JPH06168271A - Data base retrieval system utilizing dictionary by item - Google Patents

Abstract

PURPOSE:To provide the system for performing retrieval by using a dictionary previously defined for every different data base or for every item of one data base concerning retrieval processing to use the dictionary. CONSTITUTION:This system is composed of a retrieval tag information analysis/ generation part 3 for reading, analyzing a retrieval tag description from a retrieval tag description file 2 and generating retrieval tag information, retrieval formula analysis part 6 for outputting analyzed data 7 by analyzing an inputted retrieval formula, retrieval data generation part 8 for generating retrieval data 9 for each item by collating the analyzed data 7 and retrieval tag information 4, and retrieval execution part 10 to perform retrieval processing by using a dictionary 12 designated based on the retrieval data 9 and to output the final result by merging the respective retrieved results when plural data bases 11 are retrieved.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a database search system, and by predefining which dictionary of which database is to be searched using which dictionary, as a search tag, different databases are suitable for each database. The present invention relates to an item-based dictionary-based database search system that can be searched by using.

[0002]

2. Description of the Related Art Conventionally, when performing a search using a dictionary,
When searching, a method of specifying a dictionary to be used in the series of searches in advance is used. As for the dictionaries that can be specified, one dictionary can be specified for each search. Therefore, the same dictionary is used for every search item and for every database.

[0003]

In the above-described conventional search system, for example, when searching one or more databases at the same time, it is suitable for one database but unsuitable for another database. You will be using a dictionary like this. Alternatively, when searching one database, a dictionary that is suitable for a certain item but not for another item may be used.

An object of the present invention is to provide an item-based dictionary-based database search system for generating search data for each search item from search tag information, designating a database and a dictionary for each item, and performing a search. .

[0005]

The item-based dictionary utilization database search system according to the present invention reads a search tag description from a file storing the search tag description, analyzes the search tag information, and creates search tag information. And a search expression analysis means for analyzing a search expression input for database search and outputting data after analysis processing and the data after analysis processing and the search tag information are collated to obtain search data for each search item. Search data generating means for generating,
A database search system comprising: a search execution unit that refers to a dictionary specified based on the generated search data and outputs a search result across one or more databases.

[0006]

Embodiments of the present invention will now be described with reference to the drawings.

Referring to FIG. 1 showing a system configuration of an embodiment of the present invention, an item-based dictionary utilization database search system 1 includes a file 2 for storing a search tag description for designating a database for each search item and a dictionary, A search tag information analysis / generation unit 3 that reads the search tag description from the file 2 and analyzes it to generate search Doug information, and search tag information 4 generated by the search tag information analysis / generation unit 3.
And a search expression 5 that is input to perform a database search
, The search expression analyzing unit 6 that analyzes the search expression 5, the search expression analyzing unit 6 collates the post-analysis processing data 7 of the analysis result, the post-analysis processing data 7 and the search tag information 4 to obtain the search data 9. Search data generator 8 for generating and search data 9
A search execution unit 10 that performs a search process on a database group 11 using a dictionary 12 based on the above, and a database group 11 and a search execution unit 1 that are the search targets of the search execution unit 10.
The dictionary group 12 used when 0 performs a search, and the set file 13 that stores the result of the search process performed by the search execution unit 10 on the database group 11.

Further, FIG. 2 explaining the information included in the search tag description, FIG. 3 explaining the relationship between the search tag information, the database and the dictionary, and the post-analysis processing data generated by analyzing the search expression 5. 7 for explaining the search data, and FIGS. 5 (a) and 5 (b) for explaining the search data generated by associating the search tag information 4 and the post-analysis processing data 7 with each other. 6 for explaining the dictionary 12 of FIG.
A collective file 1 for associating (a) and (b) with the search result data and the database 11 that stores the data.
3 for explaining the third example, FIG. 8 for explaining the association between the set file 13 and the database 11, and the database 11
9 for explaining an example of analysis processing data 91 designating the dictionary 12 for each item and search data 92 and 93.
The operation of the system configuration of FIG. 1 will be described by using (a), (b) and (c) together.

