JPH11250069A - Information search support device - Google Patents

Abstract

(57) [Summary] [Problem] To provide an information search support device that can guide a user's search by presenting relevant information on search conditions in a data search. An information search apparatus includes a search condition input unit (1) for a user to input search conditions including a keyword, and a database (2) storing documents to be searched, and searches the database based on the search conditions. A related information dictionary 3 in which related information related to the keyword is stored, and the related information dictionary 3 based on the keyword.
And related information expanding means 2 for extracting and displaying related information corresponding to the input keyword, and related information displaying means 4 for displaying the related information, and a database based on the search condition. Prior to searching for the related information, the related information dictionary 3 is searched by the keyword or the database 6 is searched to extract the related information, and the related information is displayed.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a search dictionary for searching for a large amount of information stored in a database by using a search condition. By extracting the information from the database and organizing it and presenting it to the user, the user performs a guide when searching for further information.

[0002]

2. Description of the Related Art As shown in FIG.
1, database search means 22, final result display means 23
Etc. are provided.

[0005] As shown in FIG. 28 (B), when an information search is performed using this information search device, it is performed as follows. (S1) Enter search conditions. That is, the user inputs or selects a search condition.

(S2) According to the input search condition,
The database search means 22 searches the target database 21. (S3) The database search means 22 sends the information obtained from the search result to the final result display means 23 for display.

(S4) The user sees this display and determines whether or not the search conditions should be corrected. For example, when the number of search results is extremely large, the user corrects the search conditions and repeats the above steps S1 to S5.
3 is searched. If the search condition should not be corrected, the search is terminated.

As described above, in the conventional information search apparatus, after the user inputs or selects a search condition and executes a search, the number of data items corresponding to the search condition is displayed or The entity itself is displayed.

[0007] In many information retrieval devices, an operation that can be performed by a user is input of a retrieval condition, for example, a combination of AND, OR, and NOT of a retrieval keyword. The information presented to the user is either the number of data items or the data entity itself.

By the way, when the number of searched data exceeds several hundreds, the user is usually unable to check the contents of the searched data.
Using the R or NOT operator and a new search key, a narrow search is performed to reduce the number of search extracted data.

Conversely, even if the number of searched and extracted data is within several tens, if the user inspects the searched and extracted data and finds that there is little or no desired data, the user can perform AND and OR operations. Using a new search keyword and a NOT operator and a new search keyword, a re-search is performed to find another type of search data.

[0010]

In the prior art, when a user performs a keyword search by inputting a keyword, the result, that is, the number of searched and extracted data or the entity of the searched and extracted data is immediately transmitted to the user. Was presented to. At this time, the following two problems frequently appeared.

(1) A case where the number of searched and extracted data is several hundred or more and it is difficult for the user to check whether or not the data is desired. In this case, the user needs to perform a narrowed search to reduce the number of search extracted data by using an AND, OR, NOT operator and a new search key.

(2) When the number of retrieved data items is within several tens, and the user inspects the retrieved data,
When there is little or no data that meets your wishes.
In this case, the user needs to perform a re-search to find another type of search data using the AND, OR, and NOT operators and the new search keyword.

In any case, until the user obtains the desired data, the user devises a new keyword, or
It is necessary to re-input an appropriate search condition by devising a combination of / OR / NOT.

In the prior art, the properties, characteristics, and related information of a search condition or a data group retrieved and extracted immediately before are not presented to the user. Therefore, the user cannot use these properties, characteristics, and related information for search conditions that must be input in a narrow search or a new search.

Accordingly, an object of the present invention is to present related information on search conditions at the time of data search,
An object of the present invention is to provide an information search support device that can guide a user to search.

[0016]

FIG. 1 shows the principle configuration of the present invention.
Shown in In FIG. 1, 1 is a search condition input unit, 2 is a related information developing unit, 3 is a related information dictionary, 4 is a display unit, 5
Is a database search means, and 6 is a target database.

The above object of the present invention can be achieved by the following constitutions. (1) A search condition input unit 1 for a user to input a search condition including a keyword, and a database in which a document to be searched is stored, wherein the information search device searches the database based on the search condition. A related information dictionary 3 in which related information related to keywords is stored.
A related information expansion unit 2 for searching the related information dictionary based on the keyword, extracting related information corresponding to the input keyword, and displaying the extracted information, and a related information display unit 4 for displaying the related information. Prior to searching the database based on the search condition, the related information dictionary is searched by the keyword or the database is searched to extract the related information, and the related information is displayed.

(2) The related information expanding means 2 extracts a set of related terms such as synonyms and synonyms as related information,
Further, detailed information of the related terms is extracted from the related information dictionary or extracted by a preliminary search of the target database, and the related information display means 4 displays these related terms in a graphic format of a tagged file on the vertical axis or the horizontal axis. The detailed information of the related information of the designated tag is displayed and presented to the user as the foremost position, and the hidden detailed information is calculated as a background job.

(3) The related information developing means 2 extracts a set of related terms such as synonyms and synonyms as related information,
The related information display means 4 displays a two-dimensional tabular form in which those related terms are arranged on the related word on the vertical axis and the related word on the horizontal axis. The result of the AND search is displayed in the cell at the intersection of the vertical and horizontal axes.

(4) The related information developing means 2 extracts a set of related terms such as synonyms and synonyms as related information,
Further, the related information display means 4 displays the related terms in a graphical format of a tagged file on the vertical axis or the horizontal axis, and displays the related words specified on the vertical axis and the related words specified on the horizontal axis. The result of the AND search is displayed on the screen in the front row, and the hidden detailed information is calculated as a background job.

(5) The related information display means 4 displays a two-dimensional table format in which the related words and phrases are arranged in a related word on the vertical axis and a related word on the horizontal axis. On the outer side, an array of related words in which those related words are newly arranged is displayed, and the user sets the related words in the related word array outside the above table to (N-2) (where N is an integer of 3 or more). When the number of related words is specified, each cell of the two-dimensional table has (N-2) related words in the specified related word array, and a total of N related words on the vertical axis and related words on the horizontal axis. Is displayed in a cell at the intersection of the vertical axis and the horizontal axis.

(6) The related information display means 4 displays a two-dimensional table format in which the related words and phrases are arranged on the vertical axis and the horizontal axis, and further displays the two-dimensional table. On the outside, (N−2) (where N is an integer of 3 or more) one-dimensional arrays in which related terms different from the vertical and horizontal axes are newly arranged are displayed, and the user selects one of the one-dimensional arrays in the one-dimensional array. Related words (N-
2) When the specified words are specified, each cell of the two-dimensional table includes (N-2) related words in the specified related word array, a related word on the vertical axis, and a related word on the horizontal axis. The result of the AND search for the N related words is displayed in the cell at the intersection of the vertical and horizontal axes.

(7) As the related information dictionary 3, a time-series phrase database in which synonyms of keywords are determined based on dates and ages is held, and based on the keywords input by the user, the time-series phrase databases are used. Synonymous terms are extracted and displayed as a set of related words.

