JPH01282634A - Data base device - Google Patents

Abstract

PURPOSE:To select the desired index out of the held classification indexes and to attain the optimum retrieval in accordance with the use purpose by automatically preparing and holding the classification index of the hierarchy construction in accordance with the use purpose when a classification item in accordance with the use purpose is designated. CONSTITUTION:An index preparing means 3, when the classification item of the hierarchy construction composed of the combination for an arbitrary classification item is designated, prepares the classification index of the hierarchy construction to pass through these classification items and holds it at an index holding means 2. Consequently, when the classification item of the hierarchy construction is designated by matching to the use purpose, the classification index to pass through these classification items is held by the index holding means 2, therefore, by selecting the classification index matched to the use purpose, the desired data can be reached. Thus,the retrieval optimum for the use purpose can be executed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a database device that accumulates mutually associated data and is capable of searching for desired data therein in a variety of formats.

[Conventional technology]

Conventionally, in this type of database device, a set of record data consisting of a plurality of classification items is held in a mass storage device such as a magnetic disk memory device. For example, to give an example of personal information for company e1, as shown in FIG. 5, name, age, and age.

Data are divided into classification items such as department, department, and annual income, and are maintained for each individual. In this case, the area that holds data regarding each individual is called a record, and the area within each record that holds data classified by category such as name, age, etc. is called a field.

Therefore, when searching for desired data from data@structured data in this way, decide which fields should be large items, medium items, and small items, and create a search formula that traces each of these items from the top. Enter this search expression. Then,
A search process is performed according to this search formula, and desired data can be obtained.

On the other hand, as the number of classification items increases, it becomes difficult to create a search formula. Therefore, by setting a classification index in which classification items such as name, age, etc. are classified into a hierarchical structure, and following this classification index in order, the desired data can be retrieved. This method of searching is used in commercial databases, etc.

(Problem to be Solved by the Invention) However, the classification index used in the commercial database has only one type of hierarchical structure. For this reason, when multiple users with different purposes use the service, or when a single user uses the service for many purposes, unnecessary layers are often forced to be traced. However, the problem is that it is difficult to use.

An object of the present invention is to provide a database device that can perform optimal searches for each purpose of use.

(Means for Solving the Problems) The present invention provides a data holding means for holding a plurality of sets of data consisting of a plurality of classification items as one set of record data, and a method for searching for data of a desired classification item in the record data. For this purpose, an index creating means for creating a classification index of the structure, an index holding means for holding a plurality of sets of created classification indexes, an index selection means for selecting one of the held classification indexes, and a selection means. The above object is achieved by comprising a search means for searching the record data held in the data holding means according to the classified index.

[Operation] When a hierarchically structured classification item consisting of a combination of arbitrary classification items is specified, the index creation means creates a hierarchically structured classification index that traces these classification items,
The index is held in the index holding means. Therefore, if you specify the classification items in the 15-layer structure according to the purpose of use, the classification index that traces these classification items is held in the index holding means, so you can select the classification index that matches the purpose of use. This makes it possible to arrive at the desired data and perform an optimal search for each purpose of use.

[Embodiment] FIG. 1 is a block diagram showing an embodiment of the present invention, in which a data holding device W1 composed of a magnetic disk memory device, etc.
1 (hereinafter referred to as the database body), an index holding device 2 that holds multiple types of classification indexes,
Process for creating classification indexes and data searches M31n
It is composed of 3.

The processing device 3 is a terminal device (not shown) operated by a user.
a communication control unit 30 for transmitting and receiving search requests and grinding results to and from the processing unit 3, and a central unit 31 that controls each unit in the processing unit 3 and creates a classification index according to the purpose of use. have. Furthermore, there is a classification index processing unit 32 that switches between multiple types of indexes held in the index holding device Q2 and stores in the holding device 2 a classification index created according to the purpose of use, and a classification index processing unit 32 that edits the classification index. Index editorial department 33
, a database search unit 34 that searches data held in the database body 1 using keywords and record numbers, and a database editing unit 35 that deletes, edits, or adds data held in the database body 1. It is equipped with

In the above configuration, for example, it is assumed that the employee personal information shown in FIG. 5 is held in the database main body 1, and from this employee personal information, as shown in FIG.
The procedure for creating a classification index when searching for people in their 20s with an annual income of 5 million yen or less will be explained as an example.

In this case, first, the highest IIPi layer of the classification items is the department, the next layer is age, and the next layer is annual income, so
As shown in the flowchart of the classification index creation algorithm in FIG. 3, the field name of each layer and its search range are input and specified.

That is, the field name of department 9 age is specified, and the age is specified as 20 to 29 years old, and the annual income is specified as 0 to 5 million yen (step 100 in FIG. 3).

The classification items (i.e., field names) of each layer are input into a terminal device operated by a user, received by the communication control unit 30 via a communication line, etc., and then input into the central processing unit 31.

