JPH0581320A - Information retrieving device - Google Patents

Abstract

PURPOSE:To execute a retrieval extending over between media generated independently in an electronic filing system, etc. CONSTITUTION:By integrating keyword data bases A, B generated at every media (a), (b), an integrated keyword data base C is generated. In this integrated keyword data base C, the integrated keyword and an ID conversion table of a local keyword of every media are held. A retrieval extending over plural media is executed by using a converting mechanism of the keyword ID. Accordingly, by the integration of the keyword and its ID converting mechanism, a retrieval of a wide range extending over between the media generated independently can be executed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information retrieving device such as an electronic filing system for retrieving stored input information by retrieval information created for each medium.

[0002]

2. Description of the Related Art In recent years, electronic filing systems are becoming smaller and lighter, and a system called section (divisional unit) type that allows easy registration and retrieval of documents is becoming widespread. This is different from the conventional large-scale system, in which the keyword is stored as ID information in each document together with the image data instead of the character string itself, and the keyword database may be created and recorded for each information storage medium. This method is called a preset keyword method or a fixed keyword method.

[0003]

SUMMARY OF THE INVENTION In the above conventional configuration,
Although there is an advantage that you can freely create a keyword database for each media, but the accumulation of data progresses,
When trying to perform a keyword search across a plurality of media, there is a problem that this is impossible because the ID information is different even for the same keyword name.

The present invention solves the above conventional problems, and an object of the present invention is to provide an information retrieving apparatus capable of performing a keyword retrieval across media by integrating a keyword database created for each medium. It is a thing.

[0005]

In order to solve the above-mentioned problems, an information retrieval apparatus of the present invention is an information retrieval apparatus for retrieving stored input information by retrieval information created for each medium. Data storage means for integrating search information for each of the media and storing integrated search information created across the media, and data convertible from the integrated search information of the data storage means to the search information for each medium. It is provided with a management means.

[0006]

With the above configuration, the search information set for each information storage medium is integrated to create integrated search information across media, and the integrated storage information is stored in the data storage means, and the integrated search information of this data storage means is used. Data is converted into search information for each media by the data management means, and the stored input information is searched from the search information for each media. Therefore, when searching across multiple media, multiple search information created independently A wide range of searches can be performed with the media managed as one search information.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the system configuration of an electronic filing system according to an embodiment of the present invention. In FIG. 1, a data input unit 1 is composed of a keyboard, a pointing device, an image scanner, etc., and inputs various data from a user. The system control unit 2 to which the data input unit 1 is connected is composed of a CPU and a work RAM, and performs data input / output, control of each function, and data exchange. Further, the data storage unit 3 for the document data is specifically composed of two types of an external large-scale storage device such as an optical disk and a fixed hard disk drive. In this optical disk storage device, the storage medium is usually replaceable, and here, the media named a, b, and c are used. A document data management unit 4, to which the system control unit 2 and the data storage unit 3 are connected, manages and controls the data storage unit 3 to perform addition / update and reference of document data. Further, the document registration unit 5 connected to the system control unit 2 adds document attribute information to the document image data read by the image scanner or the like which constitutes the data input unit 1 and passes it to the data management unit 4. Further, the document search unit 6 connected to the system control unit 2 is based on the search condition input by the data input unit 1, for example, a fixed keyword as search information.
Documents are searched using the data management unit 4. Moreover,
The data display unit 7 to which the system control unit 2 is connected has a function of displaying an image of the retrieved document.

2A is a basic structural diagram of a keyword database as search information independently created in the medium a, B is a basic structural diagram of the keyword database in the medium b, and C is a medium a. It is a basic structural diagram of the integrated keyword database in, b. The keyword databases in the media a and b are those stored in the media and read by the document data management unit 4 in the form shown in FIGS. Hereinafter, the present invention applied to the document search unit 6 will be described in detail with reference to FIG. The keyword database created on the media a and b is in the form of a correspondence table of keyword IDs and keyword names. At the time of document registration, only the keyword name is displayed to the user in a table format, and only the ID of the selected keyword is held as a part of the document attributes of each document. Further, there are three types of media names, one having the same keyword name, one having a different name but the same meaning, and one having a different meaning.

The present invention provides a mechanism for integrating the keyword databases independently created as described above across media. This is the data management unit 4
The internal structure of this mechanism will be described with reference to the flowcharts of FIGS. 3 and 4. First, step S in FIG.
In step 1, "automatic extraction of keywords having the same name and table creation processing" are executed. This is to detect a keyword having the same name between media and to create a keyword using the name as an integrated keyword, and to record the local ID in each media in the integrated keyword database shown in C of FIG. Is stored in the fixed heart disk drive in the data storage unit 3 in the format of the example shown by (0). This local ID is written as a document attribute by the document registration unit 5 at the time of registration. In other words, the “minutes” of media a (the keyword whose ID is (0) is exactly the same in media b (ID
1) exists, and is automatically created with the same name in the integrated database, and a new ID (0) is assigned. Furthermore, the local ID of the media a
(0) and the local ID (1) in the medium b are recorded in the fixed hard disk in the form of a conversion table.

