JP2011164894A - Content retrieval device and content retrieval method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To list the corresponding content with a small amount of calculation by inputting characters that show the reading of vocabulary. <P>SOLUTION: A first character string of metadata of the content is decomposed, a second character string which is vocabulary of the first character string is generated. The second character string and a character showing the reading of the second character string and the first character string are associated with a content ID and stored. Based on the reading inputted by a user for retrieving the content, the second character string is referenced. The first character string associated with the content ID is referred to from the second character string, and based on the first character string, the metadata of the content is displayed. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a content search device and a content search method, and more particularly to a content search device and a content search method for searching and displaying content easily and at high speed.

  In order for a user to select a desired program from an electronic program guide or a recorded program, the function of inputting characters and selecting a program related to the corresponding program name or detailed program information and presenting it to the user is a general television or It is mounted on recorders. This method makes a judgment by comparing character strings, and if you enter the characters described in the program name and program details, you can select the corresponding program without fail. The apparatus has a problem that it takes time to input the character string used for the search.

As a method for easily realizing such character input, for example, a function for a user to input a character string by voice input has been put into practical use in a PC (Personal Computer) or the like.
The voice input uses character string linking information corresponding to the reading of the character string, and when the user inputs the voice, the character string referred to from the corresponding reading is selected. In Patent Literature 1 and Patent Literature 2, by converting a vocabulary acquired in kanji from kanji to reading kana, it is possible to perform character conversion by voice input even for a new vocabulary without reading kana.

In Patent Document 3, morphological analysis is used to select a vocabulary, and the accuracy is improved. In this case, by performing morphological analysis on the character string and performing kanji kana conversion on the extracted character string, it is possible to give the reading kana relatively accurately even to information without the reading kana. it can.
These methods can also be applied to character input using a television remote controller or the like. For example, in a character input function such as a mobile phone or a PC, when the reading of some words is input, the corresponding vocabulary is displayed as input candidates.

JP 2001-22374 A JP 2003-174597 A JP 2007-140194 A

However, when searching for content such as a program, when inputting a reading name and selecting a vocabulary and presenting a list of contents including the vocabulary, it is necessary to compare and determine the character strings of the vocabulary and the content. However, for example, the EPG data for 8 days has a problem that the number of programs is 20,000 or more, and the amount of calculation for comparison determination is very large.
An object of the present invention is to provide a content search apparatus and a content search method for searching and displaying content easily and at high speed in order to solve the above-described problems.

  In order to achieve the above object, the present invention is a content search apparatus that receives and stores metadata having a first character string representing the content of content to be distributed, and searches for the content based on the metadata. A metadata holding unit that stores and holds the received metadata, and a vocabulary included in the first character string by decomposing the first character string of the metadata held by the metadata holding unit. A metadata decomposition unit that generates a second character string, a kanji conversion unit that gives a reading kana to the second character string generated by the metadata decomposition unit, and the second character string A vocabulary holding unit that stores and holds the kana including the reading kana given by the kanji conversion unit, a first character string held in the metadata holding unit, and a second character string held in the vocabulary holding unit Metadata to associate A vocabulary associating unit; a dictionary data holding unit that holds a third character string representing a word and a fourth character string representing a reading character of the third character string; and a user for searching for the content A character input unit for inputting the fifth character string, a display generation unit for generating a display of the search result of the content for the user, and a system control unit for controlling the operation of the content search device And the system control unit is held in the vocabulary holding unit via a fourth character string held in the dictionary data holding unit based on a fifth character string input from the character input unit. Referencing a second character string, and displaying the display of metadata of the retrieved content based on the first character string associated with the second character string by the metadata vocabulary association unit Generate by generator It is characterized by controlling the so that.

