JP3193249B2 - Keyword search method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to extraction from a document file.
Stored keywords, and based on the keywords,
How to search for files.

[0002] Documents on electronic media can be used as a database by storing them. As a usage form of such a data asset, a document keyword described in a document may be displayed, or may be updated as necessary.

In an electronic document, a document keyword corresponds to an instance of a database and a document corresponds to a record of a database. Therefore, unless otherwise specified in the following description, an instance of a database includes a document keyword.

[0004]

2. Description of the Related Art Searching for a corresponding document from a keyword
Open each file for each document file
Search for the desired keyword, load it, and load the file.
The operation of closing was repeatedly performed. I
However, such a method requires a lot of search processing time.
Normally, the keyword of each document file is
-When managing and searching in a word management file
Will read only this keyword management file
Operation to open the file
One time only, and keywords for multiple document files
Search is performed continuously.

[0005]

SUMMARY OF THE INVENTION Keyword management file
Le stores keywords in the form of a table. That
The record length defined at this time is extracted from the document file.
Is usually expected because the number of characters in the keyword
The maximum length is set in advance.

Therefore, a character string exceeding the record length
Cannot be included in keyword searches,
And cannot be stored in the keyword management file. On the other hand,
Keywords extracted if code length setting is too long
Is almost the same as the number of characters in the document file
In case, the file size of the document file is effectively
Searching for similar keyword management files
As a result, the search process cannot be shortened.

[0007] The present invention has been made in view of such a point, and regardless of the record length for storing a keyword.
Any string can be stored in the keyword management file.
This keyword management file allows
It is an object of the present invention to provide a keyword search method capable of searching for a file .

[0008]

[MEANS FOR SOLVING THE PROBLEMS] To achieve the above object
In the present invention, a keyword is extracted from a document file,
The number of characters of extracted keywords is set in advance
Determines whether the number of stored characters exceeds the upper limit and extracts
The number of characters of the entered keyword exceeds the upper limit of the number of stored characters.
If not, replace the keyword string with the stored statement
If the number of characters exceeds the upper limit, before the keyword exists
The location information in the document file can be
In the file.

Further , the present invention provides the keyword management file
The record in the file keyword storage table
To determine whether the read record has location information.
If there is no location information, the search target character string and the
Keywords read from the keyword management file and
And if there is the position information, the finger is
Open the document file in which the indicated character string is described.
And extract the character string indicated by the location information from there.
It consists of steps to match with the search target string
It is characterized by.

[0010]

According to the above-mentioned means, a key for storing a keyword is stored.
Regardless of the code length, any character string can be
This keyword management file can be stored in
Files can be searched by
You.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS First, an embodiment of the present invention will be outlined. Figure
1 is a block diagram illustrating a configuration of an instance updating device. The instance update device is the instance input device 2
0 and an external storage device 30 and a display device 40 connected thereto. The instance input device 20 stores the document file 3 stored in the external storage device 30.
An instance reading device 21 for reading an instance from the first instance, a memory device 22 on which the read instance is expanded, and the instances expanded on the memory device 22 are displayed on the display device 40 in a horizontal or vertical list display format. List creating device 23 for creating a simple list data, event analyzing device 24 for analyzing an event such as an instance update instruction, and memory device 22
And an instance storage device 25 for writing back the updated instance in the external storage device 30.

To display a list of instances, first, the instance reading device 21 of the instance input device 20 opens the document file 31 stored in the external storage device 30, searches for and extracts a target document keyword. , To the memory device 22. List creation device 2
3 creates a list by arranging instances developed in the memory device 22 according to a predetermined list display format, for example, a horizontal list display format as shown in FIG. 2, and sends the list to the display device 40. The display device 40 receives the list data from the list creation device 23,
Display the list.

When an instance is changed, the event analyzer 24 analyzes an instruction for selecting and updating an instance, and notifies the list creating device 23 of the analysis. The event analysis device 24 also rewrites the corresponding instance in the memory device 22 according to the instruction to confirm the input after the change of the instance. This rewritten instance is immediately reflected in the creation of the list in the list creation device 23, and the updated list is displayed by the display device 40. Also, the rewritten memory device 22
Are written back to the document file 31 in the external storage device 30 by the instance storage device 25. For this reason, the instance storage device 25 specifies the document file 31 stored in the external storage device 30 from the storage location of the rewritten instance, searches for a corresponding instance of the document file 31, and rewrites.

In a preferred embodiment, the keywords in each document file 31 to be read by the instance reading device 21 are extracted and stored in the external storage device 30.
Management file 32 stored in one file
When the instance reading device 21 reads the keyword of each document file 31,
This keyword management file 32 is read.

Normally, when the instance reading device 21 goes to read the keyword of each document file 31, the operation of opening each file for each document file, searching and reading the target keyword, and closing the file is performed. Running repeatedly. In the present embodiment, since the keywords of each document file 31 are collectively managed in the keyword management file 32, the instance reading device 21 reads the keywords of each document file 31 when reading the keywords of each document file 31. I try to read only 32. Therefore, there is an advantage that the operation of opening the file when reading the file only needs to be performed once, and that the keywords of the plurality of document files can be read continuously, and that the list of instances can be displayed quickly. Can be.

