JPH06161844A - Data base managing device - Google Patents

Abstract

PURPOSE:To provide the device which can prevent any useless access path from being formed, can shorten processing time and enables effective support with coupling arithmetic. CONSTITUTION:This device is provided with a storage means 14 for storing the degree of relation between respective relations stored in a data base 11 and keywords defined corresponding to them, input means 13 for inputting a keyword group, weight for each keyword and threshold value of the total relation degree, total relation degree calculating means 15 for calculating the total related degree of each relation related to one keyword at least to the entire keyword group based on the keyword group and the weight for each keyword from this input means 13, and relation judging means 16 for judging and outputting the relation exceeding the threshold value of the total related degree inputted by the input means 13 among the total related degrees calculated by this means 15.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a relational database (hereinafter simply referred to as a database) in which a database stored in a knowledge database uses a relation model, and in particular, which attribute is combined to process a join operation. The present invention relates to a database management device that has a beneficial effect when used in deciding whether to perform.

[0002]

2. Description of the Related Art In recent years, relational databases have been
It is widely used in various fields due to the transparency of the data model and the high degree of data independence. In addition, the amount of data stored in databases is also increasing significantly, and there are many large databases.

When the number of relations increases in such a system, it becomes difficult for the user to grasp all the relations. In particular, the join operation, which is one of the main operations in the database, cannot be executed unless the user completely describes the names of multiple relations and their relationships. Here, the relation refers to a combination of a plurality of different types of attributes that have a predetermined relationship.

From the above, a relation search device has been developed in the related art. When a desired attribute group and a conditional attribute group are given, attributes are matched between relations to obtain these attributes. It automatically creates candidates for access paths that connect all of them. Therefore, the user can select the access path that he / she needs from the plurality of access paths searched by the relation search device, and execute the join operation using the selected access path.

[0005]

However, in the conventional relation search device, since the access path is formed for all the relations in the system, the access paths are formed by combining the meaningless attributes. It may happen. As a result, many unnecessary access paths are generated, the user takes time to select the access paths, and it takes time to calculate the unnecessary access paths, resulting in a decrease in processing efficiency. There is.

A conventional relation retrieval device will be described with reference to FIGS. 8 and 9. FIG. 8 shows relation groups stored in the database system. (A) is a toy relation, (b) is a picture book relation, (c) is a maker relation, (d) a publisher relation, and (e) is a maker's location. relation,
(F) is the publisher location relation.

[0007] In the example of such a relation group, the user wants to know the location of the picture book publisher, but since he / she does not know the relation name or the attribute of those relations, the "Price is Enter a question such as "Which location produces less than 1000 yen?" And perform a relation search.

At this time, in the example of the database shown in FIG.
The relation search device forms two access paths shown in FIG. That is, the first access path connects the toy relation (a) and the maker relation (c) with the attribute "name", and as a result, the relation and the maker location relation (e) are joined with the attribute "maker". It is an access path that can be done.

The second access path connects the picture book relation (b) and the publisher relation (d) with the attribute "ID", and as a result, the relation and the publisher location relation (f) with the attribute "publisher". It is an access path that can be combined. Such two types of access paths are automatically obtained. Now, what I really want is an access pass for picture books, so an access pass for toys is unnecessary.

Therefore, the present invention provides a database management device which can prevent the formation of meaningless access paths, can shorten the processing time, and can more effectively support the join operation. To aim.

[0011]

In order to achieve the above object, the invention according to claim 1 has a plurality of relations in which a plurality of attributes having different types are associated with each other. A stored database,
Storage means for storing keywords defined corresponding to each relation stored in this database,
An input means capable of inputting a keyword, and a relation stored in the storage means is searched based on the keyword input by this input means, and which attribute of which relation is combined and combined with respect to the relation. A procedure that indicates whether to process an operation (this is called an access path)
And a relation search means for outputting this access path.

Further, in order to achieve the above object, a second aspect is provided.
The invention corresponding to is a database that stores a plurality of relations in which a plurality of attributes having a predetermined relationship are grouped into one among a plurality of different types of attributes, and each relation stored in this database Storage means for storing the degree of association with a correspondingly defined keyword, input means capable of inputting a keyword group, weight for each keyword, and threshold value of total degree of association, and keyword input by this input means Total relevance calculating means for calculating the total relevance of the relation related to at least one keyword to the entire keyword group based on the group and the weight for each keyword, and the total relevance calculated by the total relevance calculating means. Of the relations exceeding the threshold value of the total relevance inputted by the input means, and outputs the relations. A database management apparatus and a constant section.