  The present invention is also a content search method for receiving and storing metadata having a first character string representing the content of content to be distributed, and searching for the content based on the metadata, wherein the metadata A first holding step for holding the character string; a character string generating step for decomposing the first character string of the metadata and generating a second character string that is a vocabulary included in the first character string; A metadata vocabulary associating step for associating a first character string with the second character string, a kanji kana conversion step for giving a reading kana to the generated second character string, and the second character string as A second holding step for holding the reading including the reading pseudonym, a user input receiving step for receiving a user input, and whether the input from the user input receiving step is a character input or a search A determination step for determining whether the content is an instruction to determine a content to be input, and when the determination step determines that the input from the user input reception step is a character input, The display of the metadata is created based on the above, and if it is determined that the input from the user input reception step is an instruction to determine the content to be searched, the display of the metadata of the determined content is created. It is a feature.

  According to the present invention, it is possible to provide a content search device and a content search method for searching and displaying content easily and at high speed, and there is an effect that it is possible to contribute to improvement of convenience for the user.

It is a whole block diagram in one Example of this invention. It is a flowchart which produces the corresponding vocabulary data from the metadata in one Example of this invention. 4 is a flowchart in which a user selects content using vocabulary data and metadata according to an embodiment of the present invention. It is an example of information of content metadata in one example of the present invention. It is an example of the information of vocabulary data in one Example of this invention. It is an example of the information of the dictionary data in one Example of this invention. It is an example of a display screen at the time of full-screen display in one Example of this invention. It is an example of a display screen at the time of carrying out area limited display in one example of the present invention. It is an example of a display screen in the case of searching another content from the content in one Example of this invention.

In the embodiment of the present invention, even candidate content vocabulary can be displayed even for content metadata without a reading kana, and the vocabulary can be displayed by associating the vocabulary with the content metadata. Since it is possible to easily refer to the content metadata, it is possible to search and display the content easily and at high speed by inputting the reading kana.
Also, the server device is provided with a function for converting the contents of the stream into character information and transmitted, and the client device is provided with a function for receiving and reproducing the character information so as to grasp the contents of the stream other than the stream being viewed. One of the features is that it is realized with a small hardware configuration.
Embodiments of the present invention will be described below with reference to the drawings.

  FIG. 1 is an overall configuration diagram of an embodiment of the present invention including a content search device 100 and an external content providing device 120. The content search device 100 has a function of displaying a list of content such as a television, a recorder, and a PC. The external content providing apparatus 120 is a broadcast station, a server on the Internet, a home content sharing server, or the like. As an example, when the content search device 100 is a television, the external content providing device 120 is a broadcast station or a server having a content distribution function. At this time, when content metadata such as EPG (Electric Program Guide) data is received, the metadata registration unit 101 registers (stores) the content metadata 400 in the metadata holding unit 102.

  FIG. 4 shows an example of the metadata 400. In FIG. 4, only a part of the EPG data (content ID 401 and title 402) is shown, but all of the acquired metadata may be included. The metadata holding unit 102 is a recording medium including a volatile / nonvolatile memory and a management mechanism thereof, and has a function of holding acquired metadata and data obtained by modifying the metadata. The metadata decomposing unit 103 has a function of processing a character string of metadata held in the metadata holding unit 102 or similar data. Examples of processing include morphological analysis of a character string and extraction of only independent words. Anything may be used as long as it has a function of decomposing a character string or similar data into two or more.

  FIG. 5 shows an example of data 500 decomposed by the metadata decomposition unit 103. In this example, in addition to the vocabulary 502 obtained by decomposing from the metadata 400, an example including a reading kana 503, a content ID (504), and an evaluation value 505 is shown, which will be described in detail later. The decomposed data 500 (hereinafter referred to as vocabulary data) is stored in the vocabulary holding unit 105. Further, it is not necessary to store all the decomposed data in the vocabulary holding unit 105. For example, attached words such as numbers and particles may be deleted.

