit 101 creates erroneous input kana character string information including the kana character string obj and the evaluation value pos (in [n], obj [n]) for the kana character string obj, and adds it to the additional candidate list. Store (step S16). Next, the control unit 101 moves the focus to the next kana character in [n + 1] (step S17).

  And the control part 101 repeats the process from step S14 to step S17 with respect to all the kana characters in [n]. On the other hand, if the kana character in [n] to be focused does not exist (in the case of NO in step S13), since all processing has been completed for the target kana character string in, the control unit 101 The created additional candidate list is passed to the erroneous input kana character string candidate storage processing unit 305, which is the next process, and the process proceeds.

  FIG. 5 is a diagram of a specific example of key assignment of kana characters in a mobile phone. Table 2 is a table showing a specific example of assignment of the number of keystrokes of kana characters in a mobile phone. Here, with reference to FIG. 5 and Table 2, the method for searching for an erroneous input kana character candidate in step S14 and the method for the erroneous input evaluation function to evaluate an erroneous input kana character string will be described in detail.

  In the key assignment of each kana character in the mobile phone as shown in FIG. 5 and Table 2, since there are few keys on the mobile phone, one key is not assigned for each character. Etc.) for each key. In the case of such key assignment, it is necessary to hit the key a plurality of times in order to input one character. The first embodiment targets such a character input device.

  When the user hits a key, the user may hit a key arranged at a location different from the key intended as an input error. The key-entry location error input processing unit 302 evaluates this key-entry location error.

  FIG. 6 is a diagram showing an example of an error in the keystroke location by the user in the embodiment of the present invention. Referring to FIG. 6, when the user inputs “NA”, the erroneous input evaluation function is erroneously input due to the positional relationship with respect to a key different from the key to which “NA” is assigned. Assess the possibilities. Here, the kana characters “ka”, “ya”, “ta”, and “ha” assigned to the keys arranged vertically and horizontally are evaluated as equivalent evaluation values.

  Table 3 is a table showing one calculation example of the evaluation value calculated based on the erroneous input evaluation function pos having the keystroke place erroneous input as an element. Here, Table 3 is a calculation example in the case where the evaluation value of the input kana character is 1.0 and 2.0 times is repeated for the keys arranged on the top, bottom, left, and right. The evaluation value becomes larger as the positional relationship distance is larger.

  In addition, as a method for calculating the evaluation value, an example in which the evaluation value is fluctuated in each direction instead of being uniformly multiplied by 2.0 in the vertical and horizontal directions is also conceivable. In the first embodiment, for the sake of convenience, the evaluation value is set to 2.0 times as described above. However, when the evaluation value is defined to be larger as the positional relationship distance is larger, the evaluation value is larger than 1.0. If so, it is not limited to this value.

  Next, when the control shifts to the keystroke number error input processing unit 303, the keystroke number error input processing unit 303 relates to the kana character string to be processed based on the evaluation function NUM based on the keystroke number error input process. To search for candidates for erroneously input kana characters.

  FIG. 7 is a flowchart showing the flow of an erroneous input process of keystrokes executed when the erroneous input process unit 303 is configured in the control unit 101 of the document creation apparatus 100 according to the embodiment of the present invention.

  Referring to FIG. 7, first, control unit 101 acquires a kana character string to be processed (step S21). The acquired kana character string in is set as the target kana character string in, and the kana character obj [when the user makes a mistake in the number of keystrokes for each kana character in [n] constituting the kana character string in. n].

  The control unit 101 focuses on the first kana character in [1] of the target kana character string in (step S22). If there is a kana character in [n] to be focused here (YES in step S23), the control unit 101 searches for a candidate obj [n] of an erroneously input kana character and searches for obj [ An evaluation value num (in [n], obj [n]) of the evaluation function num for n] is calculated (step S24).

  Then, the control unit 101 creates a kana character string obj by replacing the focused kana character in [n] with the searched kana character obj [n] in the target kana character string in (step S25). Next, the control unit 101 creates erroneous input kana character string information including the kana character string obj and the evaluation value num (in [n], obj [n]) for the kana character string obj, and adds it to the additional candidate list. Store (step S26). Next, the control unit 101 moves the focus to the next kana character in [n + 1] (step S27).

  And the control part 101 repeats the process from step S24 to step S27 with respect to all the kana characters in [n]. On the other hand, if there is no kana character in [n] to be focused on (in the case of NO in step S23), since all processing has been completed for the target kana character string in, the control unit 101 The created additional candidate list is passed to the erroneous input kana character string candidate storage processing unit 305, which is the next process, and the process proceeds.

  Here, the method for searching for an erroneous input kana character candidate in step S24 and the method for the erroneous input evaluation function to evaluate an erroneous input kana character string will be described in detail. The assumed key assignment environment for kana characters is the same as in FIG. 5 and Table 2 described above.

  When the user hits the key, the user may hit the key differently from the intended key hit count as an input error. The keystroke count error input processing unit 303 evaluates this keystroke count error.

  FIG. 8 is a diagram showing an example of an error in the number of keystrokes made by the user in the embodiment of the present invention. Referring to FIG. 8, “NA” can be input by pressing a key to which “NA” is assigned once. Following this, “Ni” is repeated twice, “Nu” is repeated three times, “Ne” is continued four times, “No” is continued five times, and “N” is repeated six times again. To go.

  Here, when the user inputs “Nu”, the erroneous input evaluation function evaluates the possibility that the key assigned “Nu” is erroneously input due to the difference in the number of keystrokes. Here, “ni” and “ne” assigned to the number of keystrokes of ± 1 are evaluated as equivalent evaluation values.

  Also, when “NA” is input, the contents of erroneous input differ between “NA” in which the key is pressed once and “NA” in which the key is pressed six times. The former “na” is “0” when no key is pressed and “ni” when no key is pressed and “ni” when the key is pressed twice are candidates for erroneous input kana characters. “NO” and “NI” with 7 keystrokes are candidates for erroneous input kana characters.

  Tables 4 and 5 are tables showing one calculation example of evaluation values calculated based on the erroneous input evaluation function num having the keystroke number erroneous input as an element. Here, Table 4 and Table 5 are calculation examples in the case where the evaluation value of the input kana character is 1.0 and 2.0 times is repeated for the kana character assigned to the number of keystrokes ± 1. In the first embodiment, for the sake of convenience, the evaluation value is set to 2.0 times as described above. However, the value is not limited to this value as long as the value is larger than 1.0.

  Next, when the control shifts to the keyless non-input processing unit 304, the keyless non-input processing unit 304 erroneously inputs the kana character string to be processed based on the evaluation function SYM having the keyless non-input processing as an element. Search for kana character string candidates.

  FIG. 9 is a flowchart showing a non-keystroke input process executed by configuring keyless non-input processing unit 304 in control unit 101 of document creation apparatus 100 according to the embodiment of the present invention.

  With reference to FIG. 9, first, the control unit 101 acquires a kana character string “in” to be processed (step S31). The acquired kana character string in is set as the target kana character string in, and the user has not input or mistakenly input the muffled sound / semi-voiced sound symbol for each kana character in [n] constituting the kana character string in. Search for the kana character obj [n].

  The control unit 101 focuses on the first kana character in [1] of the target kana character string in (step S32). If there is a kana character in [n] to be focused here (YES in step S33), the control unit 101 determines whether the kana character in [n] is a character that can take muddy / semi-voiced sound. (Step S34). Here, if the kana character in [n] that can take muddy / semi-voiced sound (YES in step S34), the control unit 101 searches for a candidate obj [n] of the erroneously input kana character and is searched. Then, the evaluation value sym (in [n], obj [n]) of the evaluation function sym for obj [n] is calculated (step S35).

  Then, the control unit 101 creates a kana character string obj by substituting the searched kana character in [n] with the searched kana character obj [n] in the target kana character string in (step S36). Next, the control unit 101 creates erroneous input kana character string information including the kana character string obj and the evaluation value sym (in [n], obj [n]) for the kana character string obj, and adds it to the additional candidate list. Store (step S37). Next, the control unit 101 moves the focus to the next kana character in [n + 1] (step S38).

  And the control part 101 repeats the process from step S34 to step S38 with respect to all the kana characters in [n]. If the kana character in [n] does not take muddy / semi-voiced sound (NO in step S34), the control unit 101 moves the focus to the next kana character in [n + 1] (step S38). ). On the other hand, if there is no kana character in [n] to be focused on (in the case of NO in step S33), since all processing has been completed for the target kana character string in, the control unit 101 The created additional candidate list is passed to the erroneous input kana character string candidate storage processing unit 305, which is the next process, and the process proceeds.

  Here, the method for searching for an erroneous input kana character candidate in step S35 and the method for the erroneous input evaluation function to evaluate an erroneous input kana character string will be described in detail. The assumed key assignment environment for kana characters is the same as in FIG. 5 and Table 2 described above.

  When the user hits the key, as an input mistake, forgetting to enter the muddy / semi-voiced symbols for the kana characters, or because the characters on the display are small on a small screen, the muddy / semi-voiced marks cannot be distinguished. There is a case of incorrect input. In this way, the non-key-entry processing unit 304 evaluates the error of the dakuten / semi-dakuten.

  FIG. 10 is a diagram showing an example of a dakuten / half-dakuten error by the user in the embodiment of the present invention. Referring to FIG. 10, the kana characters in the line may be accompanied by muddy / semi-voiced sounds. For example, when the user inputs “fu”, the user can input “bu” or “pu” by pressing a key to which a muddy / semi-voice symbol is assigned.

  In addition, kana characters in ka-line, ta-line, and sa-line may be accompanied by muddy sounds. For example, when the user inputs “ki”, the user can similarly input “gi”. As described above, in the work related to the muddy sound / semi-voiced sound, it is possible to delete the muddy / semi-voiced sound by inputting only one or two kana character candidates in the case of erroneous input and by pressing the key. For this reason, the erroneous input evaluation function evaluates the evaluation value as uniform due to erroneous input / non-input of muddy / semi-voiced sound.

  Tables 6 and 7 are tables showing one calculation example of the evaluation value based on the erroneous input evaluation function sym having the keyless input as an element. Here, Table 6 and Table 7 are calculation examples in the case where the evaluation value of the input kana character is 1.0 and the other candidates for the erroneous input kana character are 2.0. In the first embodiment, for the sake of convenience, the evaluation value is set to 2.0 times as described above. However, the value is not limited to this value as long as the value is larger than 1.0.