(8) As the related information dictionary 3, if the keyword is a word whose object is different depending on the situation or context,
Holding a situation phrase database in which synonyms of keywords are determined in advance by situations and contexts, and using the situation phrase database based on the keywords input by the user, extract phrases synonymous with the keywords, as a set of related words indicate.

(9) The related information developing means 2 derives a plurality of related words from the keywords, presents each keyword of the related words to the user in a table format, and displays the number of document data containing the keywords. At the same time, the appearance of these related words is checked for each document including the keyword, and the document in which the phrase appears is displayed in a table format.

(10) The related information developing means 2 derives a plurality of related words from the keywords, presents the keywords of the related words to the user in a table format, and displays the related words in a plurality of document databases. Examine the appearance of words and display in a document format which document databases these words appear in.

(11) The related information developing means 2 derives a plurality of related words from the keyword, presents each keyword of the related word to the user in a table format, and displays the number of document data including the keyword. At the same time, the appearance status of these related words is checked in each document including the keyword, and the number of appearances of those words in each document is displayed in a table format.

(12) The related information developing means 2 derives a plurality of related words from the keyword, presents each keyword of the related word to the user in a tabular form, and, in each document database including the keyword, The appearance of these related words is examined, and the document database in which those words appear is displayed in a table format.

Based on this, the following effects are obtained. (1) Before performing a data search, by obtaining relevant information from a search condition or from a database and displaying the relevant information to the user, the user can know the partial spelling without knowing the entire spelling of the keyword. As a result, a data search can be performed using the information, so that the data search can be supported. Further, since the related information is displayed, the keyword can be presented for the next search, and the function of the search support is achieved. It will be. In this way, the user can be provided with more information than can be obtained by a simple data search, and in the search again, more appropriate keyword selection and database operation can be performed.

(2) Display related words in a graphic format of a file with tags on the vertical or horizontal axis, and present only detailed information of the related information of the designated tag to the user as the forefront position, and hide the details. Since the information is calculated as a background job, it is possible to display all the detailed information at the forefront position that is easy for the user to understand, and even long detailed information. Moreover, since the hidden detailed information is calculated as a background job, even if the combination of related terms is changed, the detailed information can be displayed at the forefront in a short time.

(3) A set of related terms is extracted, and these related terms are displayed in a two-dimensional table format in which a related word on the vertical axis and a related word on the horizontal axis are displayed. Since the AND search result is displayed in the cell at the intersection, detailed information can be displayed by simple display means.

(4) A set of related terms is extracted and these related terms are displayed in the form of a tagged file on the vertical axis or the horizontal axis, and the related words specified on the vertical axis and the horizontal axis are specified. The AND search result with the related word is displayed on the front row screen, and the hidden detailed information is calculated as a background job. Therefore, even complicated ones can be displayed in an easy-to-understand and detailed manner for the user. Even if the combination is changed, the detailed information can be displayed at the forefront position in a short time.

(5) An array of related words in which those related words are newly arranged is displayed outside the two-dimensional table, and the user sets the related words in the related word array outside this table to (N-2). ), When a total of N related words of the designated (N-2) related words, related words on the vertical axis and related words on the horizontal axis are added to each cell of the two-dimensional table. Is displayed in the cell at the intersection of the vertical and horizontal axes, so even if the number of related terms is too large to be displayed in a row, N complex ANs are displayed.
D Searched items can be displayed.

(6) A one-dimensional array in which related terms different from those on the vertical and horizontal axes are newly arranged outside the two-dimensional table (N-
2) When this is displayed and the user designates (N-2) related words in the one-dimensional array, each cell of the two-dimensional table stores (N-2) of the designated related word array in the cell. Of N related words of related words, related words on the vertical axis and related words on the horizontal axis
Since the result of the D search is displayed in the cell at the intersection of the vertical and horizontal axes, the number of related words is very large and cannot be displayed even if one related word string is provided outside the two-dimensional table Even in such a case, even if a very large number of related words exist, the search state can be displayed. Furthermore, this (N-
2) By providing a one-dimensional array of books for each region, for example, for each region, for each manufacturer, or for general or special forms, it is possible to display a search state from a complicated viewpoint.

(7) For example, if the keyword is "Japanese Prime Minister /
"Prime Minister", "Emperor / Prince", "Rolling / Opposition"
For example, if the target is a phrase that differs depending on the time, such as "President of the United States," a related-information dictionary holds a time-series phrase database in which synonyms of the above keywords are determined by date and age, and the user inputs the phrase. Using the time-series word database, based on the keywords
Since words similar to the keyword are extracted and displayed as a set of related words, even if the user does not clearly recognize which target the keyword indicates,
Documents can be clearly searched in the documents of the document database.

(8) For example, if the keyword is a phrase such as "Prince (UK, Japan)" or "President (US, Russia)" depending on the situation or context,
As a related information dictionary, a situation phrase database in which synonyms of the above-described keywords are determined according to the situation and context is held in advance, and based on the keyword input by the user, the situation phrase database is used to categorize the keyword. Since the extracted words and phrases are extracted and displayed as a set of related words, even if the user does not clearly recognize which object the keyword specifically indicates, it can be clearly identified in the document of the document database. You can search for documents.

(9) Deriving a plurality of related words from a keyword, presenting each keyword of the related words to a user in a table format, displaying the number of document data containing the keyword, and displaying each keyword including the keyword. Since the appearance of these related words is checked in the entire document, and in which document the words appear, the detailed information can be displayed together with the number of cases.

(10) Deriving a plurality of related words from the keywords, presenting the keywords of the related words to the user in a table format, checking the appearance of the related words in a plurality of document databases, Displaying the document database in which the phrase appears in a tabular format enables the user to check whether the entered keyword is valid for the search and to determine which document database the user should further investigate with emphasis. Can guide the user.

(11) A plurality of related words are derived from the keyword, each keyword of the related word is presented to the user in a table format, the number of document data including the keyword is displayed, and each keyword including the keyword is displayed. By checking the appearance of these related words in the entire document and displaying the number of occurrences of those phrases in each document in a table format, it is possible for the user to check whether the entered keyword is valid for the search, Can further guide the user on which documents to focus on.

(12) Deriving a plurality of related words from a keyword, providing each keyword of the related words to a user in a table format, and examining the appearance of these related words in each document database including the keyword. Displaying the document database in which these phrases appear in a tabular format allows the user to confirm whether the entered keyword is valid for the search, and to further focus on which documents the user should investigate. Can be guided to the user.

[0041]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described with reference to FIGS. FIG. 1 shows an embodiment of the present invention, and FIG. 2 is an explanatory diagram of its operation state. In FIG. 1, 1 is a search condition input unit, 2 is a related information expanding unit, 3
Is a related information dictionary, 4 is a display means, 5 is a database search means, and 6 is a target database.