  The metadata vocabulary association unit 104 associates the metadata 400 with the vocabulary data 500. The association includes, for example, giving the ID (504) of the decomposed content to the vocabulary data created by decomposing the character string of the content of the metadata 400. For example, when the content ID “34251” of the metadata 400 in FIG. 4 is decomposed into independent words, two vocabularies “world” and “documentary” are obtained. When registering these two vocabularies in the vocabulary data 500 of FIG. 5, if the IDs of the decomposed metadata are associated with each other by describing them or the like, the decomposition source metadata can be referred to from the vocabulary data. . For this reason, the content can be searched easily and at high speed. In association, vocabularies that have already been registered may be associated by adding the original metadata to the vocabulary data 500.

  The vocabulary holding unit 105 is a recording medium including a volatile / nonvolatile memory and a management mechanism thereof, and has a function of holding the vocabulary decomposed by the metadata decomposition unit 103 as vocabulary data 500. Note that the metadata holding unit 102 and the vocabulary holding unit 105 may have any implementation form as long as it has a function of inputting information and outputting necessary information. For example, information may be stored in another device via a network. The kana-kanji conversion unit 106 has a function of assigning a reading kana (a reading kana 503 in FIG. 5) to each vocabulary of the vocabulary data 500. Yomikana is not only related to a single kanji or idiom, but also includes kana notation of English words. Accordingly, the kana-kanji conversion unit 106 may include a plurality of dictionaries for converting a certain character string into a different character string.

  The user interface 107 displays metadata and vocabulary data. For example, when the content search apparatus 100 is a television, the user interface 107 may include an input interface such as a television screen and a remote control. In addition, as will be described later, when a user searches for content, the user may include this character input unit when inputting characters. The system control unit 108 performs operation control of each component including the metadata holding unit 102, the vocabulary holding unit 105, and the user interface 107 described above.

  FIG. 2 is an operation flowchart for creating corresponding vocabulary data from metadata in the content search method according to an embodiment of the present invention. When the process is started, the metadata decomposing unit 103 in FIG. 1 decomposes the metadata stored in the metadata holding unit 102 as shown in FIG. 4 and creates vocabulary data as shown by the vocabulary 502 in FIG. (Step S201). As a vocabulary data creation method, for example, a morphological analysis is performed and an independent word is used as vocabulary data. Next, the metadata vocabulary associating unit 104 associates the vocabulary data with the metadata (step S202). The association is, for example, giving the vocabulary data the metadata ID (for example, the content ID (504) in FIG. 5) from which the vocabulary data is created. Note that this step may be executed in step S201 in order to correlate the information obtained when decomposing in step S201.

  Next, the created vocabulary data is registered in the vocabulary holding unit 105 together with the association information (step S203). Registration may be performed at an arbitrary timing, for example, after the next step S204 is executed. Next, the kana-kanji conversion unit 106 performs kana-kanji conversion on the vocabulary data and assigns a reading kana (step S204). This completes the creation of the vocabulary data to which the reading is given from the metadata.

  FIG. 3 is a flowchart when the user selects content using vocabulary data and metadata in the content search method according to the embodiment of the present invention. When the process is started, it first waits for a user input (step S301). When the user inputs a character (character input in the figure), a matching search is performed (step S303).

  Matching search is a process of comparing input characters with vocabulary data to identify the corresponding vocabulary. For example, when a single kana character is input, it is checked against the reading kana 503 of the vocabulary data 500 and searched forward. And processing such as specifying the corresponding vocabulary. In the vocabulary data 500, when one character “se” is input and a corresponding prefix search is performed, the vocabulary ID 1 corresponds. In addition, when one character “shi” is input, the vocabulary IDs 9 and 11 are applicable, and when the dakuten and half dakuten are ignored, ID12 is applicable. When the second character is input, a forward matching search using the second character is performed.

  Note that the match search method is not limited to the forward match search, but may be a search considering vocabulary fluctuation or a search including synonyms. Further, the evaluation value 505 may be used during the main search. For example, like the vocabulary indicated by 702 in the display screen example of FIG. 7, the vocabulary data is displayed in a certain area of the higher vocabulary data (in this case, the evaluation value is 10 for Switzerland, 8 for shopping, 7 for sumo wrestling) And the like.