  The erroneous input kana character string candidate storage processing unit 305 performs an erroneous input based on the additional candidate list created by the keystroke location erroneous input processing unit 302, the keystroke number erroneous input processing unit 303, and the keyless non-input processing unit 304. Incorrect kana character string information is stored in the kana character string candidate list. At this time, only the erroneous input kana character string information whose evaluation value of the erroneous input kana character string in the additional candidate list is smaller than a predetermined threshold is stored in the erroneous input kana character string candidate list.

  In addition, when the erroneous input kana character string candidate held in the additional candidate list is already stored in the erroneous input kana character string candidate list, the smaller evaluation value is updated as the erroneous input kana character string information. When the storage process is completed, the erroneous input kana character string candidate list is sorted in ascending order based on the evaluation value. This is intended to continuously target a character string having the smallest evaluation value.

  When the storage process ends, the process returns to the process control unit 301. If the evaluation value of the erroneous input evaluation function for the kana character string to be processed is larger than the threshold value, the erroneous input candidate recognition unit 103 passes the erroneous input kana character string candidate list to the kana / kanji conversion unit 104. In this way, the document creation apparatus 100 proceeds to the kana-kanji conversion work.

  FIG. 11 is a flowchart showing a flow of main processing executed by the control unit 101 of the document creation apparatus 100 in the embodiment of the present invention.

  Referring to FIG. 11, first, control unit 101 accepts an input of one or more kana character strings at character input unit 102 (step S41). Then, the control unit 101 creates erroneous input kana character string information as an evaluation value 1.0 of the erroneous input evaluation function F for this kana character string (step S42).

  Next, the control unit 101 stores the erroneously input kana character string information in the erroneously input kana character string candidate list, sets the pseudonym character string as a processing target kana character string, performs keystroke location erroneous input processing, keystroke count erroneous input processing, and Whether or not to perform each of the three keyless input processes is set as process execution information (step S43).

  Next, if the evaluation value of the erroneously input kana character string information is lower than a predetermined threshold for the processing target kana character string (in the case of YES in step S51), the control unit 101 determines that the keystroke location is incorrect. Whether or not to perform input processing is determined from the processing execution information (step S52). Here, if it is determined that the key input location error input process is to be performed (YES in step S52), the key input location error input process described with reference to FIG. 4 is performed (step S53).

  After the keystroke location error input process is completed, the control unit 101 sets in the process execution information that the keystroke location error input process for the processing target character string has already been executed, and the erroneous input kana character string information held in the additional candidate list. Is stored in the erroneous input kana character string candidate list. And the control part 101 sorts the inside of an incorrect input kana character string candidate list in an ascending order based on the evaluation value contained in incorrect input kana character string information (step S58).

  Next, when all the processes for the processing target character string are completed (YES in step S59), and there is a next character string in the erroneous input kana character string candidate list (YES in step S61), the control unit 101 designates the next kana character string candidate held in the erroneous input kana character string candidate list as the processing target kana character string (step S62).

  Next, if it is determined that the process to be performed by the erroneous input candidate recognition unit 103 remains from the process execution information (NO in step S59), the control unit 101 performs the keystroke number erroneous input process and the keyless non-input process. The same processing is performed for. If it is determined from the process execution information that the key-entry location error input process is not performed (NO in step S52), and if it is determined that the key-pressing number error input process is performed (YES in step S54), the control unit 101 Performs a keystroke count error input process (step S55).

  Further, it is determined from the process execution information that the key input location error input process is not performed (NO in step S52), it is determined that the key input number error input process is not performed (NO in step S54), and the key is not pressed. If it is determined that the input process is to be performed (YES in step S56), control unit 101 performs a key entry non-input process (step S57). If it is determined from the process execution information that no key-entry non-input process is to be performed (NO in step S56), control unit 101 proceeds to step S59.

  As described above, the processing of the processing target character string ends (in the case of YES in step S59), or the evaluation value for the processing target kana character string of the erroneously input kana character string information exceeds a predetermined threshold value. If the process target character string exists in the erroneous input kana character string candidate list (in the case of YES in step S61), the control unit 101 adds to the erroneous input kana character string candidate list. The next kana character string candidate to be held is set as a processing target kana character string (step S62), and the processing from step S51 to step S59 is repeated for this processing target character string.

  If the next character string to be processed does not exist (NO in step S61), the control unit 101 performs processing for all erroneous input kana character string candidates included in the generated erroneous input kana character string candidate list. Kana-kanji conversion processing is performed (step S63). Next, the control unit 101 outputs data to the output unit 140 to display a list of kana-kanji converted character string candidates and select a converted character string intended by the user (step S64).

  Tables 8, 9, and 10 are tables showing examples of erroneous input candidate recognition in the document creation apparatus 100 according to the embodiment of the present invention. With reference to Table 8, Table 9, and Table 10, in the first embodiment, erroneous input candidate recognition in the case where “Kaiten” is input will be described in detail. Here, the threshold used in the erroneous input evaluation function is 2.0. In addition, the evaluation value calculation methods of the keystroke location error input process, the keystroke frequency error input process, and the keyless key input process are as described above.

  In this example, “ka” key is pressed once, “ta” key is pressed four times, “a” key is pressed twice and “kana” key is input.

  Referring to Table 8, first, an erroneous input kana character string candidate when the keystroke location error input process is performed on the input kana character string is searched. For the first character “ka”, the erroneous input kana character candidates are “a”, “sa”, and “na”. A kana character string formed by replacing this with an input character is “Ate”, “Sase”, and “Nanae”. Similarly, for the second character “te” and the third character “i”, a total of eight candidates are obtained as shown in the left column of Table 8. Since the evaluation values of these erroneous input evaluation functions F are each equal to or less than the threshold value 2.0, they are stored in the erroneous input kana character string candidate list as erroneous input kana character string information.

  Next, an erroneous input kana character string candidate when the keystroke count erroneous input process is performed is searched. For the first character “ka”, the erroneously input kana character candidates are “ki” and “” where nothing is input. The kana character string formed by replacing this with the input character becomes “Kite” and “Tei”. Similarly, for the second character “Te” and the third character “I”, a total of six candidates are obtained as shown in the middle column of Table 8. Since the evaluation values of these erroneous input evaluation functions F are each equal to or less than the threshold value 2.0, they are stored in the erroneous input kana character string candidate list as erroneous input kana character string information.

  Next, an erroneous input kana character string candidate in the case where the key entry non-input processing is performed is searched. For the first character “ka”, the erroneous input kana character candidate is “gaite”. Similarly, for the second character “te” and the third character “i”, a total of two candidates are obtained as shown in the right column of Table 8. Since the evaluation value of the erroneous input evaluation function F is equal to or less than the threshold value 2.0, it is stored in the erroneous input kana character string candidate list as erroneous input kana character string information.

  Here, since there is no evaluation value equal to or less than the threshold value 2.0, the search for erroneous input kana character string candidates is completed, and an erroneous input kana character string candidate list as shown in Table 9 is obtained.

  Here, if the threshold value is set high and the threshold value is 4.0, the same processing is repeated for each erroneous input kana character string candidate. A similar process is performed on “Ate” positioned next to the kana character string that has been processed earlier in the erroneous input kana character string candidate list. Table 10 is an erroneous input kana character string candidate list of intermediate results obtained. Thereafter, the process is repeated in the same manner to create an erroneous input kana character string candidate list that satisfies the threshold.

[Second Embodiment]
Next, when the input device is a kana input mode keyboard, it holds more keys than a mobile phone, so one character is basically assigned to one key. Accordingly, there is no character to be continuously input by inputting the same key, so that it is not necessary to execute the erroneous input process of the key input number among the processes executed by the erroneous input candidate recognition unit 103.

  FIG. 12 is a block diagram illustrating an outline of the configuration of the document creation apparatus 200 according to the second embodiment. Referring to FIG. 12, document creation apparatus 200 is configured with a general-purpose computer such as a personal computer or a workstation.

  The document creation apparatus 200 includes a control unit 101 for controlling the entire document creation apparatus 200, a storage unit 220 for storing predetermined information, and an input unit for inputting predetermined information to the document creation apparatus 200. 230, an output unit 240 for outputting predetermined information from the document creation device 200, and an external storage device 250 for inputting information recorded on the recording medium 251 and recording necessary information on the recording medium 251. And a communication unit 260 for communicating with an external computer via the network 900.

Similar to the control unit 101 of the first embodiment, the control unit 101A includes an MPU (Micro Processing Unit) and an auxiliary circuit of the MPU, and includes a storage unit 220, an input unit 230, and an output unit 2.
40, controls the external storage device 250 and the communication unit 260, executes predetermined processing in accordance with programs and data stored in the storage unit 220, and from the input unit 230, the external storage device 250, or the communication unit 260. The input data is processed, and the processed data is stored in the storage unit 220 or output to the output unit 240, the external storage device 250, or the communication unit 260. The functional block diagram of the control unit 101A is the same as the functional block diagram of the control unit 101 described with reference to FIGS.

The storage unit 220 includes a RAM (Random Access Memory) used as a work area necessary for executing the program by the control unit 101A, and a ROM (Read Only Memory) for storing a program to be executed by the control unit 101A. Including. In addition, a program for executing predetermined processing is read from the external storage device 250 and the communication unit 260 and stored in the RAM. Further, as an auxiliary storage device for assisting the storage area of the RAM, a hard disk drive of the external storage device 250 (hereinafter referred to as “HDD (Hard Disk Drive)”).
Such a magnetic disk storage device is used.

  The input unit 230 is an interface for inputting signals from a keyboard, a mouse, a touch panel, and the like, and can input necessary information to the document creation apparatus 200.

  The output unit 240 is an interface for outputting a signal to a display such as a liquid crystal display device (LCD, Liquid Crystal Display) and a speaker, and can output necessary information from the document creation device 200.

  The external storage device 250 reads a program or data recorded on the recording medium 251 and transmits it to the control unit 101A. Further, the external storage device 250 writes necessary data to the recording medium 251 in accordance with an instruction from the control unit 101A.

The computer-readable recording medium 251 includes a magnetic tape, a cassette tape, a floppy (registered trademark) disk, a magnetic disk such as a hard disk, and a CD-ROM (Compact Disk).
Optical disks such as Read Only Memory (DVD), Digital Versatile Disk (DVD), magneto-optical disks such as MO (Magneto Optical disk), MD (MiniDisc), memory cards such as IC cards and optical cards, or mask ROM, EPROM (Erasable This is a recording medium that carries a fixed program including a semiconductor memory such as a Programmable Read Only Memory (EEPROM), an EEPROM (Electrically Erasable and Programmable Read Only Memory), and a flash ROM.

  The communication unit 260 is connected to a wired or wireless network 900 and communicates with other computers connected to the network 900. The communication unit 260 is controlled by the control unit 101A and transmits predetermined information to another computer or receives information from another computer.

