JPH08249341A - Document storage and retrieval device for document data base - Google Patents

Abstract

PURPOSE: To provide a retrieval device which registers and retrieves data efficiently and is high in hit rate by extracting a key word from document data and registering a representative word which the key word belongs to, together with the document data when the document data are registered in the document data base. CONSTITUTION: This device consists of a document data file accepting means 11 which accepts the document data, a key word file 22 in which key words and representative words of the key words are registered, a key word retrieval means 12 which retrieves the document data by using the key word file 22, a data storage means 13 which stores the document data and representative words, and a document retrieval means 15 which replacing a retrieval key word inputted by a user with the representative word by referring to the key word file to perform retrieval from the document data base, and displays the result on a document display means 23.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document storage / retrieval device in a document database, and more particularly to a document storage / retrieval device for adding a retrieval keyword to document data.

[0002]

2. Description of the Related Art Conventionally, an apparatus for registering an input sentence in a database and searching the database for the sentence using a keyword is a method in which a user combines a sentence registered in the database with a logical expression or the like as a keyword. A device for performing a search is known. In this device, the expression method of the sentence differs depending on the person who wrote the sentence.Therefore, when searching using a keyword having a plurality of synonyms, the keyword having the same meaning must be searched for each time using a search formula or the like. However, it causes search omissions, and the user's burden is heavy because he or she considers keywords with the same meaning.
In addition, there is a problem that it takes time to search for a keyword from a sentence when searching.

On the other hand, as a conventional device for reducing omission of search, there is known a database registration / search device disclosed in Japanese Patent Application Laid-Open No. 5-250416. The configuration is shown in FIG. 8 and the operation of the apparatus will be described.

The text input from the input / output device 30 is input to the representative word management means 33 via the input control means 31. A representative word corresponding to each keyword included in the input sentence is registered in the representative word dictionary 37 in advance, and the representative word managing unit 33 represents the keywords included in the input sentence as a representative. Word replacement data registration means 38
Register in the database 40 via. Further, when searching the registered text, the search keyword input from the input / output device 30 is replaced with the representative word in the same manner as described above, the database is searched, and the registered text is searched. obtain.

[0005]

SUMMARY OF THE INVENTION In the above-mentioned device,
The keyword included in the sentence is replaced with the representative word and registered in the database, and the keyword is replaced with the representative word when searching, which reduces the omission of the search. Since the database is searched using this, there is a problem that the search is slow.

Further, since the keywords included in the text are replaced with the representative words and registered in the database, there is a problem that the original text cannot always be obtained.

The object of the present invention is to solve the above-mentioned problems of the apparatus, reduce omission of search, obtain original sentences, further improve search speed, and reduce the burden on the user. It is to provide a document storage / retrieval device in a document database that reduces the number of documents.

[0008]

A first invention comprises a document file receiving means for receiving document data as input data, a table of keywords used when searching the document data, and a representative word corresponding to the keywords. A keyword file, keyword search means for searching the keyword from the text data using the keyword file, and extracting a representative word; and adding a new record to the document data, and extracting the record by the keyword search means. It is characterized by comprising a data storage means for recording the representative word and storing it in the document database.

In a second aspect of the present invention, in the first aspect of the present invention, there is provided document search means for receiving a search keyword from a user, referring to the keyword file, replacing the search keyword with a representative word, and searching the record. Characterize.

A third invention is characterized in that, in the first or second invention, the keyword file uses one of keywords having synonyms as a representative word.

[0011]

Next, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of the configuration of the present invention. In FIG. 1, a document file acceptance unit 11 that accepts document data as input data, a keyword file 22 including a table in which keywords used when retrieving document data and representative words of the keywords are associated, and keywords for the document data. File 22
If there is a keyword held in the keyword file 22, it is searched if there is a keyword held in
A keyword search unit 12 that searches for a representative word of the keyword using the keyword file 22 and document data are input, and a representative word of the keyword included in the document data and the document data are newly set as a record, and the document The data storage means 13 held in the database 21, the keyword input means 14 for accepting a keyword (hereinafter referred to as a keyword) to be searched by the user, the keyword file 22 for the representative word of the keyword, and the document using the representative word It is composed of a document search means 15 for searching records in the database 21 and displaying the search results on the document display means 23, and a document display means 23 for displaying the search results to the user.

Next, the operation of the present invention will be described with reference to FIGS. 1, 2, 3, and 4.

When storing a document, it operates as follows.

The document data given to the document file receiving means 11 is supplied to the keyword searching means 12. The keyword search means 12 performs a full-text search on this document data to check whether the keyword stored in the keyword file 22 is included in the document data.

An example of the structure of the keyword file 22 is shown in FIG. The keyword file 22 is a table in which keywords used when searching document data and representative words for the keywords are held in advance. For example, the keywords “extraction” and “detection” are synonyms, but these keywords are defined as “extraction” as a representative word in advance. The keyword file 22 is not limited to this example, and words having different expressions depending on the person who wrote the document (for example, data and data) are predetermined as representative keywords.