[0013]

According to the invention corresponding to claim 1, since only the relations related to the keyword inputted to the input means are searched, generation of a necessary or meaningless access path is eliminated, which is efficient and useful. Since it is possible to search only the access path, and thus unnecessary processing is not performed, the processing time can be shortened.

According to the second aspect of the present invention, the relations related to the keywords input to the input means are searched, and the weights corresponding to the respective keywords with respect to those relations and the relations and the keywords are obtained. Since the relation whose total degree of association is equal to or more than the threshold value is searched, meaningless access path is not generated, and unnecessary processing is not performed, so that the processing time can be shortened. .

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below with reference to the drawings. First, a schematic structure of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram showing a schematic configuration of the invention corresponding to claim 1, in which a plurality of attributes having different types have a predetermined relationship.
A database 1 in which a plurality of related relations are stored, a storage unit 2 for storing a keyword defined corresponding to each relation stored in the database 1, and an input unit 3 capable of inputting a keyword. Based on the keyword inputted by the input means 3, the relation stored in the storage means 2 is searched, the access path is obtained for the relation, and the relation search means 4 for outputting the access path and these 1 4 is included in the management means 10.

FIG. 2 is a block diagram showing a schematic structure of the invention corresponding to claim 2, and a plurality of relations in which a plurality of attributes having different types are associated with each other in a predetermined relation are stored. Database 1, storage means 5 for storing the degree of association between each relation stored in this database 1 and the keyword defined corresponding thereto, keyword group, weight for each keyword, and total association Inputting means 6 capable of inputting a threshold value of the degree, and based on the keyword group inputted by the inputting means 6 and the weight for each keyword, the total degree of association of the relation related to at least one keyword with respect to the entire keyword group Of the total degree of association calculated by the total degree of association calculating means 7 and the total degree of association calculated by the total degree of association calculating means 7. Determining means 8 determines a relation in excess of the total relevance of the threshold that is, outputs this
And a management means 10 for managing these 5 to 8.

FIG. 3 is a block diagram showing a schematic configuration of an embodiment of the present invention, which is a database 11 and a management means 12.
The input unit 13, the storage unit 14, the total degree-of-association calculating unit 15, the relation determining unit 16, and the relation searching unit 17 are included.

The database 11 stores the relations (a) to (f) shown in FIG. Input means 13
Includes a keyword input unit 13k, a weight input unit 13w, and a threshold value input unit 13s, and these can input the keyword k, the weight w, and the threshold value s, respectively. The storage unit 14 stores a keyword and a degree of association between the keyword and the relation.

Here, the degree of relevance is that the database designer sets the depth of the relation between the content of the relation and its keyword in relation to all the relations stored in the database 11. It is a value.

The total relevance calculating means 15 is for obtaining the total relevance related to the keyword group input from the input means 13. This is a keyword of each relation based on the calculation method of the total relevance described later. The sum of the degree of relevance to and the sum of the weights of the keywords is calculated.

The relation determining means 16 is the input means 1
Based on the threshold value input from 3, total association degree calculation means 1
The total degree of association calculated in 5 determines a relation that is equal to or greater than the threshold value.

The relation searching means 17 searches the access path by using the managing means 12 with only the relation group output from the relation judging means 16 as a target relation.

The operation of the database management apparatus configured as above will be described. FIG. 4 is a flow chart for explaining this. The degree of association between the relation stored in the storage means 14 and its keyword is given in advance by the system administrator and is as shown in FIG. 5 (S1). Here, a plurality of keywords are allowed for one relation, and the degree of association between the relation and the keyword is represented by an integer from 0 to 100.

At this time, in addition to the question for relation search from the user, the keywords "child" and "learning" are added.
It is assumed that the respective weights 1.0 and 0.8 and the threshold 150 of the total degree of association are input. The keyword "children",
“Learning” is stored in the storage unit 14, each keyword weight of 1.0 and 0.8 is stored in the total relevance calculation unit 15, and the total relevance threshold 150 is stored in the relation determination unit 16.
Sent to.