  In the present embodiment, the document creation apparatus 200 may or may not include the communication unit 260.

  Formula 2 is an example of a mathematical formula showing an erroneous input evaluation function used by the erroneous input candidate recognition unit in the second embodiment. Therefore, referring to Equation 2, in order to omit the function NUM that evaluates the keystroke count erroneous input process in the erroneous input evaluation function F, a weight is assigned to the function of three processes, and the weight of NUM is set to 0. Deal with omission of keystroke count error input process.

  That is, for the above-described erroneous input evaluation function of Equation 1, the weight Wp (0 ≦ Wp) for the keystroke location error input process, the weight Wn (0 ≦ Wn) for the keystroke error input process, and the keyless input process A weight Ws (0 ≦ Ws) is newly added.

  Here, when Wn = 0, the evaluation value NUM of the keystroke number erroneous input process is defined to be ignored. In the process described with reference to FIG. 11, at the time of initial setting (step S43), the weight of each process of the erroneous input evaluation function is referred to, and if there is a process with a weight of 0, this process is set not to be performed. To do. By setting in this way, the determination in step S54 is always determined as NO, so that the keystroke number erroneous input process can be prevented from being executed.

  Tables 11 and 12 are tables showing examples of recognition of erroneous input candidates in the document creation device 200 according to the second embodiment. Referring to Table 11 and Table 12, in the second embodiment, an erroneous input candidate when “Kaiten” is input for a kana character input mode of a keyboard that is a general input device in a personal computer. Explain the certification in detail. Here, the threshold used in the erroneous input evaluation function is 2.0. In addition, the weight of the keystroke location error input process is 1, the weight of the keystroke error input process is 0, and the weight of the keyless input process is 1. In addition, the evaluation value calculation method for each process is as described in the first embodiment.

  FIG. 13 is a diagram showing a specific example of key assignment of kana characters on a keyboard of a personal computer. Referring to FIG. 13, as a method of calculating an evaluation value of the keystroke location error input process, an evaluation value of 2.0 times is given to a key in contact with the key. Therefore, the evaluation value of the input key “ka” is 1.0, and the evaluation values of the keys “e”, “o”, “n”, “ki”, “ha”, and “su” in contact with the input key are 2.0. Note that this evaluation value calculation method is merely an example, and other calculation methods may be used.

  Referring to Table 11, first, regarding the key entry place erroneous input process, since the weight is 1, this process is performed to search for an erroneous input kana character string candidate. For the first character “ka”, the erroneous input kana character candidates are “e”, “o”, “n”, “ki”, “ha”, “su”. A kana character string formed by replacing this with an input character becomes “Eite”, “Ote”, “Nante”, “Kite”, “Hate”, and “Sute”. Similarly, as shown in the left column of Table 11, a total of 18 candidates are obtained for the second character “te” and the third character “i”. Since these evaluation values are each equal to or less than the threshold value 2.0, they are stored in the erroneous input kana character string candidate list as erroneous input kana character string information.

  Next, regarding the keystroke number erroneous input process, since the weight is 0, at the time of initial setting, the process execution information is set not to be performed. Therefore, as shown in the middle column of Table 11, there are no erroneous input kana character string candidates for this process.

  Next, since the weight is 1 for the key entry non-input processing, this processing is performed to search for an erroneously input kana character string candidate. For the first character “ka”, the erroneous input kana character candidate is “gaite”. Similarly, for the second character “te” and the third character “i”, a total of two candidates are obtained as shown in the right column of Table 11. Since these evaluation values are each equal to or less than the threshold value 2.0, they are stored in the erroneous input kana character string candidate list as erroneous input kana character string information.

  Here, since there is no evaluation value equal to or less than the threshold value 2.0 that can be the next processing target character string, the search for the erroneous input kana character string candidate is completed, and the erroneous input kana character string candidates as shown in Table 12 are completed. A list is obtained.

  In the second embodiment, the case where the kana input mode is the kana character input mode has been described. However, the present invention is not limited to this, and can be similarly applied when the mode for inputting a kana is the Roman character input mode.

[Third Embodiment]
In the first embodiment and the second embodiment, all processing weights are set to 1 or 0 in the erroneous input evaluation function. In the third embodiment, the document creation apparatus 100 suitable for the user will be described by changing the importance of each process by changing the weight and increasing the weight of the process that is easy for the user to make a mistake. The configuration of the document creation apparatus 100 in the third embodiment is the same as the configuration described in FIG.

  Table 13 is a table showing an example of erroneous input candidate recognition in the document creation device 100 according to the third embodiment. Referring to Table 13, as an example of a difference in the weight of the erroneous input evaluation function, the assumed kana character key assignment environment is input as in FIG. 5 and Table 2 described above. In this case, the erroneous input candidate recognition will be described in detail.

  Here, the threshold used in the erroneous input evaluation function is 2.0. The weight of the keystroke location error input process is 1, the weight of the keystroke error input process is 2, and the weight of the keyless input process is 1. The evaluation value calculation method for each process is the same as described in the first embodiment. Table 13 shows that “Ka” key is pressed once, “Ta” key is pressed four times, and “A” key is pressed twice for “Kate” as in the first embodiment. It is an incorrect input kana character string candidate list when a character is input.

  By referring to the erroneous input kana character string candidate list and increasing the weight of the erroneous input evaluation function for the keystroke count erroneous input process, the evaluation value of the erroneous input pseudonym character string candidate due to the erroneous input of the keystroke count becomes an error of other processing. This is a small value for the input kana character string, resulting in preferential treatment of the keystroke frequency incorrect input process. At the same time, an erroneously input character string candidate that satisfies the threshold value includes a kana character string that includes two erroneous keystrokes, so that the keystroke erroneous input process is favored.

  In this way, various evaluations using an erroneous input evaluation function are possible by changing the weight of each process. Furthermore, if the document creation apparatus 100 automatically calculates and changes the weight of each process from the user's action, more accurate evaluation can be performed. Therefore, a document creation apparatus 100 that enables automatic calculation of the weight of each process will be described.

  FIG. 14 is a functional block diagram of the document creation apparatus 100 according to the third embodiment. Referring to FIG. 14, the kana / kanji conversion determination unit 105 includes a display processing unit 501, a selection processing unit 502, and a personal information acquisition unit 503. Hereinafter, each configuration and processing form of the kana-kanji conversion determination unit 105 will be described in detail.

  The display processing unit 501 outputs to the output unit 140 data for displaying the kana-kanji converted character string candidate list of the character strings that have been converted by the kana-kanji conversion unit 104 in a form that the user can select.

  The selection processing unit 502 receives, from the input unit 130, an input that the user adopts the kana-kanji converted character string from the kana-kanji converted character string candidate list, and determines the character string to be adopted.

  The personal information acquisition unit 503 calculates, from the adopted kana-kanji conversion character string, a form of erroneous input that is easily mistaken by the user based on the evaluation value of each process in the erroneous input evaluation function, and reflects it in the weight of each process.

  FIG. 15 is a flowchart showing a flow of personal information acquisition processing executed when the personal information acquisition unit 503 is configured in the control unit 101 of the document creation apparatus 100 according to the third embodiment.

  Referring to FIG. 15, first, evaluation values of POS, NUM, and SYM of the erroneous input evaluation function F are acquired for a kana character string before conversion with respect to the kana-kanji conversion character string adopted by the user ( Step S71). If the acquired evaluation value F is 1.0, that is, if there is no erroneous input (NO in step S72), the personal information acquisition process is terminated.

  On the other hand, if the evaluation value F is not 1.0, that is, if a kana-kanji conversion character string for the erroneously input kana character string is adopted (YES in step S72), the evaluation functions POS, NUM, and , SYM evaluation values are newly stored as history data. Here, the past N times of data are stored, and if the update is Nth or more times, the old data is deleted (step S73).

  Next, from the history data, the evaluation values POS, NUM, and SYM for each process, and the average value for each N times are calculated (step S74), and these are set as the weights Wp, Wn, and Ws of each process (step S74). S75), and this is used as a new erroneous input evaluation function.

  FIG. 16 is a diagram showing an example of variation in the weight of the erroneous input evaluation function of the document creation device 100 according to the embodiment of the present invention. The assumed key assignment environment for kana characters is shown in FIG. The threshold used in the erroneous input evaluation function is 2.0. In addition, as a default setting, the weight of the keystroke location error input process is Wp = 1, the weight of the keystroke number error input process is Wn = 1, and the weight of the keyless key input process is Ws = 1. The evaluation value calculation method for each process is as described in the first embodiment.

  Here, if the user inputs “te” where “to” must be input, the kana-kanji conversion character string selected by the user is a character string converted from “to”. The value of the evaluation function NUM for erroneous input processing is 2.0. Here, the evaluation value of each process is stored in the history data, and the average value of the past history is calculated, and this is used as the weight of each new erroneous input process. That is, Wp = 1.0, Wn = 2.0, and Ws = 1.0.

  Thus, when the evaluation value of the kana-kanji conversion character string adopted by the user is not 1.0, that is, when the kana-kanji conversion character string of the erroneous input kana character string is adopted, the weight of each process is calculated, The tendency of erroneous input by the user is reflected to the erroneous input evaluation function.

  As an example of a technique for reflecting the tendency of the user's erroneous input to the erroneous input evaluation function, in the method described above, the evaluation value in the case of erroneous input for each process is linked to the weight of each process. However, the present invention is not limited to this, and various methods are conceivable, such as reflecting the ratio of the number of times for each processing as a whole. If the operation of selecting the user's kana-kanji conversion character string and the value of the three processing weights that are elements of the erroneous input evaluation function are linked, the method is not limited to the above-described method.

  In addition, as a tendency of the user's erroneous input, a method of changing the threshold value in consideration of the user's ease of error can be considered. The validity of the threshold value is determined and varied by taking a history of evaluation values by the erroneous input evaluation function of the kana character string before conversion of the kana / kanji conversion character string selected from the kana-kanji conversion character string candidate list.

[Fourth Embodiment]
In the first embodiment, the second embodiment, and the third embodiment, regarding the display method of the kana-kanji conversion character string candidate list, all kana-kanji conversion character string candidates are set to the error input evaluation function. A method of displaying a list in ascending order of evaluation values is conceivable.

  However, since there are a lot of kana-kanji conversion character string candidates to be displayed, it is difficult to find a target kana-kanji conversion character string only by displaying a list based on the evaluation value. Therefore, a display method that does not display an unimportant kana-kanji conversion character string is required.

  Therefore, only the character strings considered to be important among the kana-kanji conversion character strings of each erroneous input kana character string candidate are displayed, and the remaining characters are kana-kanji conversion character string candidates for the erroneous input kana character string only when the user needs them. Take a way to show.