Referring to FIG. 2 for explaining the search tag description 20 of the file 2, when the search tag "keyword" is specified, the item "KW" of the database "A" is converted into the dictionary "D".
This means that the search is performed using "IC1" and that the item "index code" in the database "B" is searched using the dictionary "DIC2." In addition, the search tag "author" is specified. If the search tag “hometown” is specified, the item “author” in the database “C” is searched using the dictionary “JINMEI”. If the search tag “hometown” is specified, the item “hometown” in the database “C” is dictionary “CHIMEI”. It is shown to search using ". Search tag information analysis ・
The generation unit 3 analyzes the search tag description 2 and generates search tag information (30 in FIG. 3). Referring to FIG. 3 for explaining the relationship between the search tag information (30), the database and the dictionary, the search tag “keyword” (301) includes the item “KW” (311) of the database “A” (31) and the database. "B" (32) item "index code" (321)
There is a relation with and as a search target, and further, when searching the item “KW” (311) of the database “A” (31), the dictionary “DIC1” (34) and the database “B” (32) are searched. When searching the item “index code” (321) of), the use of the dictionary “DIC2” (35) is related. Similarly, the search tag "author" (30
When searching 6), the database "C"
The item "author" (331) of (33) is converted into the dictionary "JINME".
Search using "I" (36) and search tag "birthplace"
When (307) is the search target, the item “hometown” (332) of the database “C” (33) is the dictionary “C”.
Search using HIMEI "(37).

Next, there is an instruction to use the dictionary, and
When “keyword = calculator” is input as the search expression 5, “keyword = calculator” of the search expression 4 is first analyzed by the search expression analysis unit 6 and the post-analysis processing data 7
(40 in FIG. 4) is output. Referring to FIG. 4 for explaining the post-analysis data, the search expression analysis unit 6 determines that the search target item is “keyword” (41), the search term is “computer” (42), and the search condition is exact match (43). , Using the dictionary (44) is output as a result of the analysis.
The search data generator 8 collates the post-analysis processing data 7 with the search tag information 4, and the database “A” (31),
For each "B" (32), search data (5 in FIG. 5)
1, 52) is generated. First, post-analysis processing data (4
Refer to 0) and the item to be searched is "keyword"
(41) and referring to the search tag information (30), the search item “K” is searched for the database “A” (31).
W "(302), dictionary name" DIC1 "(304), and further search item" index code "(305) and dictionary name" DIC2 "(308) for the database" B "(32). And the database "A" (3
For 1), the search target item is "KW" (511), the dictionary is "DIC1" (515), and for the database "B" (32), the search target item is " The index code is "(522)" and the dictionary is "DIC2".
Search data using (525) (51, 52 in FIG. 5)
Is generated. The search term and search condition of the search data 51 and 52 are "calculator" (513, 523), which is a perfect match (514, 524). Furthermore, referring to the dictionary "DIC1" (61 in FIG. 6), the synonyms "computer", "computer",
"Electronic computer" is defined. In addition, the dictionary "DIC
2 ”(62 in FIG. 6),“ E00102 ”is defined as a code to be used in place of“ computer ”. The search execution unit 9 first searches the database“ 52 ”based on the search data 51 and 52. A search is performed for the item "KW" of A. At that time, the dictionary "DIC1" (61) is referred to, and "computer" (611) and "computer" (610), which are synonyms for "computer" (610). Computer ”(6
12) and data containing one of the terms "electronic computer" (613) are retrieved, and the result is stored in the aggregate file of aggregate file number n (73 in FIG. 7). This aggregate file (73) is a bit string created by associating 1 bit with 1 item of data for each data stored in the search target database, and the search term specified in the data regarding the search processing is When it exists, the bit corresponding to the data is set to "1", and when the search term does not exist, the bit is set to "0". Referring to FIG. 7 for explaining the relationship between the aggregate file (73) and the data (71, 72), the data (71) in the first item "KW" of the database "A" is "computer", "computer". Since the synonyms of "", "computer", and "electronic computer" are not included, the aggregate file number n
The leading bit of the set file (73) is “0”.
Since the second data (72) includes "computer", the second bit is "1". The same applies to the third and subsequent data. Further, referring to the search result management table (80 in FIG. 8) indicating the association between the aggregate file number and the database, the search execution unit 10 determines that the item “index code” (32 in the database “B” (32).
Search for 1). At that time, the dictionary "DIC2"
In (62), "E00102" (621) is defined as a term to be used instead of "computer". Therefore, the data is searched using the term "E00102" (621), and the result is a collective file. The database B is stored in the aggregate file of number n + 1.

