JPH04342069A - Data base retrieving device - Google Patents

Abstract

PURPOSE:To retrieve a chain relation between respective items by a common key word and to improve the efficiency of retrieving work. CONSTITUTION:When an input part 1 inputs a retrieving instruction including the key word, a retrieving processing part 4 obtains at first an identification code from the key word inputted by using an index file 2 storing index data where the key word and the identification code of the item related with the key word are combined and a master file 3 storing item data where the identification code of the item and the key word related with the item are combined. Item data having this identification code is retrieved to obtain a relevant first item data. Second item data is retrieved from the key word described in the first item data. Subsequently, the data are retrieved until an (n)th item data similarly and the result is outputted to an output part 5.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a search device for a database stored on an electronic medium.

FIG. 5 shows a conventional database search.
This will be explained using the database search device diagram shown in FIG. A keyword related to the item to be searched is inputted from the terminal device 10. The index file 11 stores index data in which a keyword k1 is paired with identification codes aa, bb, and cc of items related to the keyword k1, and the search processing unit 13 identifies the item corresponding to the input keyword kn. code x
Output x, yy, zz. The master file 12 stores item data that is a set of an identification code, an item corresponding to the identification code, and keywords km and kn related to this item.

Next, the master file 12 is searched based on the identification codes xx, yy, zz searched from the index file 11, and items corresponding to the identification codes xx, yy, zz are searched, stored in the output buffer 14, and used as needed. Depending on the terminal device 1
Displayed on 5. If the user is not satisfied with this display, he/she continues the search by entering more related keywords.

0004

[Problem to be Solved by the Invention] When performing a database search as described above, searching for other items that have content related to the keyword attached to a certain item requires trial and error, depending on the index or guessing by the user. There was nothing else. For this reason, searching a database with a large number of items, such as articles from newspapers and magazines spanning several years, requires time, user ability, and
Experience required. The amount of information generated in society is rapidly increasing, and in order to cope with this, improvements in database search efficiency and automation are required.

The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide a database search device that improves the efficiency of search work by detecting chain relationships between items based on common keywords. .

[0006]

[Means for Solving the Problems] FIG. 1 is a diagram showing the principle of the present invention. 1 is an input unit for inputting search instructions including keywords; 2 is an index file that stores index data consisting of keywords and identification codes of items related to the keywords; 3 is an index file that stores index data consisting of pairs of keywords and identification codes of items related to the keywords; 3 is an input unit for inputting search instructions including keywords; A master file 4 stores item data consisting of sets of related keywords, which searches the index file 2 using the keywords input from the input section 1 to search for related identification codes, and searches the master file 4 based on this identification code. File 3
5 is a search processing unit that searches for the corresponding first item data, searches for the second item data from the keyword described in the first item data, and searches up to the nth item data in the same manner; 5 is this search processing unit; This is an output section that outputs the search results of section 4.

[0007] Also, when obtaining the n-th item data, n
The numerical value is specified in advance from the input section 1.

[0008] Furthermore, the search is terminated when at least part of the first item data and the nth item data match.

[0009]

[Operation] With the above configuration, when a keyword is entered, items related to this keyword, then other items related to this item, and so on are searched sequentially.
The information is displayed on the output unit 5, printed, or otherwise output. This makes it possible to search over a wide range of areas, and it is also possible to perform searches automatically.

[0010] Furthermore, by specifying the number of items to be detected in advance, it is possible to obtain a desired amount of desired items.

[0011] Also, since each item is often in the form of a loop via keywords, it is more efficient to terminate the search when the first item is returned or when a partially matched item is found. You can perform searches.

0012

Embodiments Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 2 is a diagram showing the configuration of this embodiment. In the figure, a terminal device 10, an index file 11, a master file 12, an output buffer 14, and a terminal device 15 have the same functions as the conventional device shown in FIG. Note that the items stored in the master file 12 are the minimum unit of a collection of information from the user's perspective. Further, the added search file 16 stores relationship information between items mutually associated with each other by keywords, as shown in FIG. The search processing unit 13 has an index file 11 and a master file 1.
2. A function for managing the exchange of information between search files 16 has been added.