  FIG. 17 is a flowchart showing a flow of kana-kanji conversion candidate display processing executed by configuring the display processing unit 501 and the selection processing unit 502 in the control unit 101 of the document creation apparatus 100 according to the embodiment of the present invention. is there.

  Referring to FIG. 17, first, control unit 101 displays a list of kana-kanji conversion character string candidates for input kana character strings included in kana-kanji conversion character string candidates created by kana-kanji conversion unit 104. Data is output to the output unit 140 (step S81). If there is a kana-kanji conversion character string to be adopted in this, and an input for selecting it is received from the input unit 130 (YES in step S82), the control unit 101 selects the selected kana-kanji conversion character string. A column is adopted (step S90). Thereafter, the control unit 101 returns the process to be executed to the caller process of the kana-kanji conversion candidate display process.

  If the input for selecting the kana-kanji conversion character string is not accepted (NO in step S82), control unit 101 determines whether or not the user has accepted an input for displaying the next candidate from input unit 130. Is determined (step S83). If it is determined that an input for displaying the next candidate is not accepted (NO in step S83), control unit 101 returns the process to be executed to step S82, and repeats steps S82 and S83. On the other hand, if it is determined that an input for displaying the next candidate has been accepted (YES in step S83), control unit 101 displays a kana-kanji conversion character string candidate upper rank for each erroneously input kana character string candidate as a next list. A set of N and erroneous input kana character strings is output to the output unit 140 as data that is displayed in a list in ascending order of evaluation values of the erroneous input evaluation function (step S84).

  If an input for selecting a kana-kanji conversion character string is accepted (YES in step 85), control unit 101 employs this kana-kanji conversion character string (step S90). Thereafter, the control unit 101 returns the process to be executed to the caller process of the kana-kanji conversion candidate display process.

  On the other hand, when the input for selecting the kana-kanji conversion character string is not accepted (NO in step S85), the control unit 101 determines whether or not the input for selecting the incorrect input kana character string is accepted ( Step S86). If it is determined that an input for selecting an erroneous input kana character string is not accepted (NO in step S86), control unit 101 returns the process to be executed to step S85, and repeats steps S85 and S86. On the other hand, if an input for selecting an erroneously input kana character string is accepted (YES in step S86), control unit 101 adds a kana-kanji conversion character string candidate for this erroneously input kana character string and displays a list ( Step S87).

  If there is a kana-kanji conversion character string to be employed in this, and an input for selecting it is received from the input unit 130 (YES in step S88), the control unit 101 selects the kana-kanji conversion character string. Adopt (step S90). Thereafter, the control unit 101 returns the process to be executed to the caller process of the kana-kanji conversion candidate display process.

  On the other hand, when an input for selecting a kana-kanji conversion character string to be adopted is not accepted (NO in step S88), control unit 101 determines whether a cancel input is accepted from input unit 130 or not. (Step S89). If it is determined that cancel input has not been accepted (NO in step S89), control unit 101 returns the process to be executed to step S88, and repeats step S88 and step S89. On the other hand, if it is determined that an input for cancellation has been accepted (YES in step S89), the control unit 101 cancels the kana-kanji conversion candidate display process on the assumption that no kana-kanji conversion character string candidate exists, and the user Return processing to the character input task.

  FIG. 18 is a diagram illustrating a display example of kana-kanji conversion character string candidates displayed by the document creation device 100 according to the fourth embodiment. FIG. 18 shows an example of display when “Kate” is input in the same environment as that of the third embodiment, and an example of selection when the kana-kanji conversion character string to be adopted is “Steamboat”. Indicates.

  First, as a result of the user inputting “Kate”, a list of kana-kanji converted character string candidates corresponding to the input kana character string “kite” (10 in FIG. 18) is displayed. However, since there is no kana-kanji conversion character string adopted by the user, the next list is displayed according to the operation by the user.

  Next, if all the kana-kanji conversion character string candidates related to the input kana character string “Kate” are displayed and there are no kana-kanji conversion character strings to be adopted, the next position is the upper position for each erroneous input kana character string candidate A list of kana-kanji converted character string candidates arranged up to the Nth (all candidates when there are N or less candidates) and an erroneous input kana character string itself are displayed in a list. Here, N = 2. In FIG. 18, two of the kana / kanji conversion character strings “low” and “boat” for the erroneously input kana character string “te” and the erroneously input kana character string “de” are displayed, and other such combinations are subsequently displayed. Is done.

  If the kana-kanji conversion character string candidate used by the user is a kana-kanji conversion character string candidate for each displayed erroneous input kana character string, the selected kana-kanji conversion character string is adopted.

  If there is no kana-kanji conversion character string adopted by the user and an erroneous input kana character string is selected, a list of kana-kanji conversion character string candidates for the erroneous input kana character string is displayed. Here, the erroneously input kana character string “Kite” is selected, and a list for this character string is displayed. Then, the “steaming” intended by the user is selected, and the selected kana-kanji conversion character string is adopted.

  Note that the kana-kanji conversion character string list display method can be any list display method that can efficiently select a desired kana-kanji conversion character string from a kana-kanji conversion character string for a kana character string erroneously input by the user. Other list display methods may be used.

  As described above, according to the document creation apparatus 100 in the present embodiment, conversion candidates are displayed for the kana character string input by the user by mistake without the user being conscious, and the intended kana-kanji conversion is performed. To be implemented.

  In the present embodiment, the invention of the document creation apparatus 100 has been described. However, the present invention is not limited to this, and the part related to character input in the document creation device 100 may be regarded as an invention of a character input device, may be regarded as an invention of a character input method performed by the character input device, The invention may be regarded as an invention of a character input program executed by the input device, or as a computer-readable recording medium on which the character input program is recorded.

[Fifth Embodiment]
FIG. 19 is a functional block diagram of the document creation device 100 according to the fifth embodiment. Referring to FIG. 19, the control unit 101 of the document creation apparatus 100 includes a character input unit 512, an evaluation calculation unit 513, a replacement character string recognition unit 514, and a display processing unit 515.

  Hereinafter, each component in the document creation apparatus 100 will be described in detail. The document creation apparatus 100 is fully equipped with some character input device connected to the input unit 130. This character input device inputs characters by pressing a general key, or a screen such as a touch panel. Any character can be used as long as the character is assigned to the key regardless of the form, such as inputting a character by touching the key.

  The character input unit 512 acquires a character string input by the user through the character input device as an input character string when the user prompts a conversion instruction. The opportunity for the character input unit 512 to acquire the input character string is not limited to the time when the user prompts the conversion instruction, but may be the time when the user inputs a predetermined number of characters or every predetermined period. Also good.

  Next, the evaluation calculation unit 513 calculates an evaluation value using the replacement evaluation function as an index for replacing the input character string with the replacement candidate character string for the acquired input character string. The relationship between the target character string and the evaluation value is stored as a replacement character string evaluation list with reference to Table 14.

  Next, the replacement character string recognition unit 514 cuts off the character string with respect to a character string whose evaluation value in the replacement character string evaluation list satisfies a predetermined condition. That is, a character string whose evaluation value satisfies a predetermined condition is recognized as a replacement character string.

  Next, the display processing unit 515 outputs data for displaying the replacement character string list in a form that the user can select to the output unit 140.

  Next, the replacement character string selection processing unit 516 receives an input from the input unit 130 that the user adopts the replacement character string from the replacement character string list, and determines the character string to be employed.

[Sixth Embodiment]
In the first embodiment, the embodiment targeting Japanese is described. However, in the sixth embodiment, an embodiment targeting languages other than Japanese will be described. The configuration of the document creation apparatus 100 according to the sixth embodiment is the same as the configuration of the document creation apparatus 100 according to the first embodiment described with reference to FIG.

  FIG. 20 is a functional block diagram of the document creation apparatus 100 according to the sixth embodiment. Referring to FIG. 20, the control unit 101 of the document creation apparatus 100 includes a character input unit 102, an erroneous input candidate recognition unit 103, a kanji conversion unit 604, and a kanji conversion determination unit 605.

  Regarding the character input unit 102 and the erroneous input candidate recognition unit 103, the target character string is only Japanese in the first embodiment, but the target character string is also applicable to languages other than Japanese in the sixth embodiment. It is the same processing operation only by changing to the configuration described above.

  Next, the kanji conversion unit 604 sets the target character string of the kana-kanji conversion unit 104 in the first embodiment to Japanese only, but in the sixth embodiment, the target character string is set to a language other than Japanese. The same processing operation is performed only by changing the configuration to the target. In the case of Korean, the kanji conversion character string created by the kanji conversion unit 604 includes Korean characters and does not necessarily target only the kanji character string.

  Next, the kanji conversion determination unit 605 sets only the target character string of the kana-kanji conversion determination unit 105 in the first embodiment to Japanese. However, in the sixth embodiment, the target character string is set to other than Japanese. The same processing operation is performed only by changing the configuration to a language.

  FIG. 21 is a functional block diagram illustrating details of the erroneous input candidate recognition unit 103 of the document creation device 100 according to the sixth embodiment.

  Referring to FIG. 21, the erroneous input candidate recognition unit 103 includes a processing control unit 301, a keystroke location erroneous input processing unit 302, a keystroke number erroneous input processing unit 303, a keystroke non-input processing unit 304, and an erroneous input character string candidate. A storage processing unit 315 is provided.

  Here, referring to Equation 1, the erroneous input evaluation function F is defined as a kana character string because the variables in and obj in the first embodiment are only for Japanese, but In the embodiment of the present invention, the evaluation is performed for a language other than Japanese, the evaluation function POS is evaluated based on the keystroke location error input process for each input, and the evaluation function is evaluated based on the keystroke count error input process. It is assumed that the function is based on the three elements of the NUM and the evaluation function SYM evaluated based on the non-keystroke input process.

  Therefore, the processing flow of the processing control unit 301, the keystroke location incorrect input processing unit 302, the keystroke number erroneous input processing unit 303, the keyless non-input processing unit 304, and the erroneous input character string candidate storage processing unit 315 according to the sixth embodiment. Is the same as in the first embodiment.

  FIG. 22 is a flowchart illustrating a keystroke location error input process executed when the keystroke location error input processing unit 302 is configured in the control unit 101 of the document creation apparatus 100 according to the sixth embodiment.

  Referring to FIG. 22, the character string to be processed in the first embodiment is a kana character string, but in the sixth embodiment, the process is also for a language other than Japanese. Therefore, the processing flow in FIG. 4 is the same as that shown in FIG. 4 except that the target character string is changed to a configuration that targets languages other than Japanese.