The search condition input means 1 searches for keywords in the target database 6 by using keywords, AND, OR,
A search condition such as a search expression in which an operator such as NOT is combined is input, and a format signal for specifying an expansion format when expanding related information is input as necessary.

The related information developing means 2 is a search condition input means 1
The related information dictionary 3 is accessed based on the keywords entered in the search condition transmitted from the user to extract related words, and the related words are displayed on the display means 4 in a specified format, and the number of appearances of the related words is counted. Or multiple related words A
The number of appearances according to the ND condition is obtained and displayed, or the target database 6 is accessed to extract relevant information and display it on the display means 4 in a specified format. In the example of FIG. 1, a case will be described in which a format as shown in FIG. 2B is specified in advance as a display format of the related information.

The related information dictionary 3 stores related information. Some related information can be extracted using a thesaurus (related word dictionary). Therefore, a thesaurus 3-1 is prepared. Details of the related information will be described later.

The display means 4 displays the related information to the user and displays the result of searching the target database 6 based on the search condition. Database search means 5
Is for searching the target database 6 based on search conditions input by the user.

The target database 6 stores documents to be searched. Next, related information for supporting a search in the present invention will be described. Such related information includes the following.

(1) Related words that can be extracted by a thesaurus (related word dictionary) based on the input keywords. The related compound phrase extracted by this related word dictionary can be a new input keyword.

(2) Related compound words that can be extracted by the target database based on the input keyword. The related compound phrase extracted by the target database can be a new input keyword.

(3) Appearance information that can be extracted by the target database based on the input keyword. Hereinafter, these (1) to (3) will be described in more detail.

(1) Related words that can be extracted by the thesaurus include the following. (1-1) Synonyms, words having the same meaning (1-2) Synonyms, words having a similar meaning (1-3) Opposite words, words having the opposite meaning (1-4) Siblings, eg, giants / Hanshin / Yakult, etc. (1-5) Higher words such as giants → baseball teams → professional baseball → ball games, etc. (1-6) Lower words (1-7) Foreign language translations, such as “computer” for “electronic computer” (1-8) Compound words, such as "Giant" for "Giant Hanshin" (1-9) Partial words, for example, "Giant" for "Giant Hanshin" (1-10) Case words for class words For example, “Hashimoto” for “Japanese Prime Minister” and “Charles” for “Prince Prince” can be used as new input keywords. FIG. 3 shows a procedure for extracting these related words and related compound words.

S1, In FIG. 1, a user inputs a search condition such as a search keyword or a search formula into the search condition input means 1. These input data are displayed on the display means 4 via the related information developing means 2. The user
When it is determined that a related word of the keyword needs to be checked, a search signal "1" for instructing a related word search is input from, for example, a keyboard (not shown).

S 2, this investigation signal “1” is transmitted from the search condition input means 1 to the related information developing means 2. Thereby, the related information developing means 2 searches the thesaurus 3-1 held in the related information dictionary 3 by the keyword to obtain the related information. Then, the target database 6 is searched based on the related information, and the frequency of occurrence can be counted.

S3, the related information expanding means 2 converts the related word obtained by searching the thesaurus 3-1 into, for example, the one shown in FIG.
The list is displayed on the display means 4 as the related information 1, 2,... At this time, the appearance frequency can be displayed.

S4: The user selects a related word necessary for the search from the search results listed on the display unit 4. S5, and the user creates a search condition using the selected related word as a new keyword. At this time, when it is deemed necessary to further investigate the related word for the new keyword, the process returns to S1 to extract the related word for the new keyword. However, when the user determines that the related word search is unnecessary, a search signal "0" is input, whereby a search formula is created without proceeding with the related word search, and the database search means 5 generates the target database 6 based on this. Search for. Then, the search result is displayed on the display unit 4.

(2) Related compound phrases that can be extracted by the target database are as follows. (2-1) Compound words such as "1-8" as in "1-8" and "Giants in Hanshin" etc. (2-2) Partial words such as "1-9" in "Giants" (2-3) Reconstructed compound words, such as "Giant Hanshin Battle" against "Giant and Hanshin Battle" etc. (2-4) Appeared frequently in the document group The related compound phrase extracted by the target database can be a newer input keyword. FIG. 4 shows a procedure for extracting these related compound phrases.

S1, In FIG. 1, a user inputs a search condition such as a search keyword or a search formula into the search condition input means 1. These input data are displayed on the display means 4 via the related information developing means 2. The user
When it is determined that it is necessary to check the related word of the keyword, for example, a check signal “1” is input from a keyboard or the like (not shown).

S2: Next, the user specifies an extraction unit.
The extraction unit is "keyword only / line / sentence / paragraph /
Section / chapter / document / file / other user designation etc. "

S3, related information expanding means 2
Searches the target database 6 in the extraction unit. Then, the related compound words shown in (2-1) to (2-4) are searched and extracted.

S4: The related information expanding means 2 displays a list of extraction units as a search result obtained by searching the target database 6 at S3, for example, as shown in FIG. , 2... Are displayed on the display means 4.

S5: The user selects a related compound word necessary for the search from the list displayed on the display means 4. S6, and the user creates a search condition using the selected related compound word as a new keyword. At this time, if it is deemed necessary to further investigate related words for the new keyword, the process returns to S1 to extract related words. However, when the user determines that the search for the related word is unnecessary, a search signal “0” is input, whereby a search formula is created without proceeding with the search for the related word.
Searches the target database 6 based on this. Then, the search result is displayed on the display unit 4.

(3) Appearance information that can be extracted from the target database includes the following. For example, there are the number of appearances of all the phrases in the document data group, the top n words in the document data group, that is, the frequently appearing phrases.

The general operation of the apparatus shown in FIG. 1 is as shown in FIG.
This will be described with reference to the flowchart shown in FIG. S1, a user inputs an appropriate search condition by combining a keyword group with an operator such as AND / OR / NOT in the search condition input means 1.

S 2, related information developing means 2 develops related information relating to the search condition using the related information dictionary 3 based on the input search condition, or develops related information using the target database 6. I do. For example, a synonym is developed using a synonym dictionary as the related information dictionary 3, or a developing process is performed using the target database 6 to extract the appearance frequency of data corresponding to the search condition as related information. At this time, the related information developing means 2 may search the target database 6 as preprocessing in order to calculate the appearance frequency and the like.

S3, the related information expanding means 2 expands the related information and sends out the expanded result to the display means 4 to display it to the user in the form of a list, a two-dimensional table described later, or the like. S4: The user looks at the related information displayed on the display means 4 and determines whether the search condition is appropriate. If the search condition is not appropriate, correct it and add a new keyword or AN
A combination of D, OR, NOT, etc. is made, and re-input from the search condition input means 1, and the process returns to S1.

S5: If the search condition is appropriate, the database search means 5 performs an actual data search on the target database 6. S6, and the database search means 5 displays the search result on the display means 4.

S7, the operator looks at this search result and
Correction of search conditions For example, if it is determined that refinement is necessary, the correction is made and the process returns to S1. However, if it is determined that the correction of the search condition is unnecessary, the search ends.