As described above, it is common practice to store a character string extracted from a document file as a keyword in a search management file in a table format. Therefore, the external storage device 30 stores the document file 3
1 is connected to a keyword management device for extracting keywords from the keyword No. 1 and generating a keyword management file 32.

FIG . 2 is a diagram showing a configuration example of the keyword management device. The keyword management device includes a keyword extraction device 33 that extracts keywords from the document file 31 of the external storage device 30 and a keyword storage device 34 that stores the extracted keywords in the keyword management file 32 of the external storage device 30. You.

The keyword extracting device 33 first extracts predetermined keywords from all the document files 31, and the extracted keywords are stored in the keyword management file 32 by the keyword storage device 34. Here, every time one of the document files 31 is updated, a keyword is extracted from the updated document file 31 and the keyword management file 32 is also updated. Instance reading device 21
When the next instance is read, the instance is read from the updated keyword management file 32.

The keyword management file 32 stores keywords in a table format. The record length of the keyword storage table defined in the keyword management file 32 is usually the maximum expected length because the number of characters of the keyword extracted from the document file 31 by the keyword extraction device 33 is arbitrary. Is set in advance.

Therefore, the keyword management file 32
The character string exceeding the record length of the keyword storage table used in the above cannot be searched for the keyword, and the keyword of the character string exceeding the record length cannot be stored. However, if the record length is set to be too long, and if the extracted keywords are almost the same as the number of characters having the set length, a large file having substantially the same contents will be stored in the keyword management file 32. And another one will be created.

Therefore, the keyword management device is configured to be able to store even a character string exceeding the record length of the keyword storage table.
That is, when the keywords extracted from the document file 31 are stored in the keyword management file 32, an upper limit of the number of characters to be stored is provided, and for the keywords less than the upper limit, the character strings are stored in the keyword management file 32 and stored. For a keyword exceeding the upper limit of the number of characters, information indicating an appearance position in a document file in which the keyword is stored is stored in the keyword management file 32. The upper limit of the number of stored characters is usually equal to the record length of the keyword storage table.

FIG . 3 is a flowchart showing the operation of the keyword extracting device. According to this flowchart, the keyword extracting device 33 first extracts a keyword from the document file 31 stored in the external storage device 30 (Step S11). Next, it is determined whether or not the extracted keyword exceeds a preset upper limit of the number of stored characters (step S12).

If it is determined in step S12 that the extracted keyword does not exceed the upper limit of the number of stored characters, the character string of the keyword is stored in the keyword storage device 3.
4 is stored in the keyword management file 32 of the external storage device 30 (step S13). Step S1
If the extracted keyword exceeds the upper limit of the number of stored characters in the determination of step 2, the position information of the keyword in the document file is stored in the keyword management file 32 of the external storage device 30 by the keyword storage device 34. (Step S14).

FIG . 4 is a diagram exemplifying the document contents of the document file. According to the illustrated example, there are three document files 31a, 31b, and 31c whose document names are "document 1", "document 2", and "document 3". In these documents, character strings extracted as keywords are underlined to distinguish them from other character strings. For example, character strings set as keywords in “document 1” are “marking”, “standard processing format”, and “automatic generation”. The keywords in these documents are extracted and stored in the keyword management file 32 by the keyword management device.

FIG . 5 is a diagram exemplifying the contents of the keyword management file. The keyword management file 32 is a correspondence table of two items of “document name” and “keyword”. The upper limit of the number of stored characters in the “keyword” field is, for example, 10 characters (20
Byte). Therefore, a keyword having the number of stored characters up to this upper limit is stored as it is in the “keyword” field.

Here, when the number of characters in the extracted character string exceeds the upper limit of the number of stored characters, as in the case of the second keyword of “document 3” shown in FIG. Instead of storing the column in the keyword management file 32 as it is, information on the position where the character string appears in the document is stored in the keyword management file 32. For example, in the example shown in the figure, the second keyword of “document 3” is counted from the beginning of the document to the 121st byte and the 1st byte.
It indicates that it exists between the 54th byte.

When the instance reading device 21 reads an instance from the external storage device 30 not from the document file 31 but from the keyword management file 32 in which keywords and location information are stored in a mixed manner, It is considered whether the information stored as is a keyword or position information.

That is, the keyword management file 32
If the information read from is the keyword location information,
First, a document file 3 in which the keyword is stored
1c is opened, and then a character string in the document file 31c pointed to by the position information of the keyword is extracted, and the character string is used as an instance for list display. When the character string has been extracted, the document file 31c is closed at an appropriate time.

The keyword management file 32 in the external storage device 30 can be used not only for searching for a keyword for reading an instance, but also for searching for a mere keyword. If a keyword search is to be performed without using the keyword management file 32, when performing a full-text search, all document files are opened, and a search target character string is compared with all character strings in the document. There must be. On the other hand, the keyword management file 32 in which the keyword and the position information are stored mixedly
In the case where is used, in addition to opening the keyword management file 32, it is only necessary to open a document file relating to the location information if it has been registered.