In the following, the flow of processing will be described using Korean as an example of a language other than Japanese.
FIG. 23 is a diagram of a specific example of Hangul key assignment in a mobile phone. A method for inputting Korean characters will be described with reference to FIG. Hangul combines a vowel and a consonant to form one character. Table 15 shows symbols of Korean vowels and consonants. Referring to Table 15, there are 10 “basic vowels”, 11 “composite vowels”, and 19 consonants, which are combined to form one Hangul character.

  Table 16 is a table showing a specific example of assignment of the number of keystrokes of Korean characters in the mobile phone.

  According to FIG. 23, basic vowels and compound vowels are input by combining a plurality of “1” to “3” keys. The consonant is input by pressing the “4” to “0” keys and then pressing each key continuously. When inputting one Hangul character, “consonant + vowel” or “consonant + vowel + batchum (second consonant)” can be input.

  Referring to Table 17, in order to input the character 1, the “8” key is pressed twice for the input of the character 1-1 as a consonant, and the “1” key is input for the input of the character 1-2 as a vowel. , Once with the “2” key, and once with the “5” key to input the character 1-3 as a batch. For character 2, the "7" key is input twice to input the character 2-1 that is a consonant, the "2" key is input once to input the character 2-2 that is a vowel, and the "1" key is input twice. Finally, the user inputs the key by pressing the “*” key indicating the end of the input.

  Here, with reference to FIG. 23 and Table 16, the method of searching for an erroneous input character candidate in step S114 of FIG. 22 and the method of the erroneous input evaluation function evaluating the erroneous input character string will be described in detail.

  Similar to the case of Japanese described in the first embodiment, there is a system in which a plurality of symbols can be assigned to one key even in Korean. The sixth embodiment targets such a character input device.

  FIG. 24 is a diagram illustrating an example of an error in the keystroke location by the user in the sixth embodiment. Referring to FIG. 24, when the user inputs character 1, a key to which character 1-1 as a consonant is assigned, a key constituting character 1-2 as a vowel, and a character 1-3 as a batch. For each keystroke to which the key is assigned, the possibility of erroneous input due to the physical distance of the key positional relationship is evaluated.

  First, it is a case where the input of the character 1-1 is targeted. Symbols assigned to keys arranged at the same distance in the vertical and horizontal directions are evaluated as equivalent evaluation values.

  Table 18 is a table showing one calculation example of the evaluation value calculated based on the erroneous input evaluation function pos having the keystroke place erroneous input as an element. Here, Table 18 is a specific example in the case where the evaluation value of the input symbol is 1.0 and 2.0 times is repeated for keys arranged at the same distance in the vertical and horizontal directions. In the sixth embodiment, for the sake of convenience, the evaluation value is set to 2.0 times as described above. However, when the evaluation value is defined to be larger as the distance is larger, the evaluation value is larger than 1.0. The value is not limited to this.

  Next, it is assumed that vowel input is similarly evaluated. When inputting the character 1-2 that is a vowel, the key “1” is pressed and the key “2” is pressed. Tables 19 and 20 are tables showing one calculation example of the evaluation values calculated based on the erroneous input evaluation function pos having the keystroke place erroneous input as an element in the same manner as Table 18.

  Next, batch batch input is evaluated in the same manner. Table 21 is a table showing one calculation example of evaluation values calculated based on the erroneous input evaluation function pos having the keystroke place erroneous input as an element when inputting characters 1-3 as a batch, as in Table 18. It is.

  In this way, in order to input one Korean character, it is necessary to press a plurality of keys. Unlike the first embodiment, these evaluations are performed a plurality of times for one character. become. Therefore, by multiplying these evaluations, the erroneous input evaluation function pos having the keystroke place erroneous input as an element is calculated.

  Next, when the control shifts to the keystroke number error input processing unit 303, the keystroke number error input processing unit 303 erroneously inputs the character string to be processed based on the evaluation function num that causes the keystroke number process. Search for character candidates.

  FIG. 25 is a flowchart illustrating the flow of an erroneous input process of keystrokes executed by configuring the controller 101 of the document creation apparatus 100 according to the sixth embodiment with an erroneous input process number of keystrokes 303.

  Referring to FIG. 25, the character string to be processed in the first embodiment is a kana character string. However, in the sixth embodiment, the processing is for a language other than Japanese. Therefore, the processing flow in FIG. 7 is the same as that shown in FIG. 7 except that the target character string is changed to a configuration that targets languages other than Japanese.

  Here, the method for searching for erroneous input character candidates in step S124 and the method for evaluating the erroneous input character string by the erroneous input evaluation function will be described in detail. The assumed character key assignment environment is the same as that in FIG. 23 and Table 16 described above.

  FIG. 26 is a diagram illustrating an example of an error in the number of keystrokes made by the user according to the sixth embodiment. Referring to FIG. 26, characters 1-1-1 assigned to the “8” key can be input by pressing the “8” key once. Following this, the pattern of characters 1-1 being repeated twice, characters 1-1-2 being consecutively three times, and characters 1-1-1 being consecutively repeated four times, each of which can be input by typing, followed by repeated patterns. Go.

  Here, when the character 1-1 is input, the erroneous input evaluation function evaluates the possibility that the “8” key to which the character 1-1 is assigned is erroneously input due to the difference in the number of keystrokes. Here, it is assumed that the characters 1-1-1 and 1-1-2 assigned to the number of keystrokes of ± 1 are evaluated as equivalent evaluation values.

  Table 23 is a table showing one calculation example of evaluation values calculated based on an erroneous input evaluation function num caused by erroneous input of the number of keystrokes. Here, Table 23 is a calculation example in the case where the evaluation value of the input character is 1.0 and 2.0 times is repeated for the character assigned to the keystroke count ± 1. In the sixth embodiment, for the sake of convenience, the evaluation value is set to 2.0 times as described above. However, when the evaluation value is defined to be larger as the difference in the number of consecutive keystrokes is larger, it is larger than 1.0. If it is a value, it is not restricted to this value.

  Next, when the control shifts to the keyless non-input processing unit 304, the keyless non-input processing unit 304 regards a character string to be processed based on the evaluation function sym caused by the keyless non-input processing as a factor. Search for candidates.

  FIG. 27 is a flowchart of the non-keystroke input process executed when the keyless non-input processing unit 304 is configured in the control unit 101 of the document creation apparatus 100 according to the sixth embodiment.

  Referring to FIG. 27, the character string to be processed in the first embodiment is a kana character string, but in the sixth embodiment, the process is also for a language other than Japanese. Therefore, the processing flow in FIG. 9 is the same as that shown in FIG. 9 except that the target character string is changed to a configuration that targets languages other than Japanese.

  Here, in the case of Korean, there is no symbol that can be treated as a symbol that can be added to the character to be processed in step S134. There is no difference due to the evaluation value. However, if there is something to which a symbol is added depending on the input method, it is processed as in the first embodiment.

  The erroneous input character string candidate storage processing unit 305 is based on the additional candidate list created by the keystroke location error input processing unit 302, the keystroke number error input processing unit 303, and the key input non-input processing unit 304. Store incorrect input character string information in the candidate list. At this time, only erroneous input character string information in which the evaluation value of the erroneous input character string in the additional candidate list is smaller than a predetermined threshold is stored in the erroneous input character string candidate list.

  In addition, when the erroneous input character string candidate held in the additional candidate list is already stored in the erroneous input character string candidate list, the smaller evaluation value is updated as erroneous input character string information. When the storage process is completed, the incorrect input character string candidate list is sorted in ascending order based on the evaluation value. This is intended to continuously target a character string having the smallest evaluation value.

  When the storage process ends, the process returns to the process control unit 301. If the evaluation value of the erroneous input evaluation function for the character string to be processed is larger than the threshold value, the erroneous input candidate recognition unit 103 passes the erroneous input character string candidate list to the kanji conversion unit 604. As described above, the document creation apparatus 100 proceeds to the kanji conversion work.

  FIG. 28 is a flowchart illustrating a flow of main processing executed by the control unit 101 of the document creation apparatus 100 according to the sixth embodiment.

  Referring to FIG. 28, the character string to be processed in the first embodiment is a kana character string, but in the sixth embodiment, the processing is also for a language other than Japanese. Therefore, the processing flow in FIG. 11 is the same as that shown in FIG. 11 except that the target character string is changed to a configuration that targets languages other than Japanese.

  Tables 24 and 25 are tables showing an example of erroneous input candidate recognition processing when character 1 is input in the document creation apparatus 100 according to the sixth embodiment. With reference to Table 24 and Table 25, the erroneous input candidate recognition process when the character 1 is input in the sixth embodiment will be described in detail. Here, the threshold value of the evaluation value used in the erroneous input evaluation function is 2.0. In addition, the evaluation value calculation methods of the keystroke location error input process, the keystroke frequency error input process, and the keyless key input process are as described above. The assumed character key assignment environment is shown in FIG.

  Character 1 can be input by pressing the “8” key twice, the “1” key once, the “2” key once, and the “5” key once.

  Referring to Table 24, first, an erroneous input character string candidate when the keystroke place erroneous input process is performed on the input character string is searched. Since the threshold value is 2.0, there are four wrong input character string candidates for the consonant character 1-1, four candidates from the top of the leftmost column, and no character 1-2 for the vowel. First, there are a total of 7 candidates, 3 candidates from the bottom of the leftmost column for the characters 1-3 that are batches. These are stored as erroneous input character string information in the erroneous input character string candidate list.

  Next, an erroneous input character string candidate when the keystroke count erroneous input process is performed is searched. Since the threshold value is 2.0, the erroneous input character string candidates are two candidates from the top of the center column for the consonant character 1-1, and the top of the center column for the character 1-2 of the vowel. From the third candidate to the character 1-3 that is a batch, there are a total of four candidates, one candidate at the bottom of the center column. These are stored as erroneous input character string information in the erroneous input character string candidate list.

  Next, a wrong input character string candidate when the key entry non-input process is performed is searched. As described above, there is no candidate in Korean because no candidate for the key entry non-input process is found.

  As described above, since there is no evaluation value equal to or less than the threshold value 2.0, the search for erroneous input character string candidates is completed, and an erroneous input character string candidate list as shown in Table 25 is obtained.

  Also, the flow of processing in the case of Chinese as a language other than Japanese and Korean will be described. Take Pinyin input, which is common in Chinese input, as an example. Pinyin is a phonetic representation of Chinese in Roman characters, and the Chinese characters are input by inputting Roman characters. FIG. 29 is a diagram of a specific example of alphabetic key assignment used for pinyin input in a mobile phone. Table 26 is a table showing a specific example of assignment of the number of times of keystrokes of alphabets in the mobile phone.