FIGS. 5 to 26 show a second embodiment of the present invention.
It will be described based on. FIG. 5 shows a second embodiment of the present invention and its operation explanatory diagram, FIG. 6 shows a two-dimensional display format of related information of the present invention, and FIG. 7 shows a tag display format of related information of the present invention. FIG. 8 shows the AND of N pieces of related information of the present invention.
FIG. 9 shows an AND display format of two related phrases of the present invention, FIG. 10 shows a correlation display format between two related phrases of the present invention, and FIG.
FIG. 1 shows an AND display format of a tag format of two related phrases.
2 indicates an AND display format of N related phrases of the present invention;
13 shows an AND display format of N related axes of the present invention, FIG. 14 shows an AND display format of N related axes of the present invention, and FIG. 15 shows a time-series database of related words of the present invention. FIG. 16 shows a display format of the preliminary search of the related word status database of the present invention, and FIG.
18 shows an appearance display format for each data of the related information of the present invention. FIG. 18 shows a distributed database (D) of the related information of the present invention.
FIG. 19 shows an appearance number display format for each data of related information of the present invention, and FIG. 20 shows an appearance number display format for each database of related information of the present invention. 21 shows an example of a procedure of the related information expanding means of the present invention, FIG. 22 shows a synonym appearance frequency display format of the related information of the present invention, and FIG. 23 shows a synonym appearance frequency tag of the related information of the present invention. A display example is shown, 24 shows an example of the procedure of the expansion means of the related information of the present invention, FIG. 25 shows an appearance frequency display format of a high-frequency phrase of the related information of the present invention, and FIG. 26 shows the related information of the present invention. 3 shows a tag display format of a high-frequency phrase.

In FIG. 5 (A), the same symbols as those in the other figures indicate the same parts, 7 is related information expansion determining means, and 8 is expansion format storage means. The related information development determining means 7 determines whether or not the related information should be developed, and when displaying the related information, determines based on which of a plurality of display formats to be displayed later.

The expansion format storage means 8 stores a plurality of display formats in which related information is to be displayed. For example, the expansion format storage means 8 is constituted by a ROM.

Next, the operation of the second embodiment shown in FIG. 5A will be described with reference to the flowchart shown in FIG. S1. The user inputs a keyword group (including the case of one keyword) in the search condition input means 1 by AND / O.
Enter appropriate search conditions combined with R / NOT, etc.
In addition, a search signal for requesting search output display of related information, an extraction unit signal for specifying a unit for extracting related information, a display format signal indicating a display format of related information, and the like are input. The search condition input means 1 transmits these inputs to the related information development determining means 7.

S2. The related information development determining means 7 decodes these signals, determines whether or not the related information should be developed, and determines that the related information should be developed when the investigation signal is, for example, “1”. Expansion of related information expansion means 2
To instruct. At this time, when the search signal is "0", it is determined that the search is of the conventional system in which the related information is not expanded, and the process proceeds to S6 described later, where the database search means 5 searches the target database.

S3. By this instruction, related information developing means 2
Develops related information related to keywords using the related information dictionary 3 or the target database 6. For example, a synonym, an upper word, or a lower word is extracted by using a synonym dictionary of the related information dictionary 3 or extracting a synonym using the thesaurus 3-1. Further, the frequency of appearance of related information corresponding to the search condition is extracted using the target database 6. At this time, the related information developing means 2 searches the document database 6 as preprocessing to calculate the appearance frequency and the like.

S4. After developing the related information, the related information developing means 2 sends out the developed result to the display means 4. At this time, the related information development determining means 7 decodes the display format signal and recognizes the format to be displayed, accesses the development format storage means 8 based on this, reads out the format to be displayed, and reads it out. To send to. The display means 4 writes the development information in the transmitted format and displays the development information to the user.

S5. The user sees the related information displayed on the display unit 4 and determines whether or not the search condition should be corrected.
If it is determined that the search condition should be corrected,
Return to step 1 and enter the search condition again.

S6. In S5, if the user determines that it is not necessary to correct the search condition, the user inputs this fact. This signal and the search condition are transmitted to the database search unit 5 via the related information expansion determination unit 7.
Then, the database search means 5 searches the target database 6 based on the search condition.

S7. The database search means 5 sends the search result to the display means 4, and the display means 4 displays the search result. S8. The user determines whether or not the search condition should be corrected based on the search result. If the user determines that the search condition needs to be corrected, the process returns to S1 and reenters the search condition. However, if it is determined that the search conditions do not need to be modified, the search is terminated.

The feature of the present invention is that FIG.
This relates to the display of related information in S4 of (B), which will be described in detail below. FIG. 6 shows a case of a two-dimensional display format as a display format of related information, and is an example of “computer” as an input keyword.

FIG. 6A shows a case where the related information 1, 2, 3, 4,... Obtained by the input keyword "computer" are displayed side by side on the display surface. Detailed information for each related information will be displayed below the related information.

FIG. 6B shows a case where the related information 1, 2, 3, 4... Obtained by the input keyword "computer" are displayed side by side in the vertical direction on the display surface. in this case,
Since not only the related information 1, 2, 3, 4,... Can be easily displayed even if the related information is long, the detailed information for each related information can be displayed beside each related information.
The display area of the detailed information can be enlarged as compared with the case of FIG. Therefore, even when detailed information such as the number of output cases or the document name is long, it can be easily displayed.

FIG. 7 shows a case where a tag display format is used as a display format of related information, and is an example of "computer" as an input keyword. FIG. 7A shows a case where the related information 1, 2, 3,... Obtained by the input keyword “computer” is displayed as tags arranged in the horizontal direction on the display surface. In FIG. 7A, even if the detailed information of the related information 1, 2,... Is very long, the display area is large, so that it can be easily displayed. FIG. 7A shows the related information 2
Here is an example in which details are displayed.

FIG. 7B shows a case where the related information 1, 2, 3,... Obtained by the input keyword "computer" is displayed as tags arranged in the vertical direction on the display surface.
In FIG. 7B, even if the related information 1, 2, 3,... Is long, it can be easily displayed, and even if the detailed information is very long, for example, when the document title is long, the display area thereof Is large, so that it can be displayed easily. FIG. 7B shows an example in which the related information 2 is displayed in detail.

In the case of the tag display format, only the detailed information of the related information of the designated tag is presented to the user as the foremost position, and the hidden detailed information can be calculated as a background job.

FIG. 8 shows a tag display format of an AND search of N pieces of related information, and is an example of “computer” as an input keyword. FIG. 8A shows related information 1, 2, 3, 4, 5,... Obtained by the input keyword "computer".
Shows a case where... Are displayed side by side on the display surface, and a case where an AND search result of the related information 2 and the related information 5 is displayed is shown. In this case, the related information 2 and the related information 5 to be AND-searched can be shown to the user by, for example, blinking.