  Next, the vocabulary selected by the search is displayed (step S304). It should be noted that although the reading is the kana 503 or the vocabulary 502, other related information may be displayed. An example of a state in which the corresponding vocabulary is displayed is shown in FIG. 7, where the vocabulary and the reading kana are displayed in 703. When the corresponding vocabulary is listed in a format that can be selected by the user, the process returns to step S301, and the user subsequently inputs characters or performs a selection decision operation on the selected vocabulary.

  The selection determining operation refers to, for example, an operation in which “Seikai” or “World” of the vocabulary ID 1 is selected. Then, referring to the content ID (504) of the vocabulary data, the content of the corresponding content ID is acquired from the content metadata 400 and displayed (step S302), and the flow ends. In 705 of FIG. 7, information on the content selected from the vocabulary using the title 402 or the like is displayed.

  By selecting a vocabulary based on the input kana in this way and searching for the metadata content associated with the content ID, the content can be searched and displayed easily and at high speed.

  In addition to this flow, when a genre is selected after the start or at an arbitrary timing, only the vocabulary corresponding to the genre may be displayed. In this case, the genre information accompanying the content metadata may be inherited by the vocabulary, or the genre information may be given to the dictionary information of the kana-kanji conversion unit 106 in units of words. In this way, genre information can be given to each vocabulary of the vocabulary data 500, and a logical product of the genre information of the vocabulary data and the genre information input by the user is added to the search condition by the matching search 303, so that it corresponds to the genre. Vocabulary data can be displayed.

  FIG. 4 is an example of content metadata 400. In this example, only the content ID 401 and the content title 402 are shown, but any information related to the content may be described in the data. 401 is a unique ID assigned to the content. Normally, the content can be specified only with the real ID. 402 is the title of the content. In addition to these, for example, detailed information on content, character information, and the like can be given.

  FIG. 5 is an example of vocabulary data 500. 501 is a unique ID assigned to each vocabulary. Usually, the vocabulary can be specified only with this ID. 502 is a vocabulary. A vocabulary refers to a group of data that forms a meaningful group of words, compound words, or other words. 503 is a vocabulary reading kana. The reading kana may include not only kanji and idiom reading kana but also all data associated with the vocabulary, such as kana notation of pronunciation such as English. Reference numeral 504 denotes an ID of content metadata that is a source of the vocabulary. With this ID, content metadata can be referenced from the vocabulary.

  505 is a vocabulary evaluation value. The vocabulary evaluation value is a value used when determining the search / display order of the vocabulary, and may represent, for example, the number of reference content IDs for each vocabulary. Further, not a single value as in this example, but an evaluation value of any format as commonly used may be included. For example, when viewing content for a certain period of time, it detects that the content has been viewed and raises the evaluation value of the vocabulary that is a component of the content. For example, the evaluation value is distributed, or the evaluation value of the vocabulary included in the recorded content is increased. If an evaluation value set by a broadcasting station or server is distributed, the evaluation value of a vocabulary contrary to public order and morals can be lowered, or the evaluation value of a vocabulary having an advertising effect can be raised.

  In addition, this evaluation value may be provided with a device shared with other users. For example, a mechanism for uploading the evaluation value to the server or a mechanism for directly sharing the evaluation value between apparatuses may be used. By sharing the evaluation value, it is possible to increase the evaluation value of a popular keyword. Moreover, if the evaluation value of the vocabulary obtained from the content metadata of the content existing in the home is increased, the vocabulary according to the user's collection tendency can be preferentially presented to the user. In this method, first, character information such as titles and detailed information of contents in the home, including via a network, is collected. Next, the collected content metadata is decomposed in step S201. If the vocabulary obtained by the decomposition is registered in the vocabulary holding unit 105, the evaluation value of the corresponding content is increased. In this case, for example, if the target of the content metadata is only EPG data, and the home content is used only for setting the evaluation value, the display order of the vocabulary is determined from the content existing in the home, which reflects the user's preference Display order.