  First, when the control shifts to the keystroke location error input processing unit 302, the keystroke location error input processing unit 302 performs an error on the character string to be processed based on the evaluation function pos having the keystroke location error input process as an element. Search for input character candidates. Here, with reference to FIG. 29 and Table 26, a method for searching for an erroneous input character candidate in step S114 in FIG. 22 and a method for the erroneous input evaluation function to evaluate an erroneous input character string will be described in detail.

  FIG. 30 is a diagram illustrating an example of an erroneous input of a keystroke place by a user according to the sixth embodiment. Referring to Table 27, when the user inputs the character 3, the first character 3-1 can be input by inputting "PIN", and the second character 3-2 can be input. Input is made possible by inputting “YIN”, and the character 3 is input in this way.

  Referring to FIG. 30, when the user inputs “P”, the erroneous input evaluation function is erroneously input due to the physical distance of the positional relationship with respect to a key different from the key to which “P” is assigned. Evaluate the possibility of being. Here, the alphabets assigned to the “4” key and the “8” key arranged on the top, bottom, left, and right are evaluated as equivalent evaluation values.

  Table 28 is a table showing one calculation example of the evaluation values calculated based on the erroneous input evaluation function pos having the keystroke place erroneous input as an element. Here, Table 28 is a calculation example in the case where the evaluation value of the input alphabet is 1.0, and 2.0 times is repeated for keys arranged on the top, bottom, left, and right. For convenience, the evaluation value is set to 2.0 times as described above. However, when the evaluation value is defined to be larger as the distance is larger, the value is not limited to this value as long as the value is larger than 1.0.

  Next, when the control shifts to the keystroke number error input processing unit 303, the keystroke number error input processing unit 303 relates to the processing target character string based on the evaluation function num having the keystroke number error input process as an element. Search for incorrect input character candidates.

  FIG. 31 is a diagram illustrating an example of erroneous input of the number of keystrokes by the user according to the sixth embodiment. Referring to FIG. 31, in order to input the alphabet “P”, the “7” key is pressed once. Here, when “P” is input, the erroneous input evaluation function evaluates the possibility that the “7” key is erroneously input due to the difference in the number of keystrokes. Here, “Q” assigned to the number of keystrokes of ± 1 or “no keystroke” state “” is evaluated as an equivalent evaluation value.

  When “P” is input, the contents of erroneous input differ between “P” in which the key is pressed once and “P” in which the key is pressed five times. For the former “P”, “” with no keystrokes and “Q” with no keystrokes and “Q” with two keystrokes are candidates for wrong input characters, while the latter “P” indicates “S” with four keystrokes. “Q” with the number of keystrokes of 6 is the closest candidate to the erroneously input character.

  Tables 29 and 30 are tables showing one calculation example of evaluation values calculated based on the erroneous input evaluation function num having the keystroke number erroneous input as an element. Here, Table 29 and Table 30 are calculation examples in the case where the evaluation value of the input alphabet is 1.0 and 2.0 times is repeated for the characters assigned to the number of keystrokes ± 1. In the sixth embodiment, for the sake of convenience, the evaluation value is set to 2.0 times as described above. However, when the evaluation value is defined to be larger as the difference in the number of keystrokes is larger, a value larger than 1.0 is set. If so, it is not limited to this value.

  Next, when the control shifts to the keyless non-input processing unit 304, the keyless non-input processing unit 304, regarding the character string to be processed, erroneously input characters based on the evaluation function sym having the keyless non-input processing as an element. Search for column candidates.

  Here, in the case of Chinese, in step S134 of FIG. 29, there is no symbol that can be treated as a symbol that can be added to the character to be processed. There is no difference due to the evaluation value of the input evaluation function SYM. However, if there is one that adds a symbol depending on the input method, the same processing as in the first embodiment is performed.

  Similarly to the case of Korean, the erroneous input character string candidate storage processing unit 305 is an additional candidate list created by the keystroke location erroneous input processing unit 302, the keystroke number erroneous input processing unit 303, and the keyless non-input processing unit 304. Is stored in the erroneous input character string candidate list. At this time, only erroneous input character string information in which the evaluation value of the erroneous input character string in the additional candidate list is smaller than a predetermined threshold is stored in the erroneous input character string candidate list.

  In addition, when the erroneous input character string candidate held in the additional candidate list is already stored in the erroneous input character string candidate list, the smaller evaluation value is updated as erroneous input character string information. When the storage process is completed, the incorrect input character string candidate list is sorted in ascending order based on the evaluation value. This is intended to continuously target a character string having the smallest evaluation value.

  When the storage process ends, the process returns to the process control unit 301. If the evaluation value of the erroneous input evaluation function for the character string to be processed is larger than the threshold value, the erroneous input candidate recognition unit 103 passes the erroneous input character string candidate list to the kanji conversion unit 604. As described above, the document creation apparatus 100 proceeds to the kanji conversion work.

[Seventh Embodiment]
In the sixth embodiment, the erroneous input character string candidate storage processing 305 includes an additional candidate list created by the keystroke location error input processing unit 302, the keystroke number error input processing unit 303, and the key input non-input processing unit 304. Based on this, erroneous input character string information was stored in the erroneous input character string candidate list. At this time, only the erroneous input character string information in which the evaluation value of the erroneous input character string in the additional candidate list is smaller than the predetermined threshold value is stored in the erroneous input character string candidate list. However, in this method, the threshold value is too large. In this case, a large number of erroneous input character string candidates are created. Therefore, in the seventh embodiment, a method of storing a character string having a higher N (> 1) evaluation value in the erroneous input character string candidate list rather than a method of comparing the evaluation value with a threshold value. Describe.

  FIG. 32 is a flowchart illustrating a flow of main processing executed by the control unit 101 of the document creation apparatus 100 according to the seventh embodiment. Here, the flow processing other than step S151 is the same as in FIG.

  By the processing from step S141 to step S43, erroneous input character string information for the input character string is created, stored in the erroneous input character string candidate list, and processing execution information is set with this character string as a processing target character string.

  Next, if the evaluation value of the erroneous input character string information is lower than the evaluation value of the Nth character string in the erroneous input character string candidate list with respect to the processing target character string, or an erroneous input If the total number of character strings stored in the character string candidate list is less than N (YES in step S151), the processing from step S52 to step S57 is performed.

  Then, the control unit 101 sorts the erroneous input character string candidate list in ascending order based on the evaluation value included in the erroneous input character string information, and is positioned at the Nth position from the erroneous input character string candidate list. Erroneous input character string information having an evaluation value that exceeds the evaluation value of the character string is deleted from the erroneous input character string candidate list (step S158).

  As described above, the processing of the processing target character string ends (in the case of YES in step S59), or N evaluation values for the processing target character string of the erroneous input character string information are listed from the erroneous input candidate list. If the evaluation value is equal to or higher than the evaluation value of the character string located in the eye and the total number of character strings stored in the erroneous input kana character string candidate list is N or more (NO in step S51), step 61 is performed. The subsequent processing is performed.

[Eighth Embodiment]
Next, in the seventh embodiment, the erroneous input character string candidate storage processing 305 is created by the keystroke location erroneous input processing unit 302, the keystroke number erroneous input processing unit 303, and the keyless non-input processing unit 304. Based on the additional candidate list, erroneous input character string information is stored in the erroneous input character string candidate list. At this time, when the character string having the evaluation value of the top N (> 1) is stored in the erroneous input character string candidate list, if N is large and the evaluation value is large, an excessive number of erroneous input characters There is a possibility of creating a column candidate. Therefore, in the eighth embodiment, a method for storing a character string satisfying a threshold value with respect to an evaluation value and a character string having a higher N (> 1) evaluation value in an erroneous input character string candidate list is described. To do.

  FIG. 33 is a flowchart illustrating a flow of main processing executed by the control unit 101 of the document creation device 100 according to the eighth embodiment. Here, the flow processing other than step S251 is the same as in FIG.

  By the processing from step S141 to step S43, erroneous input character string information for the input character string is created, the input character string is stored in the erroneous input character string candidate list, and the process execution information is set with the input character string as the processing target character string. To do.

  Next, for the processing target character string, the evaluation value of the erroneous input character string information is the smaller of the evaluation value of the character string located at the Nth position in the erroneous input character string candidate list or the threshold value. If the total number of character strings stored in the erroneous input character string candidate list is less than N and the evaluation value of the erroneous input character string information is below the threshold (in step S251). YES), the processing from step S52 to step S57 is performed.

  Then, the control unit 101 sorts the incorrect input character string candidate list in ascending order based on the evaluation value included in the erroneous input character string information, and the Nth character from the erroneous input character string candidate list. The erroneous input character string information having the evaluation value of the column or the evaluation value exceeding the threshold value is deleted from the erroneous input character string candidate list (step S158).

  As described above, the processing of the processing target character string ends (in the case of YES in step S59), or the evaluation value for the processing target character string of the erroneously input character string information is Nth in the erroneous input candidate list. Is equal to or higher than the evaluation value of the character string located at, and the total number of character strings stored in the erroneous input character string candidate list is N or more (in the case of NO in step S51), the steps after step 61 are performed. Perform processing.

[Ninth Embodiment]
In the first embodiment, the erroneous input evaluation function performs an evaluation based on a keystroke location erroneous input process in which evaluation is performed based on a physical distance between character input keys, and a difference in the number of keystrokes of the character input key. This is intended for three elements: an erroneous input process for the number of keystrokes to be performed, and a keyless input process for performing an evaluation based on a difference in symbols with respect to characters. In the ninth embodiment, a description will be given of an erroneous input evaluation function further including a keystroke direction erroneous input process for performing evaluation based on the positional relationship between the upper, lower, left, and right between character input keys.

  FIG. 34 is a functional block diagram showing details of the erroneous input candidate recognition unit 103 of the document creation apparatus 100 according to the ninth embodiment.

  Referring to FIG. 34, the erroneous input candidate recognition unit 103 includes a processing control unit 301, a keystroke location erroneous input processing unit 302, a keystroke number erroneous input processing unit 303, a keystroke non-input processing unit 304, and a keystroke direction erroneous input processing unit 606. , And an erroneously input character string candidate storage processing unit 315. Formula 3 is an example of a mathematical formula showing an erroneous input evaluation function used by the erroneous input candidate recognition unit 103 according to the ninth embodiment.

  Here, with reference to Equation 3, the erroneous input evaluation function F evaluates the character string obj with respect to the input character string in, and is evaluated based on the keystroke location error input process. This is a function based on the four elements of the evaluation function NUM evaluated based on the processing, the evaluation function SYM evaluated based on the keyless non-input processing, and the evaluation function DIR based on the keying direction incorrect input processing. The functions POS, NUM, SYM, and DIR for evaluating the character string obj with respect to the input character string in are the n-th input character in [n] constituting the two character strings in and obj, respectively. Is composed of functions pos, num, sym and dir which evaluate the character obj [n].

