JPH03246764A - Data base retrieving system - Google Patents

Abstract

PURPOSE:To select one or more files suitable for a desired content by generating the histograms of all key words attached on data on which a retrieval key word is attached at every file or designated plural files. CONSTITUTION:A key word conversion part 1 performs conversion to the retrieval key word X' by performing the retrieval of a key word dictionary 2 based on a word inputted as a retrieval word, and retrieves every file comprising data bases 4, 4',... based on the key word X', and extracts the data on which the retrieval key word X' is attached. Next, all the key words attached on extracted data are extracted at every file or designated plural files, and they are outputted to a histogram calculation part 3. The histogram is generated by arranging the key word at every file or designated plural files, and the retrieval key word X' and the histogram are displayed on a display part 5. In such a way, it is possible to easily select the file optimum for a retrieval content from the data bases 4, 4',....

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a database search system in which a database user searches a database for a file that best matches the search content.

[Conventional technology]

Traditionally, when searching a database, the task of finding and selecting files that contain a lot of the content desired by the searcher from among the large number of files that make up the database is a process that the searcher has spent many years working on. A lot of it depended on experience.Also, database services such as DIALOG provide a service that displays the number of data hits for a specified keyword for each file.

[Problem to be solved by the invention]

However, even in this case, it was insufficient for the searcher to know the field of each hit data. Furthermore, when setting this keyword, there was a problem with the wording, for example, the wording ``computer'' would get a hit, but the wording ``computer'' would not get a hit. Furthermore, there are similar problems with abbreviations, such as expressing ``National Athletic Meet'' with ``Kokutai,'' especially in English, problems with spelling errors and hyphenation in compound words, and ``computer'' with ``calculator.'' There was a problem with synonyms or derivative words.

The present invention solves these problems, and when an amateur searches a database, he or she can immediately find words that should be used as keywords even if he or she does not have many years of experience in search operations.

The purpose is for the searcher to select one or more files from the database that are most suitable for the content desired by the searcher.

[Means to solve the problem]

The present invention is designed to solve the problem of file selection, which is encountered especially by beginners in database searches, and the problem of keyword setting that arises. , has a means for converting keywords into keywords that can be searched in a database by translation etc., and when selecting a file, based on the search keyword set by the keyword conversion means, all the data attached to the search keyword are added. A means is provided to create a histogram of keywords for each file or for each specified multiple files.

[For production]

By the above means, it is possible to convert the words manually created by the database searcher into keywords for database search, and furthermore, for one or more data in the file to which the search keyword is attached, the words added to these data can be From the distribution of all the keywords in the file, we can determine the content and content of the data stored in the file.
You can easily learn about fields, trends, etc. It goes without saying that the search keyword conversion means is useful not only when searching for files, but also during normal searches after fixing files. Furthermore, once a file has been fixed, it is sufficient to enter a search keyword into the fixed file and display the search results on the display unit.

〔Example〕

FIG. 1 shows an embodiment of the present invention. Reference numeral 1 in the figure is a keyword conversion unit, which performs a search in the keyword dictionary 2 based on the word input as a search word via the signal line 11, and if necessary, searches from the input word X. Convert to keyword X゛.

A signal line 12 is used to input a search word to the dictionary 2 and output the converted search keyword. In addition.

If the input word can be used as a search keyword as is, it will not be converted. 3 in keyword dictionary 2
For example, the word "interface" is registered as a keyword, and "interface" can be entered by a database searcher.

Conversion pairs with words such as "interface" are registered. Keyword Dictionary 2 provides synonyms, similar words, abbreviations,
In addition to registering word pairs such as translated words, you can also use a thesaurus etc. to solve problems with derived words, and especially in English, use a dictionary to correct spelling errors and omissions.
“First C1ass +”First-class
It has a function to solve problems with hyphenation and compound words such as s + "Firstclass", and a function to register and delete new dictionaries as necessary (
It will be done like this. The search keyword X°, which is obtained by converting the search word X input by the user by the keyword conversion unit l, is sent to the histogram calculation unit 3 via the signal line 13. Transferred to database 4 and database 4°. Note that it is clear that the search word input to the keyword conversion unit 1 may be in the form of not only a word but also an expression in which words are combined using operators such as and and or. It is also clear that similar processing is possible even if there are three or more databases.

As shown in FIG. 1, the database is composed of multiple files with different contents.

Each file stores a plurality of pieces of data each assigned one or more keywords.

In database search, among these data,
Data hinted at the searcher's keywords is output.

A database 4 containing a plurality of files, ie, "File 1", "File 2J", . . . , "File n".
Furthermore, “File 1°”, “File 2゛”, etc.
In the database 4° containing "file n", first, each file constituting the database is searched based on the search keyword X' sent from the keyword conversion unit 1, and the search keyword X' is assigned. Extract the data that is

Next, all keywords assigned to the extracted data are extracted for each file or for each specified plurality of files, and the histogram calculation unit 3 uses the signal line 14 to
Output to. Naturally, these keywords include the search keyword X'.

The histogram calculation unit 3 organizes the search keywords input via the signal line 14 for each file or for each specified plurality of files to create a histogram. Furthermore, the search keyword X" inputted using the signal line 13 is outputted using the signal line 15.

The display unit 5 displays the input search keyword X'' and histogram using the signal line 15.

An example of the display results is shown in FIG. In FIG. 2, in "File 1" and "File 2", the number of data containing keyword A is "3" and "2", and similarly, the number of data containing keyword B is "10" and "2", respectively. 1"
, the number of data containing keyword C is "3" and "2",
The number of data containing keyword D is “1” and rlOJ
, Furthermore, the number of data containing the search keyword
"File l" and "File 2" are both "15".

Here, even though the number of data to which the search keyword Since there is a large amount of data containing D, it can be seen that the properties of the data stored in "File 1" and "File 2" are different. Therefore, the database searcher needs keyword B and keyword D.
By comparing these files, you can select files that contain many keywords that are close to the field you want to search.

In traditional database search systems.

In such a case, the searcher is only given the information that the number of data containing the search keyword , I couldn't decide on the spot which file was going through me.

By the way, after fixing the search file.

The search keyword output from the keyword conversion unit 1 is
It is sufficient to output the search results to the search file and output the search results to the display section 5 via the signal line 16.

〔Effect of the invention〕

As explained above, according to the present invention, even when an amateur searches a database, it is possible to easily set search keywords and to easily select the file most suitable for the search content from the database. can.

[Brief explanation of drawings]

FIG. 1 shows an embodiment of the present invention, and FIG. 2 shows a display example of a histogram calculated by a histogram calculating section. 1...Keyword converter, 2...Keyword dictionary. 3... Histogram calculation section, 4.4'... Database, 5... Display section, 11 to 16... Signal line.

Claims (1)

[Claims] In a database search device that selects one or more files containing a large amount of data with content desired by a searcher from one or more databases containing a plurality of files, the database search device includes: a search keyword or a search; Search each file for data that matches a search formula that combines keywords using a logical formula, and calculate the frequency of occurrence of the keywords assigned to each of the search data within the same file or within multiple specified files. What is claimed is: 1. A database search system characterized in that the histogram is obtained by calculating the histogram, and the histogram is displayed.