FIG. 8B shows related information 1, 2, 3, 4, 5,... Obtained by the input keyword “computer”.
Are displayed side by side in the vertical direction of the display surface, and among these, the case where the AND search result of the related information 2 and the related information 5 is displayed is shown. In FIG. 8B, related information 1, 2, 3,...
・ Even if the length is long, it can be displayed easily. In addition, the related information 2 and the related information 5 are represented by A
The ND search item may be indicated, or the line on the right side of the frame of the related information 2 and the related information 5 may be removed and displayed to display A.
An ND search item may be indicated.

FIG. 8 also shows the tag display format.
Only the detailed information of the related information of the specified tag is presented to the user as the foremost position, and the hidden detailed information can be calculated as a background job.

FIG. 9 (A) shows an AND display format of two related terms, and the AN of two related terms of related terms 1, 2, 3,... Obtained by the input keyword "computer".
D display is shown. In FIG. 9 (A), “related word 1 AND computer” is 120 times, “related word 1 AND related word 3 AND computer” is 32 times,... This indicates that the number of documents for which information has been detected is 121.

According to the flow chart of FIG.
The search procedure for the two-dimensional arrangement (A) will be described. S10. In FIG. 5, when search data such as a search condition display format is input to the search condition input means 1, the search data is transmitted to the related information developing means 2 and the like via the related information development determining means 7.

S11. If the related information developing means 2 determines that the related information should be developed, the process proceeds to the next step S12, where control for developing the related information is performed. If it is determined that the related information should not be developed, the process proceeds to S14.

S12. When it is determined that the related information is to be developed, the related information developing means 2 obtains the related information, that is, the related words from the related information dictionary 3 and counts the number of appearances of these related words from the target database 6. Then, the developed related words are sent to the display means 4. At this time, since the related information development determining means 7 extracts the display format specified by the display format signal from the development format storage means 8 and sends it to the display means 4, the display means 4 First, FIG. 9 shows the transmitted related words.
(A) is sequentially entered in the related word entry area in the horizontal axis direction. For example, related words 1, related words 2,... Are sequentially written in the horizontal direction, and the number of appearances is written. In this way, the related information is displayed.

S13. The operator sees this and determines whether or not the search condition should be corrected. If the operator determines that the search condition should be corrected, the process returns to S10. S14. If it is not corrected, the related information developing means 2 performs an AND search between the related words to obtain the number of times for two-dimensional display. Then, the related word obtained in S12 is converted to the AN
D The result is sent to the display means 4 together with the search result. At this time, the display unit 4 sequentially writes the transmitted related words in the related word entry area in the vertical axis direction in FIG. 9A, and also writes the AND search result. For example, related words 1, related words 2,... Are sequentially written in the horizontal direction, and the AND search result is displayed together with related information.

S15. The operator sees this and determines whether or not the search condition should be corrected. If it is determined that the search condition should be corrected, the process returns to S14. S16. If it is determined that it should not be corrected, the database search means 5 searches the target database 6 based on the search conditions.

S17. The operator sees this and determines whether or not the search condition should be corrected.
Return to, and terminate if not corrected. FIG. 10 shows a correlation display between two related phrases, where the related words obtained by the input keyword “computer” are “hardware”, “software”, “PC”, “internet”, “printer”,. One thing, "Hardware AND
"Computer" 120 times, "Hardware AND
"PC AND computer" 32 times, "software AND computer" 147 times, "software AND printer AND computer" 7 times
Time: Indicates that the search was performed in the target database and the number of documents in which the relevant information was detected in the target database was 121.

FIG. 11 shows a case where AND display of two related terms is displayed in a tag format, and shows an AND display of related terms 1, 2, 3,... Obtained by the input keyword "computer". . FIG. 11 shows a display of a search result of “related information 2 AND related information 4 AND computer”.

In the case of FIG. 11 as well, since the tag display format is used, the result of the AND search for the specified related word is displayed on the screen in the front row, and the hidden detailed information is calculated as a background job. Can be.

FIG. 12 shows an AND display of N related phrases. In the case of FIG. 12, the input keyword “computer”
... of the related words 1, 2, 3, 4, 5, 6,... Obtained in accordance with the common search condition “computer AND related word 2 AND related word 5” and A of related words 11 and 23
When the ND condition is used, the search is performed 32 times, and when the AND condition of the related words 13 and 26 is used, the search is performed 12 times.
Indicates that there are cases.

FIG. 13 shows a case where AND display is performed on N related axes. In this case, the related axis 1 indicates a document in which, for example, a maker name is entered, the related axis 2 indicates a document in which the type of a computer (for example, a personal computer or a general purpose) is entered, and the related axes 3, 4, 5 Indicates documents in which the names of regions in the United States, Japan, and Europe are entered. Then, a related word B is selected from the related axis 3, a related word d is selected from the related axis 4, and a related word c is selected from the related axis 5, and these AND conditions are obtained. 13, 14, 1
5 ... from related axis 2 to related words 21, 22, 23, 2
A case where AND conditions based on 4, 25, and 26 are obtained is shown. According to FIG. 13, among the related words obtained by the input keyword “computer”, the search condition “computer”
AND related word B AND related word d AND related word c ”plus 1 for the AND condition of related words 11 and 21
The AND condition of the related words 11 and 23 indicates that the search was performed 32 times, and that the number of documents in which the related information was detected is 121.

FIG. 14 shows a case in which the related axes 1 to 5 are displayed in the tag display format in FIG. In this case, since the display area of the search result is large, the display of the AND search result, that is, detailed information can be displayed in detail.

In this case also, since the tag display format is used, the hidden detailed information can be calculated as a background job. FIG. 15 shows a case where a time-series database is prepared as a related information dictionary for a keyword whose target changes with time as related information, and the target object is identified.

For example, the keywords are “Japanese Prime Minister / Prime Minister”, “Emperor / Prince”, ruling party / opposition party, “US President”
For example, if the object is a phrase that varies depending on time, a time-series phrase database in which synonyms of the keyword as described above are determined according to date and age, and based on the keyword input by the user, the time-series phrase database is used. Using the phrase database, phrases similar to the keyword are extracted and displayed as related words.

FIG. 15 shows the keyword “Japanese Prime Minister” as a keyword.
Shows the related information display when is input. Displaying the prime minister's name for a certain time span and its term provides accurate relevant information.

If the keyword is "(UK, Japan)
"Prince", "President (of the United States, Russia)"
Maintain a situation phrase database that defines synonyms of the keyword as described above according to the situation and context, and based on the keyword input by the user, extract a phrase synonymous with the keyword using the situation phrase database, It must be a word.

For this reason, as shown in FIG. 16A, when "Prince" is input as an input keyword, the name of the Japanese Crown Prince and the name of the British Crown Prince are displayed as related information. . Also, "President" as an input keyword
When a is input, the most frequently appearing information, such as the names of the US president, the name of the Russian president, and other presidents of the Middle East, is displayed as related information.