  Hereinafter, each configuration and processing form of the erroneous input candidate recognition unit 103 will be described in detail. The processing control unit 301 determines control shift to the next processing to be performed based on the input character string acquired by the character input unit 102 and the erroneous input character string candidate list. The contents to be controlled are the determination of the character string to be processed, the control shift to the keystroke location error input processing unit 302, the control shift to the keystroke number error input processing unit 303, the control shift to the keyless non-input processing unit 304, These are transition to the keystroke direction erroneous input processing unit 606, determination of termination of processing in the erroneous input candidate recognition unit 103, and control transition to the kanji conversion unit 104. Here, the processes of the keystroke location error input processing unit 302, the keystroke number error input processing unit 303, and the key input non-input processing unit 304 are the same as those in the sixth embodiment.

  FIG. 35 is a flowchart illustrating the flow of a keying direction erroneous input process executed by configuring the keying direction erroneous input processing unit 606 in the control unit 101 of the document creation apparatus 100 according to the ninth embodiment.

  Referring to FIG. 35, first, the control unit 101 acquires a character string “in” to be processed (step S91). Using the acquired character string in as the processing target character string in, search for the character obj [n] when the user erroneously inputs the keystroke direction of each key in [n] constituting the character string in. To do.

  The control unit 101 focuses on the first character in [1] of the processing target character string in (step S92). If there is a character in [n] to be focused here (YES in step S93), the control unit 101 searches for a candidate obj [n] of an erroneously input character, and finds the searched obj [n]. An evaluation value dir (in [n], obj [n]) of the evaluation function dir is calculated (step S94).

  Then, the control unit 101 creates a character string obj by replacing the focused character in [n] with the searched character obj [n] in the target character string in (step S95). Next, the control unit 101 creates erroneous input character string information including the character string obj and the evaluation value dir (in [n], obj [n]) for the character string obj, and stores it in the additional candidate list ( Step S96). Next, the control unit 101 moves the focus to the next character in [n + 1] (step S97).

  And the control part 101 repeats the process from step S94 to step S97 with respect to all the characters in [n]. On the other hand, if the character in [n] to be focused does not exist (NO in step S93), since all processing for the target character string in is completed, the control unit 101 is created. The additional candidate list is transferred to the erroneous input character string candidate storage processing unit 315, which is the next process, and the process proceeds.

  Here, with reference to FIG. 5 and Table 2, the method for searching for erroneous input character candidates in step S94 and the method for evaluating the erroneous input character string by the erroneous input evaluation function will be described in detail.

  In the key input location error input process, the physical distance of the positional relationship is evaluated as an element. In the key input direction error input process, the possibility of erroneous input due to the positional relationship direction is additionally evaluated.

  FIG. 36 is a diagram illustrating an example of erroneous input of the keystroke direction by the user in the ninth embodiment. Referring to FIG. 36, when the user inputs “NA”, the erroneous input evaluation function is erroneously input due to the orientation of the positional relationship with respect to a key different from the key to which “NA” is assigned. Assess the possibilities. Here, it is assumed that the keys arranged in the regions divided into the eight directions are evaluated using the keys in the respective regions as equivalent evaluation values.

  Table 31 is a table showing one calculation example of evaluation values calculated based on the erroneous input evaluation function dir caused by keystroke direction erroneous input. Here, in Table 31, the evaluation value of the input character is 1.0, and 2.0 (top), 2.0 (right), 3.0 (bottom), 4.0 (left) for each of the eight directions. ), 5.0 (upper right), 6.0 (lower right), 7.0 (upper left), and 8.0 (lower left). The evaluation value depends on the direction. For convenience, the evaluation value is assigned as described above as a method for calculating the evaluation value. However, if the evaluation value is defined to be large as the possibility of erroneous input is small, Not limited to values. Further, it is necessary to assign a more appropriate value by investigating the tendency of the user's erroneous input. Here, the characteristics of each user may be acquired as history information to make the evaluation value dynamic.

  FIG. 37 is a flowchart showing a flow of main processing executed by the control unit 101 of the document creation apparatus 100 according to the ninth embodiment. Referring to FIG. 37, the processing is the same as that in FIG. 11 except for the processing in steps S156 and S157.

  Here, in step S56, it is determined that the control unit 101 does not perform the keystroke non-input process following the keystroke location error input process and the keystroke number input process (in the case of NO in step S56), and the keystroke direction error is determined. If it is determined that the input process is to be performed (YES in step S156), control unit 101 performs a keystroke direction erroneous input process (step S157). If it is determined from the process execution information that the keying direction incorrect input process is not performed (NO in step S156), the control unit 101 shifts the process to step 59.

[Tenth embodiment]
In the third embodiment, the erroneous input evaluation function used by the erroneous input candidate recognition unit 103 does not include elements related to the keystroke direction erroneous input process. In the tenth embodiment, a description will be given of a document creation apparatus 100 that performs an evaluation suitable for a user by changing the weight of each process including this process.

  Formula 4 is an example of a mathematical formula showing an erroneous input evaluation function used by the erroneous input candidate recognition unit in the tenth embodiment. A weight (0 ≦ Wd) for the keystroke direction erroneous input process is newly added to the above-described erroneous input evaluation functions of Equations 2 and 3.

  If the weight Wd of the keystroke direction erroneous input process in Equation 4 is set to 0, the weight of each process of the erroneous input evaluation function is referred to, and if there is a process with a weight of 0, the process is set not to be performed. Therefore, by always determining NO in step S156 of FIG. 37, the keystroke direction erroneous input process is not executed, and is equivalent to the third embodiment.

  FIG. 38 is a functional block diagram of the document creation device 100 according to the tenth embodiment. With reference to FIG. 38, the functional block diagram described in FIG. 14 of the third embodiment is the same as the processing configuration except that the target character string is changed to a configuration that targets languages other than Japanese. It is.

  FIG. 39 is a flowchart showing a flow of personal information acquisition processing executed when the personal information acquisition unit 503 is configured in the control unit 101 of the document creation apparatus 100 according to the tenth embodiment.

  Referring to FIG. 39, first, evaluation values of POS, NUM, SYM, and DIR of the erroneous input evaluation function F are acquired for the character string before conversion for the kanji conversion character string adopted by the user (step S171). ). If the acquired evaluation value F is 1.0, that is, if no erroneous input has been made (NO in step S172), the personal information acquisition process is terminated.

  On the other hand, if the evaluation value F is not 1.0, that is, if a Kanji conversion character string for the erroneously input character string is adopted (YES in step S172), the evaluation functions POS, NUM, SYM and Each DIR evaluation value is newly stored as history data. Here, the past N times of data are stored, and if the update is Nth or more times, the old data is deleted (step S173).

  Next, from the history data, the evaluation values POS, NUM, SYM, and DIR for the respective processes are averaged for N times (step S174), and these are set as the weights Wp, Wn, Ws, and Wd for each process ( In step S175), this is used as a new erroneous input evaluation function.

  As described above, when a Kanji conversion character string of an erroneously input character string is employed, the weight of each process is calculated and varied to reflect the tendency of the erroneous input of the user in the erroneous input evaluation function.

  As an example of a technique for reflecting the tendency of the user's erroneous input to the erroneous input evaluation function, in the method described above, the evaluation value in the case of erroneous input for each process is linked to the weight of each process. However, the present invention is not limited to this, and various methods are conceivable, such as reflecting the ratio of the number of times for each processing as a whole. If the user's operation of selecting a Kanji conversion character string and the value of the four processing weights that are elements of the erroneous input evaluation function are linked, the method is not limited to the above-described method.

[Eleventh embodiment]
Next, referring to FIG. 19, the document creation apparatus 101 does not necessarily have to be integrated as one apparatus, and the character input unit 512 is configured by the remote controller 800, and a system for inputting characters by remote operation is also considered. It is done. FIG. 40 shows a method of inputting characters to the display device 700 using the remote controller 800. In this example, the remote controller 800 simply inputs a character, and the evaluation calculation unit 513 and the replacement character string recognition unit 514 are attached to the display device 700 on the remote control 800 receiving side.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