Similarly, by changing the search tag information, the same database can be searched by changing the dictionary for each item. When the search expression “author = sato AND birthplace = Tokai” is input, the search expression analysis unit 6 generates post-analysis data (91 in FIG. 9). That is, the items to be searched are "author" (911) and "hometown" (912), and the search terms are "sato".
(913) and “Tokai” (914), and the search condition is a logical product. The search data (92, 93 in FIG. 9) is generated by collating this data (91) with the search tag information (30). This search data (92, 9
3) is used to refer to the search data 92 for the item “author” (331) of the database “C” (33),
The search term is "Sato" (921) and the dictionary is "JIN".
MEI "(922) is used. Also, the item" Hometown "
The search data 93 is referred to for (332), the search term is “Tokai” (932), and the dictionary is “CHIMEI”.
The database C is searched using (933). After that, the two search results are logically ANDed and stored in the aggregate file of the aggregate file number m.

[0012]

As described above, according to the present invention,
Searching one or more databases using the dictionary suitable for each database by predefining which dictionary in which database should be used to search which items, and the items in one database You can search by using the appropriate dictionary for each.

[Brief description of drawings]

FIG. 1 is a diagram showing a system configuration of an embodiment of the present invention.

FIG. 2 is a diagram illustrating search tag description.

FIG. 3 is a diagram illustrating a relationship between search tag information, a database, and a dictionary.

FIG. 4 is a diagram illustrating data after a search expression analysis process.

FIG. 5 is a diagram illustrating search data.

FIG. 6 is a diagram illustrating a dictionary file.

FIG. 7 is a diagram illustrating a relationship between a collective file and data.

FIG. 8 is a diagram illustrating a search result management table.

FIG. 9 is a diagram illustrating post-analysis processing data and search data.

[Explanation of symbols]

 1 Item-based dictionary usage database search system 2 Search tag description file 3 Search tag information analysis / generation unit 4 Search tag information 5 Search formula 6 Search formula analysis unit 7 Data after analysis processing 8 Search data generation unit 9 Search data 10 Search execution unit 11 database group 12 dictionary group 13 set file 20 search description tag 30 search tag information 31 database A 32 database B 33 database C 34 dictionary DIC1 35 dictionary DIC2 36 personal name dictionary JINME 37 place name dictionary CHIME 40 post-analysis processing data 51 search data 52 search Data 61 Dictionary DIC1 62 Dictionary DIC2 71 First data of database A 72 Second data of database A 73 Collective file 80 Search result management table 91 Post-analysis processing data 92 Search data 93 Search data

Claims (1)

[Claims] 1. A search tag information analysis / generation means for reading and analyzing the search tag description from a file storing the search tag description and generating search tag information, and a search expression input for database search, and analyzing. Based on the generated search data, a search expression analysis unit that outputs post-processing data, a search data generation unit that collates the post-analysis processing data and the search tag information, and generates search data for each search item. An item-based dictionary-based database search system comprising: a search execution unit that outputs a search result across one or more databases by referring to the dictionary specified in.