FIG. 17 shows a display format of the relation state between the related information and the document when the keyword is “computer” and the related information is detected, for example, 121 in the target database. FIG.
The related documents “Hardware”, “Software”, “PC”, “Internet”, “Printer”, etc. obtained by keyword “Computer” are displayed horizontally. Are displayed side by side, and for each document, those in which the related information is described are indicated by circles. In FIG. 17A, “hardware” and “printer” are described in the first document, and “hardware”, “software”, and “internet” are described in the second document.
Indicates the case where is described.

FIG. 17B shows the first to 121st documents side by side and displays related information “hardware”, “software”, “PC”, and “computer”.
This shows a case where "Internet" is displayed vertically. In FIG. 17B, “hardware” is described in the first document, and “hardware” is described in the second document.
The case where "software", "Internet", and the like are described, and "PC" and "Internet" are described in the third document are shown.

FIG. 17C shows the related information “hardware”, “software”, “PC”, Internet, “printer”, etc. obtained by the keyword “computer” in the tag format in the horizontal direction. Then, which of the related information is described for each document is displayed in the central display area. In the example of FIG. 17C, “hardware” (for example, the fact that it is being selected by the bling display) indicates that the first document, the second document, the fourth document, the fifth document,・ ・ It is shown in the description.

FIG. 17D shows related information “hardware”, “software”, “PC”, “internet”, etc. obtained by the keyword “computer”.
Are displayed in the tag format in the horizontal direction, and which of the related information is described for each document is displayed in the central display area. In the example of FIG. 17C, it is shown that the related information “PC”, “Internet”,.

FIG. 18 shows a plurality of related information “hardware”, “software”, “PC”, “internet”, “printer”... Arranged side by side or vertically from a keyword “computer”. In each document database, the appearance status of the related information is checked, and in which document database the related information appears is displayed in a table format.

In FIG. 18, databases DB1 and DB
2, DB3, DB4,... Constitute a distributed database. And databases DB1, DB2, DB3,
DB4... Can be configured according to a tendency such as, for example, countries such as the United States, Japan, and Europe, or companies.

FIG. 18A shows the related information “hardware”, “software”, “PC”, “internet”, “printer”... Arranged side by side, and “hardware” and “hardware” in the database DB1. "Printer" is described, and "Hardware", "Software", and "Internet" are described in the database DB2.

In FIG. 18B, databases DB1, D
B2, DB3,... Are displayed side by side, and "hardware" is the database DB1, DB2, DB4, DB5,.
・ ・ "Software" is a database DB
2...

FIG. 19 shows the number of appearances of related information for each document. A plurality of related information "hardware", "software", "personal computer", and "internet" derived from the keyword "computer" are displayed. , "Printer" ... are displayed side by side or vertically, and the number of appearances of these related information in a plurality of documents is displayed in a table format. Indicates that the number is 121.

In FIG. 19A, related information “hardware”, “software”, “PC”, “internet”, “printer”... Are displayed horizontally, and documents are displayed vertically. The number of appearances of the related information is displayed for each document. In the first document, the related information “hardware” is displayed seven times, “printer” is displayed once, and so on.
In the second document, "hardware" appears five times, "software" appears six times, and "Internet" appears four times.

In FIG. 19B, 121 documents are displayed side by side, and related information “hardware”, “software”, “PC”, “internet”, “printer”. The related information appearing in each document is displayed side by side by the number of times.

FIG. 20 shows databases DB1 and D of the related information “hardware”, “software”, “PC”, “Internet”, “printer”,.
B2, DB3, DB4,... Are displayed in a table format. In FIG. 20A, the related information “hardware”, “software”, “PC”, “internet”, “printer” .. Are displayed horizontally, and the databases DB1, DB2,.
The number of appearances of the related information is displayed for each database. The database DB1 indicates that the related information “hardware” appears seven times, the “printer” appears once,. hardware"
5 times, "Software" 6 times, "Internet"
... Appear four times.

In FIG. 20B, the database DB
1, DB2, DB3, DB4... Are displayed horizontally, and related information “hardware”, “software”, “PC”, “Internet”. Each related information is displayed by the number of times.

Next, examples of related information development will be described with reference to FIGS.
Will be described based on the flowchart shown in FIG. FIG. 21 is an explanatory diagram of the operation of the expansion procedure of the related information expansion means.
FIG. 22 shows an example of the related information display in the case of the appearance frequency of a synonym, and FIG. 23 shows the synonym in a tag display format.

S20. In FIG. 5, when a search condition is input to the search condition input unit 1, this is transmitted to the related information developing unit 2 via the related information expansion determining unit 7. S21. Thereby, the related information developing means 2 expands the related word from the search condition using the related word dictionary 3 or the target database 6. For example, a set of synonyms of the keyword in the search condition is extracted, and if the input keyword is “computer”, the synonyms “computer”, “electronic computer”, “computer”, “computer”, "Computer" is extracted.

S22. Next, the related information developing means 2 calculates the appearance frequency of each related word in the target database 6, in this example, the appearance frequency of the synonym, that is, the number of appearances. The number of appearances may be checked in advance how many times each synonym appears in the target database 6, and the number of appearances may be stored in the related information dictionary 3 as an attribute of the synonym. Alternatively, a keyword full-text search may be actually performed in the target database 6 for each synonym to count the number of appearances.

S23. After obtaining the related words and their appearance frequencies in this way, the related information developing means 2 displays them on the display means 4. The display means 4 displays a pair of a synonym and the number of appearances of the synonym, as shown in FIG. In this example, if the entered keyword is "computer"
The number of appearances of the synonym “computer” is 121
And that the number of appearances of “electronic computer” is 34.

S24. The user sees this display and determines whether or not to modify the search condition.
Return to 20. S25. If no correction is required, the database search means 5 searches the target database 6 based on the search conditions. Then, the search result is displayed on the display unit 4.

The display format of the related information is not limited to the format shown in FIG. It can be displayed in another format, for example, a tag display format as shown in FIG. FIG.
In the example of, each synonym is displayed in a file format with a tag, the number of appearances of the synonym “electronic computer” is displayed as 34, and the number of appearances of other synonyms cannot be hidden. When displayed in this way, when the number of occurrences has already been calculated is displayed on the front side, and for synonyms for which the number of occurrences has not been calculated, only tags are displayed, and these can be calculated in the back job. There is also an advantage that the number of appearances of no synonyms can be calculated.

Another example of developing related information will be described with reference to FIGS. 24 to 26 based on the flowchart shown in FIG. FIG. 24 is an explanatory diagram of a developing procedure operation of the related information developing means when a high-frequency phrase is displayed as related information, FIG. 25 shows an example of related information display in the case of the appearance frequency of the same frequency phrase, and FIG. A high-frequency phrase is displayed in a tag display format.

S30. In FIG. 5, when a search condition is input to the search condition input unit 1, this is transmitted to the related information developing unit 2 via the related information expansion determining unit 7. S31. Thereby, the related information developing means 2 extracts the high-frequency words and phrases from the target database 6.