It is a block diagram which shows the outline of a structure of the document preparation apparatus in embodiment of this invention. It is a functional block diagram of the document preparation apparatus in embodiment of this invention. It is a functional block diagram which shows the detail of the erroneous input candidate recognition part of the document preparation apparatus in embodiment of this invention. It is a flowchart which shows the flow of the keystroke location error input process performed when a keystroke location error input process part is comprised in the control part of the document preparation apparatus in embodiment of this invention. It is a figure of one specific example of the key assignment of the kana character in a mobile telephone. It is a figure which shows an example of the error of the keystroke place by the user in embodiment of this invention. It is a flowchart which shows the flow of the keystroke frequency incorrect input process performed when a keystroke frequency error input process part is comprised in the control part of the document preparation apparatus in embodiment of this invention. It is a figure which shows an example of the error of the keystroke frequency by the user in embodiment of this invention. It is a flowchart which shows the flow of the non-keystroke input process performed when a keystroke non-input process part is comprised in the control part of the document preparation apparatus in embodiment of this invention. It is a figure which shows an example of the error of the dakuten and the semi-dakuten by the user in embodiment of this invention. It is a flowchart which shows the flow of the main process performed by the control part of the document preparation apparatus in embodiment of this invention. It is a block diagram which shows the outline of a structure of the document preparation apparatus in 2nd Embodiment. It is a figure which shows one specific example of the key assignment of the kana character in the keyboard of a personal computer. It is a functional block diagram of the document preparation apparatus in 3rd Embodiment. It is a flowchart which shows the flow of the personal information acquisition process performed when a personal information acquisition part is comprised in the control part of the document preparation apparatus in 3rd Embodiment. It is the figure which showed an example of the fluctuation | variation example of the weight of the erroneous input evaluation function of the document preparation apparatus in embodiment of this invention. It is a flowchart which shows the flow of the kana-kanji conversion candidate display process which is an example which the display process part and the selection process part perform in the display part of the document preparation apparatus in embodiment of this invention. It is a figure which shows the example of a display of the kana / kanji conversion character string candidate displayed by the document preparation apparatus in 4th Embodiment. It is a functional block diagram of the document preparation apparatus in 5th Embodiment. It is a functional block diagram of the document preparation apparatus in 6th Embodiment. It is a functional block diagram which shows the detail of the erroneous input candidate recognition part of the document preparation apparatus in 6th Embodiment. It is a flowchart which shows the flow of the keystroke location incorrect input process performed when a keystroke location incorrect input process part is comprised in the control part of the document preparation apparatus in 6th Embodiment. It is a figure of one specific example of the key assignment of Hangul in a mobile phone. It is a figure which shows an example of the error of the keystroke place by the user in 6th Embodiment. It is a flowchart which shows the flow of the keystroke frequency incorrect input process performed when a keystroke frequency error input process part is comprised in the control part of the document preparation apparatus in 6th Embodiment. It is a figure which shows an example of the error of the keystroke frequency by the user in 6th Embodiment. It is a flowchart of the keystroke non-input process performed when a keystroke non-input process part is comprised in the control part of the document preparation apparatus in 6th Embodiment. It is a flowchart which shows the flow of the main process performed by the control part of the document preparation apparatus in 6th Embodiment. It is a figure of one specific example of the alphabet key allocation used for the pinyin input in a mobile telephone. It is a figure which shows an example of the incorrect input of the keystroke place by the user in 6th Embodiment. It is a figure which shows an example of the incorrect input of the keystroke frequency by the user in 6th Embodiment. It is a flowchart which shows the flow of the main process performed by the control part of the document preparation apparatus in 7th Embodiment. It is a flowchart which shows the flow of the main process performed by the control part of the document preparation apparatus in 8th Embodiment. It is a functional block diagram which shows the detail of the erroneous input candidate recognition part of the document preparation apparatus in 9th Embodiment. It is a flowchart which shows the flow of the keystroke direction incorrect input process performed when a keystroke direction incorrect input process part is comprised in the control part of the document preparation apparatus in 9th Embodiment. It is a figure which shows an example of the incorrect input of the keystroke direction by the user in 9th Embodiment. It is a flowchart which shows the flow of the main process performed by the control part of the document preparation apparatus in 9th Embodiment. It is a functional block diagram of the document preparation apparatus in 10th Embodiment. It is a flowchart which shows the flow of the personal information acquisition process performed when a personal information acquisition part is comprised in the control part of the document preparation apparatus in 10th Embodiment. In this method, characters are input to the display device using a remote controller.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Document creation apparatus, 101, 101A control part, 102 Character input part, 103 Incorrect input candidate recognition part, 104 Kana / kanji conversion part, 105 Kana / kanji conversion determination part, 120 memory | storage part, 130 input part, 140 output part, 170 wireless Communication unit, 171 antenna, 180
Voice input / output unit, 200 Document creation device, 220 Storage unit, 230 Input unit, 240 Output unit, 250 External storage device, 251 Recording medium, 260 Communication unit, 301 Processing control unit, 302 Keystroke location error input processing unit, 303 Keystroke Number-of-times mistaken input processing unit, 304 Key input non-input processing unit, 305 Error input kana character string candidate storage processing unit, 315 Error input character string candidate storage processing unit, 501 Display processing unit, 502 Selection processing unit, 503 Personal information acquisition unit, 512 character input unit, 513 evaluation calculation unit, 514 replacement character string recognition unit, 515 display processing unit, 516 replacement character string selection reception unit, 604 kanji conversion unit, 605 kanji conversion determination unit, 606 keystroke direction incorrect input processing unit, 700 Display device, 800 remote control, 900 network.

Claims (24)

A character input means for accepting character input by a character input key;
For an input character string composed of one or more characters accepted by the character input means, using a replacement evaluation function as an index for replacing the input character string with a replacement candidate character string, and based on the replacement evaluation function Evaluation calculation means for calculating an evaluation value of the replacement candidate character string for the input character string;
A replacement character string certifying unit that certifies the replacement candidate character string whose evaluation value calculated by the evaluation calculation unit satisfies a predetermined condition as a replacement character string that replaces the input character string. A character input device.   The character input device according to claim 1, wherein the replacement evaluation function is a function for evaluating the possibility that the input character string is an erroneous input with respect to the replacement candidate character string. The replacement evaluation function is a function that gives a higher evaluation value to the replacement candidate character string as the input character string is more likely to be erroneous input,
The replacement character string recognition unit recognizes the replacement candidate character string whose evaluation value exceeds a predetermined threshold as a replacement character string for replacing the input character string. Character input device. The replacement evaluation function is a function that gives a higher evaluation value to the replacement candidate character string as the input character string is more likely to be erroneous input,
The replacement character string recognition unit recognizes a predetermined number of replacement candidate character strings as replacement character strings to replace the input character string in descending order of the evaluation value. Character input device.   When the input character string and the replacement candidate character string recognized as the replacement character string are Japanese, the input character string and the replacement character string are converted from kana to kanji, The character input device according to claim 1, further comprising character conversion means for column candidates.   When the input character string and the replacement candidate character string recognized as the replacement character string are Korean, the input character string and the replacement character string are converted into Korean characters by converting Hangul to Kanji. The character input device according to claim 1, further comprising character conversion means for column candidates.   If the input character string and the replacement candidate character string recognized as the replacement character string are Chinese, the input character string and the replacement character string are converted into Pinyin vs. Kanji characters, and converted characters The character input device according to claim 1, further comprising character conversion means for column candidates. Conversion character string candidate display processing means for displaying a list of the conversion character string candidates created by the character conversion means;
6. The apparatus according to claim 5, further comprising conversion character string selection receiving means for receiving selection of a conversion character string from conversion character string candidates displayed by the conversion character string candidate display processing means. Item 8. The character input device according to any one of Items 7 to 7.   The replacement evaluation function inputs a character input key that is keyed to input a character in the character input means and a character corresponding to the character in the replacement candidate character string for an input character string that includes the character. The character input device according to claim 1, wherein the character input device is a function that gives an evaluation value according to a physical distance from the character input key.   The replacement evaluation function inputs a character input key that is keyed to input a character in the character input means and a character corresponding to the character in the replacement candidate character string for an input character string that includes the character. The character input device according to any one of claims 1 to 8, wherein the character input device is a function that gives an evaluation value corresponding to an orientation with a character input key. The character input means accepts input of one character from a plurality of characters assigned to the character input key according to the number of times the character input key is pressed,
The replacement evaluation function includes the number of keystrokes of a character input key that is keyed to input a character in the character input means, and a character corresponding to the character in the replacement candidate character string for the input character string including the character. The character input device according to any one of claims 1 to 8, wherein the character input device is a function that gives an evaluation value corresponding to a difference from the number of keystrokes for inputting.   The replacement evaluation function can be added to a character including a character received by the character input means and a character corresponding to the character in the replacement candidate character string for an input character string including the character. The character input device according to any one of claims 1 to 8, wherein the character input device is a function that gives an evaluation value corresponding to a difference in symbols.   When the character received by the character input unit is Japanese, the replacement evaluation function is configured to include the character received by the character input unit and the replacement candidate character string in the replacement candidate character string for the input character string including the character. The character input device according to any one of claims 1 to 8, wherein the character input device is a function that gives an evaluation value according to a difference between a muddy note or a semi-voiced note from a character corresponding to the character.   The replacement evaluation function inputs a character input key that is keyed to input a character in the character input means and a character corresponding to the character in the replacement candidate character string for an input character string that includes the character. An evaluation value corresponding to a physical distance from the character input key for the character, a character input key keyed to input a character in the character input means, and the replacement candidate character string for the input character string including the character An evaluation value corresponding to the direction of the character input key for inputting the character corresponding to the character of the character, the number of times the character input key is pressed to input the character in the character input means, and the character An evaluation value corresponding to the difference in the number of keystrokes for inputting a character corresponding to the character in the replacement candidate character string with respect to the input character string to be received, and received by the character input means Evaluation of at least two of evaluation values according to a difference between a character and a character corresponding to the character in the replacement candidate character string with respect to an input character string including the character The character input device according to any one of claims 1 to 8, wherein the character input device is a function that gives an evaluation value calculated based on the value.   The replacement evaluation function inputs a character input key that is keyed to input a character in the character input means and a character corresponding to the character in the replacement candidate character string for an input character string that includes the character. An evaluation value corresponding to a physical distance from the character input key for the character, a character input key keyed to input a character in the character input means, and the replacement candidate character string for the input character string including the character An evaluation value corresponding to the direction of the character input key for inputting the character corresponding to the character of the character, the number of times the character input key is pressed to input the character in the character input means, and the character An evaluation value corresponding to a difference in the number of keystrokes for inputting a character corresponding to the character in the replacement candidate character string with respect to the input character string to be received, and received by the character input means Weighting each of the evaluation values according to the difference between the character and the character corresponding to the character in the replacement candidate character string with respect to the input character string including the character can be added to the character The character input device according to any one of claims 1 to 8, wherein the character input device is a function that gives a calculated evaluation value.   Weight updating means for updating the weight according to the difference between the character string before conversion to the conversion character string candidate received by the conversion character string selection receiving means and the input character string received by the character input means; The character input device according to claim 8, further comprising:   The character input device according to claim 8, wherein the conversion character string candidate display processing means displays the conversion character string candidates in a list in descending order of evaluation values of the replacement evaluation function.   The converted character string candidate display processing means converts the converted character string candidate converted from the input character string by the character converting means and the replacement character string recognized by the replacement character string authenticating means into converted character string candidates. Including a list of the conversion character string candidates, and when the conversion character string selection accepting unit accepts the conversion character string selection, the conversion character string candidates converted from the replacement character string by the character conversion unit are displayed. The character input device according to claim 8, wherein the character input device displays a list. Replacement character string candidate display processing means for displaying a list of the replacement character string candidates recognized by the replacement character string recognition means;
The replacement character string selection receiving means for receiving selection of a replacement character string from the replacement character string candidates displayed by the replacement character string candidate display processing means, further comprising: Character input device.   20. A document creation device comprising the character input device according to claim 1.   20. A document creation apparatus comprising the character input device according to claim 1, wherein the character input means is a remote controller. A step of accepting a character input by a character input key;
Targeting an input character string composed of one or more accepted characters, using a replacement evaluation function as an index for replacing the input character string with a replacement candidate character string, and based on the replacement evaluation function, the input character Calculating an evaluation value of the replacement candidate character string for a sequence;
And a step of recognizing the replacement candidate character string for which the calculated evaluation value satisfies a predetermined condition as a replacement character string for replacing the input character string. A step of accepting a character input by a character input key;
Targeting an input character string composed of one or more accepted characters, using a replacement evaluation function as an index for replacing the input character string with a replacement candidate character string, and based on the replacement evaluation function, the input character Calculating an evaluation value of the replacement candidate character string for a sequence;
A computer-executable program for causing a computer to execute a step of recognizing the replacement candidate character string for which the calculated evaluation value satisfies a predetermined condition as a replacement character string for replacing the input character string.   A computer-readable recording medium on which the character input program according to claim 23 is recorded.

Images