  FIG. 6 is an example of dictionary data 600. The dictionary data 600 includes at least a word 601 and a reading kana 602. Note that this data is data that is referred to in order to acquire the reading kana 602 from the word 601, and may be in any form as long as the reading kana can be acquired. Moreover, this dictionary may be updated via broadcasting or a network.

  FIG. 7 is an example of a display screen 700 when a full screen is displayed in one embodiment of the present invention. Reference numeral 701 denotes an area for inputting a reading pseudonym in step S301. In this example, the reading kana “shi” is input. An area 702 is displayed using the evaluation value 505. In this example, the top three evaluation values are displayed in descending order. Reference numeral 703 denotes an area for displaying the vocabulary corresponding to step S303, and the vocabulary 502 is displayed together with the reading kana 503. Reference numeral 704 denotes a focus. For example, the determination operation in step S301 is executed by pressing a determination button of an input device such as a remote controller. Note that the focus 704 may be the one in which the determination operation in step S301 is automatically executed when another vocabulary is focused. For example, when the user inputs one character and the focus is displayed as the initial state, the same processing as the determination operation in step S301 is performed, and the search result of the matching search is displayed in the area 705. Reference numeral 705 denotes an area for displaying a search result of the matching search, and the focus 704 may be able to move to this area.

  In this case, when content is selected, reserved recording of the corresponding content may be executed, or content may be played back. When content stored in the device or content is played back via a network, information for playing back the content, such as a URL (Uniform Resource Locator) of the content, may be stored in the content metadata 400. In addition, when the number of corresponding vocabulary becomes 0 by inputting 701 reading kana, the mode may be automatically switched to the kana-kanji conversion mode. The transition to the kana-kanji conversion mode means that the input mode of the 701 is read, not the kana input, but the kana-kanji conversion mode, and a normal IM (Input Method) is activated. Here, IM refers to a Japanese input tool (for example, Japanese input software) including the Kana-Kanji conversion unit 106. For example, as in Japanese romaji input, a single character is formed and output by a plurality of keystrokes, or when a hiragana is input and a conversion is instructed, a corresponding kanji or word candidate is output. . As a result, it is possible to support the input of a vocabulary in which the reading kana is not correctly assigned in step S202 due to reasons such as not being written in the dictionary 600.

  FIG. 8 is an example of a display screen 800 when the area limitation display is performed in one embodiment of the present invention. Reference numeral 801 denotes an area for inputting a reading pseudonym corresponding to 701. An area 802 displays a corresponding vocabulary corresponding to 703. A focus 803 corresponds to 704. Reference numeral 804 denotes an area for displaying a search result of the matching search corresponding to 705. A characteristic point of this screen example is that other content such as a broadcast screen can be searched and understood while the content is being displayed.

Further, in the above embodiment, when the reading kana is input, the vocabulary data is displayed and the content corresponding to the vocabulary data is displayed. However, by selecting the content as in the example of the display screen 900 shown in FIG. Corresponding vocabulary data may be displayed, and content including corresponding vocabulary data may be displayed. For example, assume that a program is selected in the electronic program guide and the function of the present invention is activated. The program selected at this time is 901 search source programs. The vocabulary associated with this search source program is shown at 902. By displaying other programs including 902 on 903, it is possible to obtain a search result without inputting a reading pseudonym.
With the configuration described above, even for content metadata having no reading name, it is possible to search for desired content easily and at high speed by using only the vocabulary from which the search result can be obtained by inputting the reading name.

  The overall configuration diagram, the flowchart, the display screen, and the like described so far are merely examples, and do not limit the present invention. Examples with some modifications can be considered, all within the scope of the present invention.

  101: Metadata registration unit, 102: Metadata storage unit, 103: Metadata decomposition unit, 104: Metadata vocabulary association unit, 105: Vocabulary storage unit, 106: Kana-Kanji conversion unit, 107: User interface, 108: System control Section 120: External content providing apparatus.