S32. Next, the related information developing means 2 extracts a high frequency word from the target database 6. S33. The related information developing means 2 extracts the appearance frequency of each high-frequency word, arranges the high-frequency words in the order of the appearance frequency, and sends the words to the display means 4 together with the appearance frequency. Based on this, as shown in FIG. 25, the display means 4 displays the high-frequency words in pairs with the frequency of appearance, that is, the number of appearances, together with the number of appearances. In this example, "hardware", "software"
The number of appearances is large in the order of “PC”, and the number of appearances is 226, 179, 158,.

S34. The user sees this display and determines whether or not to modify the search condition.
Return to 30. S35. If no correction is required, the database search means 5 searches the target database 6 based on the search conditions. Then, the search result is displayed on the display unit 4.

The display format of the related information is not limited to the format shown in FIG. It can also be displayed in another format, for example, a tag display format as shown in FIG. FIG.
In the example, each high-frequency phrase is displayed in a tagged file format, the number of appearances of the high-frequency phrase “software” is displayed, and the number of appearances of other high-frequency words is not hidden.

Next, as shown in FIG. 27, as the related information that can be extracted from the target database, the number of appearances of all the terms in the document data group, that is, all the words in the target database, the number of occurrences of the target database in the top n frequently occurring words, A procedure for extracting the number of appearances of all the words in the phrase will be described. The operation in FIG. 27 is performed by the related information developing means 2 in FIGS.

S40. The related information developing unit 2 determines whether or not the user checks the total number of occurrences of the keyword based on input data such as a search condition and an extraction unit input to the search condition input unit 1. When it is determined not to check, the process proceeds to S43.

S41. When it is determined that the total number of appearances is to be obtained, the related information developing means 2 searches the target database 6 with the keyword and counts the number of appearances for each keyword. S42. The related information developing means 2 displays the total number of appearances for each keyword thus obtained on the display means 4.

S43. Next, the related information developing means 2 determines whether or not the user checks whether or not the keyword has appeared. When it is determined not to check, the process proceeds to S47. S44. When checking for the presence or absence of a keyword, the user specifies a unit of measurement of the frequency distribution. A unit of measure is a paragraph / section / chapter / document / file,
Judge this. This measurement unit can be input as input data such as the extraction unit.

S45. The related information developing means 2 searches the target database 6 with the keyword, and counts the number of appearances of the keyword for each measurement unit. S46. The related information developing means 2 sends the presence or absence of the keyword for each measurement unit to the display means 4 and displays it.

S47. The related information developing means 2 determines whether or not the user checks the appearance frequency distribution of the keyword. The process ends when it is determined not to check. S48. When checking the appearance frequency distribution of a keyword, the user specifies a unit of measurement of the frequency distribution. The unit of measurement is a paragraph / section / chapter / document / file.

S49. The related information developing means 2 searches the target database 6 with a keyword, and counts the number of appearances of the keyword for each measurement unit. S50. The related information developing unit 2 sends this to the display unit 4 and displays the appearance frequency distribution state of the keyword for each measurement unit on the display unit 4.

In this manner, the total number of occurrences of all keywords in the target database, the presence / absence of occurrence of keywords,
It is possible to obtain the keyword appearance frequency distribution, that is, the top n frequently appearing phrases and the like.

As described above, in the present invention, before performing a data search based on a search condition input by a user, related information is derived from the search condition by using a related information expanding means, and By providing the relevant information to the user in the form of, for example, a table, the user can be given more information than can be obtained by a simple data search. Appropriate keyword selection and database operation are possible.

[0136]

According to the present invention, the following effects can be obtained. (1) Before performing a data search, by obtaining relevant information from a search condition or from a database and displaying the relevant information to the user, the user can know the partial spelling without knowing the entire spelling of the keyword. As a result, a data search can be performed using the information, so that the data search can be supported. Further, since the related information is displayed, the keyword can be presented for the next search, and the function of the search support is achieved. It will be. In this way, the user can be provided with more information than can be obtained by a simple data search, and in the search again, more appropriate keyword selection and database operation can be performed.

(2) Display related words in the form of a tag-attached file on the vertical or horizontal axis, and present only detailed information of the related information of the designated tag to the user as the forefront position, and display the hidden details. Since the information is calculated as a background job, it is possible to display all the detailed information at the forefront position that is easy for the user to understand, and even long detailed information. Moreover, since the hidden detailed information is calculated as a background job, even if the combination of related terms is changed, the detailed information can be displayed at the forefront in a short time.

Further, a set of related terms is extracted, and these related terms are displayed in a two-dimensional table format in which a related word on the vertical axis and a related word on the horizontal axis are displayed. Since the result is displayed in the cell at the intersection, detailed information can be displayed by simple display means.

In addition, a set of related terms is extracted and these related terms are displayed in the form of a tagged file on the vertical axis or horizontal axis, and the related words specified on the vertical axis and the relevant words specified on the horizontal axis are displayed. Is displayed on the screen in the front row, and the hidden detailed information is calculated as a background job, so that even complex items can be displayed in a clear and easy-to-understand manner for the user, and the combination of related words and phrases can be displayed. Even if it is changed, the detailed information can be displayed at the forefront position in a short time.

(3) An array of related words in which the related terms are newly arranged is displayed outside the two-dimensional table, and the user sets the related words in the related word array outside this table to (N−2). ), When a total of N related words of the designated (N-2) related words, related words on the vertical axis and related words on the horizontal axis are added to each cell of the two-dimensional table. Is displayed in the cell at the intersection of the vertical and horizontal axes, so even if the number of related terms is too large to be displayed in a row, N complex ANs are displayed.
D Searched items can be displayed.

Further, outside the two-dimensional table, N-2 one-dimensional arrays in which related terms different from those on the vertical and horizontal axes are newly displayed are displayed, and the user can display the related words in the one-dimensional array by ( N-2)
When specified, N cells of (N-2) related words of the specified related word array, related words on the vertical axis, and related words on the horizontal axis are added to each cell of the two-dimensional table. Since the result of the AND search for the word is displayed in the cell at the intersection of the vertical and horizontal axes,
Even if the number of related words is very large and a single related word string cannot be displayed even if it is provided outside the two-dimensional table, it is displayed, and even if there are a large number of related words, the search state can be changed. Can be displayed. Further, by providing the N-2 primary arrays for each area, for example, for each area, for each manufacturer, and for general and special types, it is possible to display a search state from a complicated viewpoint.

(4) A plurality of related words are derived from the keyword, each keyword of the related word is presented to the user in a table format, the number of document data including the keyword is displayed, and each keyword including the keyword is displayed. Since the appearance of these related words is checked in the entire document, and in which document the words appear, the detailed information can be displayed together with the number of cases.

Also, a plurality of related words are derived from the keywords, each keyword of the related words is presented to the user in a table format, and the appearance status of these related words is checked in a plurality of document databases, and the phrases are determined. Since which document database appears in a tabular format, the user can check whether the entered keyword is valid for the search and ask the user which document database the user should focus on and further investigate. Can guide.