Next, the operation will be explained. (2) The user inputs into the terminal device 10 a keyword related to a desired item and a search level indicating the range of related items desired. This search level is set in advance and depends on the number of items and keywords. ■ The device searches file 16 for the input keyword.
Register. (2) Items in the master file 12 related to the keyword input in (2) and keywords related to the identification code are displayed on the display device 15, so select any one of these keywords and instruct a search. (2) The device copies each item on the master file 12 into the work area using the identification code in the identification code area on the index file 11 that stores keywords. The following operations are automatically performed by the device.

■ Pick up keywords other than the keyword of interest for each item in the work area, and search for search file 1.
Register on 6. ■ Compare the keyword registered in ■ with the keyword registered in ■, and if there is a matching item, write that item to the output buffer 14. ■ Repeat the processes from ■ to ■ for each keyword in each item. ■ The search level set by the user in ■ (This is ~
(Responsible for the number of repetitions of ■)) If the number of repetitions is exceeded, the process is terminated. ■ The search ends when the chain of searched keywords completes one cycle and returns to the original item displayed with ■.

FIG. 3 is a diagram showing the above-mentioned operation. A search is started by focusing on the keyword km indicated by a star in the figure. It can be seen that the keyword km exists in common for items aa and bb. Furthermore, focusing on item bb, an item having the keyword kp is searched. Thereafter, by sequentially repeating this process, it becomes possible to detect a chain relationship of keywords that goes around once as shown below. aa → (km) → bb → (kp) → cc → (kx)
→dd→(kz) →aa

[0016] This detected result is generated and registered as an item xxx at a higher level using an identification code xxx. Since the information in item xxx is in a compressed format, it provides an efficient working environment for database users.

FIG. 4 shows items aa and bb as shown in (b).
, cc and the upper level item xx as shown in (a)
It is a figure which shows the specific example of x. Each item a in item xxx
A situation is shown in which the information of a, bb, and cc is compressed and displayed.

As described above, by focusing on a keyword in a certain item and specifying a search range, the user searches for items having the same keyword in a chain. Then, the search continues until the same item is found again. For the combination of items that have gone through one cycle, the identification code of each item is recorded and registered as a higher level item. The user can obtain the elements of the combination of extracted items by referencing only these higher-level items. In this way, automation of information compression centered on the focused keywords is achieved.

By performing the above-described search for the higher-level items themselves, it is possible to derive a summary of a higher-level hierarchy. By repeating this operation, multi-layered item summaries can be realized, and search efficiency can be achieved.

[0020]

[Effects of the Invention] As is clear from the above explanation, the present invention compresses information by searching for relationships between keywords and extracting combinations of items with a strong degree of connection.
Improve search efficiency.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the principle of the present invention.

FIG. 2 is a diagram showing the configuration of an embodiment.

FIG. 3 is a diagram showing the operation of this embodiment.

FIG. 4 is a diagram showing a specific example of search items in this embodiment.

FIG. 5 is a diagram showing the configuration of a conventional database search device.

[Explanation of symbols]

10, 15 Terminal device 11 Index file 12 Master file 13 Search processing section 14 Output buffer 16 Search file

Claims (3)

[Claims] 1. An input unit (1) for inputting search instructions including keywords, an index file (2) for storing index data consisting of keywords and identification codes of items related to the keywords, and A master file (3) that stores item data consisting of a combination of the identification code and a keyword related to the item, and the index file (2) that is searched using the keyword input from the input section (1) and associated items are searched. Search for the identification code, search the master file (3) for the corresponding first item data based on this identification code, search for the second item data from the keyword described in the first item data, and do the same. A database search device comprising: a search processing unit (4) that searches up to the n-th item data; and an output unit (5) that outputs the search results of the search processing unit (4). 2. The database search device according to claim 1, wherein a numerical value of n for obtaining the n-th item data is specified in advance through an input section (1). 3. The database search device according to claim 1, wherein the search is terminated when at least a portion of the first item data and the n-th item data match.