In the storage means 14, all the relations related to at least one keyword, in this case, the six relations of FIG. 8 are selected, and the relation name and the relation degree information as shown in FIG. Then, the total degree of association is calculated (S2). For example, in the case of a toy relation, total relevance = (relevance of keyword “child”) × (weight) + (relevance of keyword “learning”) × (weight) = 100 × 1.0 + 30 × 0.8 = 124 Is calculated as FIG. 7 shows the calculation result of the total degree of association of each relation.

The total relevance calculating means 15 sends the result to the relation determining means 16. Relation judging means 16
Then, the relation having the total degree of association equal to or more than the threshold value is selected (S3). Here, three relations of picture book, publisher, and publisher's whereabouts value are selected. The relation determining means 16 sends this relation group to the relation searching means 17. The relation search means 17 executes those relation searches. As a result, only the access path for the picture book is generated, and the path for the toys is not generated. Therefore, only the required access path is required (S4).

According to the above-described embodiment, formation of a meaningless access path, which is a problem of the conventional relation retrieval apparatus, can be prevented to some extent. Further, since wasteful processing is not performed, the processing time can be shortened, and it becomes possible to provide more effective support to the user during comprehensive calculation.

In the above-mentioned embodiment, the keyword group, the weight and the threshold value are inputted from the input means 13. However, when only the keyword group is inputted, only the relations related to the keyword are searched. As a result, unnecessary or meaningless access path generation is eliminated, only useful access paths can be efficiently searched, and unnecessary processing is not performed, so that processing time can be shortened.

[0029]

According to the present invention, it is possible to prevent a meaningless access path from being formed, shorten the processing time, and provide a database management device capable of more effectively supporting a join operation. be able to.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an invention corresponding to claim 1 of a database management device of the present invention.

FIG. 2 is a block diagram showing a schematic configuration of an invention corresponding to claim 2 of the database management apparatus of the present invention.

FIG. 3 is a block diagram showing a schematic configuration of an embodiment of a database management device of the present invention.

4 is a flowchart for explaining the operation of the embodiment of FIG.

5 is a diagram showing a relationship between keywords and degrees of association for explaining the operation of the embodiment of FIG.

FIG. 6 is a diagram showing the relationship between keywords and degrees of association for explaining the operation of the embodiment of FIG.

FIG. 7 is a diagram showing the total relation degree of relations for explaining the operation of the embodiment of FIG.

FIG. 8 is a diagram for explaining a relation group of a conventional database management device.

FIG. 9 is a diagram of access paths that can appear in a conventional database management device.

[Explanation of symbols]

1 ... database, 2 ... storage means, 3 ... input means, 4 ...
Relation search means, 5 ... storage means, 6 ... input means,
7 ... Total relevance calculation means, 8 ... Judgment means, 10 ... Management means, 11 ... Database, 12 ... Management means, 13 ... Input means, 14 ... Storage means, 15 ... Layer relevance degree calculation means, 16
... relation determination means, 17 ... relation search means.

Claims (2)

[Claims] 1. A database that stores a plurality of relations that have a predetermined relation among a plurality of different types of attributes, and a relation that corresponds to each relation stored in this database. A storage means for storing the defined keyword, an input means capable of inputting the keyword, and a relation stored in the storage means are searched based on the keyword input by the input means, and the relation is targeted. A database management device comprising: a relation search unit that obtains a procedure (which is referred to as an access path) that indicates which attribute of which relation is combined to process a combination operation, and that outputs the access path. 2. A database in which a plurality of relations having a predetermined relationship among a plurality of attributes of different types are stored, and each relation stored in this database and corresponding thereto. Storage means for storing the degree of association with a keyword defined by the above, input means capable of inputting a keyword group, a weight for each keyword, and a threshold value of the total degree of association, and a keyword group input by this input means And a total relevance calculating means for calculating the total relevance of the relation related to at least one keyword with respect to the entire keyword group based on the weight of each keyword, and a total relevance calculating means for calculating the total relevance. Of these, the relation that exceeds the threshold of the total relevance inputted by the input means is judged and the judgment is made to output this. Database management device having means, a.