Claims (12)

A content search device that receives and stores metadata having a first character string representing details of content to be distributed, and searches for the content based on the metadata,
A metadata holding unit for storing and holding the received metadata;
A metadata decomposition unit that decomposes the first character string of the metadata held in the metadata holding unit and generates a second character string that is a vocabulary included in the first character string;
A kana-kana conversion unit for giving a reading kana to the second character string generated by the metadata decomposition unit;
A vocabulary holding unit for storing and holding the second character string including a reading kana given by the kanji kana conversion unit;
A metadata vocabulary associating unit that associates a first character string held in the metadata holding unit with a second character string held in the vocabulary holding unit;
A dictionary data holding unit that holds a third character string representing a word and a fourth character string representing a reading character of the third character string,
A character input unit for a user to input a fifth character string for searching for the content;
A display generation unit for generating a display of the search result of the content for the user;
A system control unit for controlling the operation of the content search device;
The system control unit
Based on the fifth character string input from the character input unit, the second character string held in the vocabulary holding unit is referred to via the fourth character string held in the dictionary data holding unit. The display generation unit generates a display of the metadata of the searched content based on the first character string associated with the second character string by the metadata vocabulary association unit. A content search apparatus characterized by:   The content search apparatus according to claim 1, wherein the method for generating the second character string from the first character string is morphological analysis.   2. The content search device according to claim 1, wherein the metadata vocabulary association unit detects whether or not the second character string has already been generated by a different first character string, and has already been generated. Is associated with a second character string generated by the different first character string.   The content search apparatus according to claim 1, wherein the metadata vocabulary association unit further associates the second character string with an evaluation value of the second character string.   The content search apparatus according to claim 4, wherein the evaluation value is a value determined in association with an operation on the content related to the first character string.   5. The content search device according to claim 4, wherein when the metadata vocabulary association unit generates the second character string using a different first character string, the metadata vocabulary association unit includes the different first vocabulary association unit. A content search apparatus that updates an evaluation value of a second character string generated by a character string.   5. The content search device according to claim 4, wherein the metadata vocabulary associating unit views content related to the first character string including the second character string, watches more than a certain time, schedules recording, records and watches, Alternatively, the content search apparatus updates the evaluation value of the second character string when recorded and viewed for a predetermined time or longer.   The content search device according to claim 1, wherein genre information indicating a genre to which the content belongs is assigned to the second character string, and the display generation unit is configured to display content of the content input from the character input unit. A content search apparatus that generates a display of metadata of the content searched based on a genre to which it belongs.   The content search apparatus according to claim 1, wherein when the second character string referred to based on the fifth character string does not exist, the kanji / kana conversion unit shifts to a kana / kanji conversion mode. A content search device.   The content search apparatus according to claim 1, wherein the system control unit selects content information related to the first character string, refers to the second character string, and refers to the second character string referred to. A content search apparatus that searches for information on content related to another first character string related to, and generates a display of content metadata related to the other character string in the display generation unit.   The content search apparatus according to claim 1, wherein the second character string is supplied to another apparatus. A content search method for receiving and storing metadata having a first character string representing details of content to be distributed, and searching for the content based on the metadata,
A first holding step for holding the metadata;
A character string generation step of decomposing the first character string of the metadata and generating a second character string that is a vocabulary included in the first character string;
A metadata vocabulary associating step associating the first character string with the second character string;
A kanji kana conversion step for giving a reading kana to the generated second character string;
A second holding step for holding the second character string including the reading kana,
A user input receiving step for receiving user input;
A determination step for determining whether the input from the user input reception step is a character input or an instruction for determining content to be searched;
If it is determined in the determination step that the input from the user input reception step is a character input, the display of the metadata is created based on the input character string, and the input from the user input reception step When it is determined that is an instruction to determine content to be searched, a display of metadata of the determined content is created.

Images