Here, using the flowchart of FIG.
The details of the “automatic extraction of the same name keyword and the table creation process” in step S1 will be described below. First, in step S11, a keyword database of all the media to be integrated is stored in a buffer formed in the RAM of the system control unit 2, the media name and the keyword ID.
And paired together to create one array. This array is an array of structures having three members of a keyword name, a media name, and a keyword ID, and the number of elements N is the total number of keywords in the keyword database of each medium. Next, in step S12, the elements of this array are sorted in the order of character code with respect to the keyword name.
As a result, the same keyword is positioned adjacent to each other in the array. After that, the keyword name matching test of the adjacent element of this array is continued to classify the keyword having the same name and the keyword existing independently.

That is, in step S13, the subscript i for array reference is initialized. And step S
In 14, first, two adjacent elements Ki and Ki are adjacent to each other.
The keyword names of +1 are compared to determine whether they are the same. If they are the same in step S14, the process proceeds to step S15, and the integrated keyword is registered in the integrated keyword buffer on the RAM. The format of this integrated keyword buffer is the same as that of C of FIG. 2, and is a list of newly added integrated keyword ID, integrated keyword name, and local keyword ID in each medium.

Further, in step S16, the array reference subscript i is incremented, and in step S17 it is determined whether "the number of elements N> array reference subscript i", that is, the end check at the time of the additional test is performed. If the array reference subscript i is larger than the number N of elements, the processing is terminated, and the number N of elements is the array reference subscript i.
If it becomes larger than that, the process proceeds to step S18. Step S18
Then, it is tested whether or not there is another keyword having the same name as the integrated keyword previously registered in step S15.
If the elements are the same, the process returns to step S16, and the same name keyword is added to the corresponding media column of the integrated keyword buffer. If the same keyword does not exist in step S18, the additional test for this integrated keyword ends, and in step S19, “i = i +
1 ", that is, the target element is advanced, and it is determined in step S20 whether the number of elements N> array reference subscript i. Number of elements N
> If the array reference subscript i, the process returns to step S14. If the number N of elements is not greater than the array reference subscript i, the keyword K _N is once registered as a single keyword in step S21. Step S21 is processing of the final data to be tested, which remains unprocessed.

On the other hand, if the adjacent keywords do not match in step S14, the keyword Ki is registered as existing independently in step S22, and is once subjected to interactive integration of the single keywords. Then, the process proceeds to step S19.

After the above-mentioned "automatic extraction of keywords having the same name and table creation processing" in step S1, step S2
The process moves to "Interactive integration of keywords having the same meaning and table creation process", and integration is performed by a user interactive process when keywords having different names but the same meanings exist in a plurality of media. This interactive processing is performed by the data display unit 7
Then, the element list of the single keyword buffer created in step S2 is displayed, and the user inputs the instruction of the synonymous keyword and the new keyword name using the pointing device or the keyboard which is the data input unit 1. is there. For example, in FIG. 2, "report" of A, "report" of B, "manual" of A, "manual" of B, "patent publication" of A and "patent" of B. These are synonymous keywords, and these integrated names may be selected by the user or devised newly. Here, as shown in FIG. 2C, the names are summarized as “report” in the first example, “manual” in the second example, and “patent” in the third example. Although not illustrated, the conversion table in the corresponding column of the medium for which the corresponding keyword does not exist is blank (-).

Further, after step S2, step S
3. "Independently existing keyword table creation process", that is, the table creation process is performed for keywords that independently exist only in a specific medium. In the example of C in FIG. 2, "takalog", "decision form" and "document" correspond to this. In this type of keyword, all conversion tables except one medium are blank. As described above, the integrated keyword database as the integrated search information is obtained.

Therefore, first, an integrated keyword database is created from a keyword database created independently for each of a plurality of media. This integrated keyword database holds a conversion table relating to corresponding keyword IDs, that is, an ID conversion table of integrated keywords and local keywords for each medium. In addition, a search over a plurality of media is performed using a keyword ID conversion mechanism. By such keyword integration and its ID conversion mechanism, a plurality of media in which keywords are independently created are managed in a single keyword database, and a wide range of keyword searches across a plurality of media becomes possible. That is, the integrated keyword database is recorded in the hard disk of the data storage unit 3, and by using this, the keyword I of the integrated database designated by the user is recorded.
It can be easily realized by converting D into a local keyword ID of each medium and searching the medium.

[0017]

As described above, according to the present invention, by integrating independent search information for each medium, a plurality of media having a search information system constructed without being collectively managed using a hard disk or the like. It is possible to perform a wide range of searches spanning in between.

[Brief description of drawings]

FIG. 1 is a block diagram showing a system configuration of an electronic filing system according to an embodiment of the present invention.

FIG. 2A is a basic structural diagram of a keyword database of medium a, B is a basic structural diagram of a keyword database of medium b, and C is a basic structural diagram of an integrated keyword database of media a and b.

FIG. 3 is a flowchart showing a process of integrating keyword databases created independently in the electronic filing system according to the embodiment of the present invention.

FIG. 4 is a flowchart showing automatic extraction of a keyword having the same name and table creation processing in step S1 of FIG.

[Explanation of symbols]

 1 data input unit 2 system control unit 3 data storage unit 4 data management unit 5 document registration unit 6 document search unit 7 data display unit

Claims (1)

[Claims] 1. An information retrieval device for retrieving stored input information by retrieval information created for each medium, wherein the retrieval information for each medium in the stored input information is integrated and created across the media. An information search apparatus comprising: a data storage unit that stores integrated search information; and a data management unit that can convert the integrated search information of the data storage unit into search information for each medium.