FIG . 6 is a flowchart showing a keyword search method. When trying to search for the corresponding keyword in the document file by specifying the search target string,
First, the keyword management file 32 is opened (step S21). Next, one record in the keyword storage table is read (step S22). Here, it is checked whether or not the end of the file (EOF) has been reached (step S23). If it is determined that the end of the file has been reached, the search is completed.

If it is not the end of the file, it is checked whether or not the read record has position information (step S24). If the position information does not exist, the search target character string specified in advance is compared with the keyword read from the keyword management file 32 (step S2).
5), proceed to search for the next record. If it is determined in step S24 that there is position information, a document file in which the character string specified by the position information is described is opened, and the character string indicated by the position information is extracted therefrom. Is performed (step S26). And
The result of this collation is displayed by the display device.

FIG . 7 is a diagram showing an example of a hardware configuration of a computer system embodying the present invention. In the figure, a computer system includes a processor 51, a read-only memory (ROM) 52, and a main memory (R).
AM) 53, a graphic control circuit 54 and a display device 55, a mouse 56, a keyboard 57, a hard disk device (HDD) 58, a floppy disk device (FDD) 59, and a printer 60.
These components are interconnected by a bus 61.

The processor 51 controls the entire computer system. The read-only memory 52 stores, for example, programs required at the time of startup. The main memory 53 stores data such as read instances and the like in addition to the system programs and the application programs for list display.

The graphic control circuit 54 has a video memory and the like, converts a list display screen of the instance group read into the main memory 53 into a display signal, and sends the display signal to the display device 55. The display device 55 displays a horizontal or vertical list display screen based on the received display signal.

The mouse 56 moves the cursor displayed on the screen of the display device 55, and selects an instance to be updated on the list display screen by clicking a button. This is a pointing device for selecting and instructing a command. The keyboard 57 is used for inputting characters and the like when rewriting an instance.

The hard disk device 58 stores a system program, an application program for displaying an instance list, a document file, a keyword management file, and the like. The floppy disk device 59 is an external storage device that can take a document file stored in the floppy disk 59a into the hard disk device 58 and store data of a list display result in the floppy disk 59a.

The list of instances can be printed on paper by passing the display data to the printer 60.

[0038]

As a method of storing keywords extracted from a plurality of document files in a keyword management file, an upper limit of the number of stored characters is set, a keyword less than the upper limit is a character string of the keyword, and a keyword exceeding the upper limit is a keyword. The appearance position of the keyword in the document is stored in the keyword management file. For this reason, the size of the keyword management file can be optimized, and by managing keywords collectively in one file, a character string of any number of characters extracted from a document file can be managed as a keyword while maintaining search efficiency to some extent. can do.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an instance update device.
It is.

FIG. 2 is a diagram illustrating a configuration example of a keyword management device.

FIG. 3 is a flowchart showing the operation of the keyword extraction device .
It is.

FIG. 4 is a diagram exemplifying document contents of a document file;

FIG. 5 is a diagram exemplifying the contents of a keyword management file.
It is.

FIG. 6 is a flowchart showing a keyword search method .
You.

FIG. 7 is a computer system embodying the present invention;
FIG. 3 is a diagram illustrating an example of a hardware configuration.

[Explanation of symbols]

 Reference Signs List 20 instance input device 21 instance reading device 22 memory device 23 list creation device 24 event analysis device 25 instance storage device 30 external storage device 31 document file 32 keyword management file 33 keyword extraction device 34 keyword storage device

Continued on the front page (72) Inventor Chikako Hyakuyama 2-2-1 Honai, Yao-cho, Neguro-gun, Toyama Fujiyama Co., Ltd. (72) Inventor Takahiro Hayashi 2-2-1 Honai, Yao-cho, Meguri-gun, Toyama Co., Ltd. JP-A-60-24657 (JP, A) JP-A-5-181907 (JP, A) JP-A-5-81101 (JP, A) JP-A-62-189552 (JP, A A) JP-A-5-128159 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G06F 17/30 210

Claims (2)

(57) [Claims] A keyword extracted from a document file;
Keys stored in the keyword management file and stored
Keyword search to search document files based on words
In the method, a keyword is extracted from a document file , and the number of characters of the extracted keyword is set in advance.
Judge whether the number of stored characters exceeds the upper limit of the number of stored characters.
If not, the character string of the keyword is
If the number of characters exceeds the upper limit, the keyword exists.
The location information in the document file
When storing in a management file and performing search processing, the stored keyword management file is used.
A keyword search method characterized by comprising a step of searching for a document file based on a file
Law. 2. A record of the keyword management file.
To determine if the read record has location information
And, if there is the position information, the searched character string keywords
Check with keyword read from file management file
If there is the position information, it is indicated by the position information.
Open the document file that describes the character string
From there, extract the character string indicated by the location information and
2. The key according to claim 1, further comprising a step of performing matching with an elephant character string.
Word search method.