When the keyword search means 12 searches the document data for a keyword stored in the keyword file 22, a representative word corresponding to the keyword is extracted by referring to the keyword file 22 (FIG. 2).
Output to the data storage unit 13. Also, when a plurality of keywords are detected from the document data by the keyword search means 12, the representative word corresponding to the keywords is similarly extracted and output to the data storage means 13.

In the data storage means 13, the document data accepted by the document file acceptance means 11 is input, and the representative word extracted by the keyword search means 12 is also input at the same time. Here, the data storage means 13 stores in the document database 21 in a format in which a representative word group is added to the document data as in the example shown in FIG. 3 or 4.

The document database 21 will be described below. In the document database 21, as shown in FIG. 3, a record of a representative word group included in the document data is newly added to the document data having the document number, and a keyword included in the document data is included. Of course, the number of representative words of V is different for each document data.

FIG. 4 shows another example of the document database. In this example, what kind of document data exists for a certain representative word is represented by a tree structure. Of course, in the case of document data having a plurality of representative words, there are a plurality of branches.

By the operation described above, the document data accepted by the document file accepting means 11 is stored in the document database 21.

When searching for a document, it operates as follows.

The keyword input by the keyword input means 14 is supplied to the document search means 15. The document search unit 15 refers to the keyword file 22 and extracts a representative word corresponding to the input keyword. Next, the document database 21 is searched using the representative word, predetermined document data is obtained, and the result is displayed on the document display device 23. Here, as the data displayed on the document display device 23, only the document data desired by the user may be displayed, but the related representative word of the document may also be displayed. Further, a part or all of the tree structure shown in FIG. 4 may be displayed to the user.

By the operation described above, the document database is searched from the keyword input by the keyword input means 14, and the search result is displayed on the document display means 23.

Next, an example of a series of operations of the present invention when the document file receiving means 11 receives the document data shown in FIG. 5 will be described. The keyword file 22 is assumed to have the keyword file shown in FIG. 6 in advance.

The document data shown in FIG. 5 is input by the keyword search means 12, and the keyword file 22 is referred to to extract the representative word of the keyword contained in the document data.

Now, in the document data shown in FIG.
The keywords, “text”, “database”, and “search” defined in 1. exist. Keyword search means 1
In step 1, these keywords are detected, the respective representative words “text”, “database”, and “search” are obtained, and the representative words are output to the data storage means 13.

In the data storage unit 13, the document data is input from the document file receiving unit 11, the representative word is input from the keyword searching unit 12, and the document database 2 is input.
Store in 1. As shown in FIG. 3, the storage may be performed by adding a record of a representative word to the document data (see FIG. 7), or as shown in FIG. It may be a tree structure indicating whether it exists.

Here, when the keyword "database" is input through the keyword input means 14,
The document search means 15 refers to the keyword file 22 and obtains a representative word "database" corresponding to the keyword "database" (FIG. 6). Then, using the representative word, a record of the representative word in the document database 21 is searched, desired document data is obtained, and displayed on the document display means 23.

Further, in the search, one keyword is not always used for the search, and a plurality of keywords may be searched for in combination with a logical expression or the like.

[0031]

As described above, the document storage / retrieval apparatus for a document database according to the present invention adds a record of a representative word to the document database and sets the record as a search target in the conventional search. This has the effect that the search is faster than when all the texts of the technology are searched.

Since the record holds the representative words, a wide range of searches can be performed even when synonyms whose expressions are different by the user are input as keywords.

Further, as compared with Japanese Patent Laid-Open No. 5-250416, since the representative word is held as an independent record, the original sentence can be obtained.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the structure of the present invention.

FIG. 2 is a diagram showing an example of a table structure of a keyword file of the present invention.

FIG. 3 is a diagram showing an example of a record structure of a document database of the present invention.

FIG. 4 is a diagram showing another example of the structure of records in the document database of the present invention.

FIG. 5 is a diagram showing an example of document data.

FIG. 6 is a diagram showing an example of the structure of a keyword file table.

FIG. 7 is a diagram showing an example of the structure of a document database.

FIG. 8 is a block diagram showing the structure of a conventional technique.

[Explanation of symbols]

 11 Document File Reception Means 12 Keyword Search Means 13 Data Storage Means 14 Keyword Input Means 15 Document Search Means 21 Document Database 22 Keyword Files 23 Document Display Means 30 Input / Output Devices 31 Input Control Means 32 Keyword Conversion Means 33 Representative Word Management Means 34 Representatives Word registration means 35 Representative word determination means 36 Output control means 37 Representative word dictionary 38 Data registration means 39 Data search means 40 Database

Claims (3)

[Claims] 1. A document file accepting means for accepting document data as input data, a keyword file comprising a table of keywords used when searching the document data and a representative word corresponding to the keyword, and using the keyword file. A keyword search means for searching the text data for the keyword and extracting a representative word, and adding a new record to the document data, recording the representative word extracted by the keyword search means in the record, and storing it in a document database. A document storage / retrieval device for a document database, comprising: a data storage unit for holding the document database. 2. The document according to claim 1, further comprising a document search unit that receives a search keyword from a user, refers to the keyword file, replaces the search keyword with a representative word, and searches the record. Database document storage and retrieval device. 3. The document storage / retrieval device for a document database according to claim 1, wherein the keyword file uses one of keywords having synonyms as a representative word.