Central processing unit 31G, once the classification items of all the layers have been input, create a logical expression for a search formula that traces the input classification items of each layer from the top layer to the bottom layer (
Steps 101 and 102 in Figure 3).

Here, each layer of the classification index is described using keywords such as AAA, BBB, and EEE, as shown in Figure 4(a), where AAA is the department, BBB is the age, and EE is the keyword.
Assuming that E is a keyword for annual income, these keywords and their corresponding contents are stored as one classification index file in the classification index holding device c! 2 is stored in one of files 2A to 2C. For example, A.A.
Assuming that the classification index file with keywords A, BBB, and EEE is 28, this file 2A has keywords and their contents, as shown in w54 figure (b).
and identifiers of classification items that are subordinate to the classification item indicated by each keyword. In this case, the identifier rOJ,
"0" in NJ indicates that the classification item in the relevant hierarchy is stored, "1" indicates that the record number of the condition that matches the logical expression is stored, and AAA
The top level classification item indicated by the keyword is $
Searched using the keyword T-OP $.

Therefore, the central processing vR unit 31 transfers the created logical formula to the database search section 34, and searches the database main body 1 for data that match the conditions indicated by the logical formula.

If the database search unit 34 detects a record in which data matching the condition expressed by the logical formula is stored from among the employee personal information, the database search unit 34 returns the number of the record to the central processing unit @31.

Here, if the file that stores the classification index for searching for people in their 20s who belong to the general affairs department and have an annual income of 5 million yen or less is 2A of files 2A to 2C, the central processing unit 31 inputs The recorded record number is stored in the table area corresponding to the classification item "annual income" at the lowest level in file 2A (step 103 in Figure 3).
).

That is, the record number of the database main body 1 in which data matching the logical formula is stored is stored in the table area corresponding to the keyword EEE in the classification index file 2A having the table structure shown in FIG. 4(b).

As a result, the age of the person belonging to the General Affairs Department shown in Figure 2 is 2.
A classification index for people under the age of 5 with an annual income of 5 million yen or less is created in a hierarchical structure as shown in FIG. 4(a).

Therefore, we created a classification index according to the search purpose,
If the index file 2 is stored in advance, the desired data can be reached by selecting a classification index according to the purpose and tracing the hierarchy therein. Therefore, there is no need to trace unnecessary hierarchies during a search, making it possible to perform an efficient search that suits the user's purpose. In addition, since there is no need to input a search formula each time, the operation becomes simple, and even inexperienced users can easily proceed with the search work.

By the way, when new WA layer data is added or deleted from data classified into a hierarchical structure as shown in FIG. 2, special processing must be performed.

For example, when there is data in hierarchical divisions A, B, and C, if new data is added below layer A in 5th layer, the record number for accessing the added data in other layers B and C is It becomes unknown. Therefore, in such a case, by setting an item called "unorganized" in the top layer A and storing the record number of the added data in this unorganized item, the lower layer B .

In C, you can know the record number of the added data.

On the other hand, if a lower record in a certain hierarchy is deleted,
If the data itself is deleted, it becomes impossible to access it from other hierarchies. Therefore, in such a case, only the record number assigned to that hierarchy is deleted, but the record itself is not deleted.

This allows access even from other hierarchies.

Note that if the search formula is stored in the lowest hierarchy, it becomes possible to access the database main body 1 dynamically.

[Effects of the Invention] As explained above, according to the present invention, by specifying classification items according to the purpose of use. A hierarchical classification index according to the purpose of use is automatically created and maintained, so by selecting the desired classification index from among the maintained classification indexes, the optimum grinding according to the purpose of use can be performed. It can be carried out.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing an embodiment of the present invention, Fig. 2 is an explanatory diagram showing an example of data of an agency that creates a hierarchical classification index, and Fig. 3 is a flowchart showing an algorithm for creating a classification index. , FIG. 4 is an explanatory diagram showing the hierarchical structure and storage format of the classification index, and FIG. 5 is an explanatory diagram showing an example of data stored in the database main body and its storage format. DESCRIPTION OF SYMBOLS 1... Database main body, 2... Classification index holding device, 3... Processing device, 31... Central wave peak device, 32... Classification index processing part, 33... Classification index editing part, 34 ...Database search department,
35...Database editorial department. Figure 1 轡49P Pointer Boiling Figure 2 Figure 3

Claims (1)

[Scope of Claims] Data holding means for holding a plurality of sets of data consisting of a plurality of classification items as one set of record data, and a classification index with a hierarchical structure for searching data of a desired classification item in the record data. index creating means for creating a plurality of sets of created classification indexes; index selection means for selecting one of the stored classification indexes; and index selection means for selecting one of the stored classification indexes; A database device comprising a search means for searching record data held in a data holding means.