Then, a plurality of related words are derived from the keyword, each keyword of the related word is presented to the user in a table format, the number of document data including the keyword is displayed, and the entire document including the keyword is displayed. In order to check the appearance of these related words and display the number of occurrences of those words in each document in a tabular format,
It is possible to confirm whether the input keyword is effective for the search, and to guide the user on which document the user should further investigate.

In addition, a plurality of related words are derived from the keywords, each keyword of the related words is provided to the user in a table format, and the appearance status of these related words is checked in each document database containing the keywords, and the related words are obtained. The tabular display of the document database in which the word or phrase appears allows the user to check whether the entered keyword is valid for the search, and to determine which documents the user should focus on and further investigate. Can guide you.

[Brief description of the drawings]

FIG. 1 is an embodiment of the present invention.

FIG. 2 is a diagram illustrating the operation of the present invention.

FIG. 3 is a development explanatory diagram of related information using a thesaurus in the present invention.

FIG. 4 is a development explanatory diagram of related information using a target database according to the present invention.

FIG. 5 is a second embodiment of the present invention.

FIG. 6 is a display format (two-dimensional display format) of related information of the present invention.

FIG. 7 is a display format (tag display format) of related information according to the present invention.
It is.

FIG. 8 is a display format of related information (tag display format of AND search of N pieces of related information) according to the present invention.

FIG. 9 is a display format (an AND display format of two related phrases) of related information according to the present invention.

FIG. 10 is a display format (correlation display between two related phrases) of related information of the present invention.

FIG. 11 is a display format of related information of the present invention (an AND display format of a tag format of two related phrases).

FIG. 12 is a display format (an AND display format of N related phrases) of related information of the present invention.

FIG. 13 is a display format of related information (AND display format with N related axes) according to the present invention.

FIG. 14 is a display format of the related information of the present invention (AND display with N related axes).

FIG. 15 is a display format (preliminary search display of a time-series database) of related information of the present invention.

FIG. 16 is a display format (display of a preliminary search of a situation database) of the related information of the present invention.

FIG. 17 is a display format (appearance status for each data) of related information of the present invention.

FIG. 18 is a display format (appearance status for each distributed DB) of related information according to the present invention.

FIG. 19 is a display format (number of appearances for each data) of related information according to the present invention.

FIG. 20 is a display format (number of appearances for each database) of related information of the present invention.

FIG. 21 is an example of a procedure of a related information expanding unit according to the present invention.

FIG. 22 is an example of related information display (frequency of appearance of synonyms) according to the present invention.

FIG. 23 is a display example (appearance frequency tag display) of a synonym of the related information of the present invention.

FIG. 24 is an example of a procedure of a related information expanding means of the present invention.

FIG. 25 is an example of related information display (frequency of appearance of a high-frequency phrase) according to the present invention.

FIG. 26 is a display example (tag display) of a high-frequency phrase as related information of the present invention.

FIG. 27 is an explanatory diagram of a related information extraction procedure that can be extracted by the target database of the present invention.

FIG. 28 is an explanatory view of a conventional example.

[Explanation of symbols]

 1 search condition input means 2 related information development means 3 related information dictionary 4 display means 5 database search means 6 target database 7 related information development determination means 8 development format storage means

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Kiyoshi Okura 4-1-1, Kamidadanaka, Nakahara-ku, Kawasaki-shi, Kanagawa Prefecture Inside Fujitsu Limited (72) Kuni Matsui 4-1-1 Kamiodanaka, Nakahara-ku, Kawasaki-shi, Kanagawa Prefecture No. 1 Inside Fujitsu Limited

Claims (4)

[Claims] 1. An information search apparatus comprising: a search condition input unit for inputting a search condition including a keyword by a user; and a database storing documents to be searched, wherein the database is searched based on the search condition. A related information dictionary in which related information related to the keyword is stored; and searching the related information dictionary based on the keyword,
A related information developing means for extracting and displaying related information corresponding to the input keyword; and a related information display means for displaying the related information, wherein prior to searching a database based on the search condition, An information search device, wherein the related information dictionary is searched by a keyword or a database is searched to extract related information, and the related information is displayed. 2. The related information developing means extracts a set of related terms such as synonyms and synonyms as related information, and further extracts detailed information of the related terms from the related information dictionary or preliminarily searches a target database. The related information display means displays these related words in a graphic format of a tagged file on the vertical axis or the horizontal axis, and presents the detailed information of the related information of the designated tag to the user as the foremost position. Or that the hidden detailed information is calculated as a background job, or the related information developing means performs a two-dimensional table in which those related terms are arranged in a related word on the vertical axis and a related word on the horizontal axis. The format is displayed, and the result of an AND search between the related word on the vertical axis and the related word on the horizontal axis is displayed in a cell at the intersection of the vertical axis and the horizontal axis, or the related information display means further includes: Their related words Is displayed in a graphic format of a tagged file on the vertical axis or horizontal axis, and the result of an AND search between the related word specified on the vertical axis and the related word specified on the horizontal axis is displayed on the front row screen 2. The information search apparatus according to claim 1, wherein the hidden detailed information is calculated as a background job. 3. The related information display means displays a two-dimensional table format in which the related words and phrases are arranged on a related word on a vertical axis and a related word on a horizontal axis, and further outside the two-dimensional table. A related word array in which the related words are newly arranged is displayed, and the user specifies (N-2) (where N is an integer of 3 or more) related words in the related word array outside the above table. Then, in each cell of the two-dimensional table, a total of N related words of (N-2) related words of the designated related word array, related words on the vertical axis and related words on the horizontal axis are provided. The result of the AND search of the word is displayed in a cell at the intersection of the vertical axis and the horizontal axis, or a one-dimensional array in which related terms different from the vertical axis and the horizontal axis are newly arranged (N-2 ) (Where N is an integer of 3 or more)
The book is displayed, and the user selects the related word in the one-dimensional array (N
-2) When the specified words are assigned to each cell of the two-dimensional table, (N-2) related words in the specified related word array, related words on the vertical axis and related words on the horizontal axis AND of a total of N related words
The information search device according to claim 1, wherein a search result is displayed in a cell at an intersection of a vertical axis and a horizontal axis. 4. The related information developing means derives a plurality of related words from a keyword, presents each keyword of the related words to a user in a table format, and displays the number of document data including the keyword. , Check the appearance of these related words in each document including the keyword, and display in a tabular form which documents appear in those words, or together with this presentation, in each of a plurality of document databases,
Investigate the appearance of these related words and display in which document database those words appear in a tabular format, or display the number of document data items containing the keyword and the entire document containing the keyword Then, the appearance of these related words was examined, and the number of appearances of each of the phrases in each document was displayed in a tabular form. 2. The information retrieval apparatus according to claim 1, wherein the information retrieval device displays the document database in which the words and phrases appear in a table format.