JP2819567B2 - Classification index creation device - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a classification index creation device that creates an index for accessing a database.

2. Description of the Related Art A database is a "base" storing various information, and can be represented as a shared file in which various files are integrated so that various users and various types of businesses can effectively use the database. Prior to the advent of databases, files were proprietary for each task. For this reason, there is a problem in updating and maintaining the file because data is duplicated between the files and the content is inconsistent. The database solves this problem.Data is drawn according to the position of various tasks and processes, and data can be selectively extracted and processed and processed in accordance with the data. It is a large aggregate.

Conventionally, a thesaurus has been used for searching data from such a database. Here, the thesaurus is an index word dictionary (term list) that defines the range of keywords used in the search and the relationship between the keywords. The words or keywords used in the thesaurus are defined in advance, and in a system using a thesaurus in which the relations between synonyms, broad terms, lower terms, related words, etc. are defined between keywords, predetermined keywords are used. Is added to data to be stored and stored in a database. Alphabetic characters and numbers are used as keywords.

By the way, if a thesaurus is used as the only index, a system that is effective for a specific application but is not always easy to use for other various applications is not necessarily obtained. Therefore, in order to improve this drawback, a classification index creation device having a plurality of indexes has been proposed.

"Problem to be Solved by the Invention" In the proposed device, an index for performing hierarchical classification is created. Because of this, you need to enter a search expression each time you create an index,
It was troublesome. Due to such complexity, the creation of a hierarchical index is hindered, and its usefulness cannot be exhibited.

Accordingly, it is an object of the present invention to provide a classification index creation device that can easily create a hierarchical index.

[Means for Solving the Problems] In the present invention, a database device that holds a plurality of sets of records in which data consisting of a plurality of classification items is one set of records is provided with (a) the plurality of classification items corresponding to the plurality of classification items. Means for storing a plurality of types of classification methods that store the classification methods to be performed by these classification items, and (b) a hierarchy to be created from among the plurality of classification methods stored in the classification method storage means. (C) a classification method reading means for reading a corresponding classification method from the classification method holding means in accordance with the specification of the classification method by the classification method specifying means; A) combining means for combining the classification methods read by the classification method reading means, and (e) a hierarchical structure using the classification method combined by the combination means. And all the nodes of the same hierarchy have a classification index having a subordinate node group having the same structure as each other. (F) a classification index generated by the classification index generation means And index holding means for holding a plurality of index items for retrieving data of a desired category item from a plurality of category items in the record.

That is, according to the present invention, a search formula is created while sequentially traversing the hierarchy, and a classification index is created.

[Examples] Hereinafter, the present invention will be described in detail with reference to examples.

FIG. 1 shows a configuration of a classification index creating apparatus according to an embodiment of the present invention. This device includes a central processing unit 21. The central processing unit 21 is connected to each unit described below by a bus 22, and controls each unit in response to a request from a user.

(I) Communication control section 23: A section for controlling requests from a plurality of terminal devices in a time-division manner.

(Ii) Classification index processing unit 24: A unit that performs all accesses to a plurality of classification indexes. The classification index processing unit 24 performs the following operation.

(B) The category index is stored as a file in the auxiliary storage device.

(B) Switching from a plurality of classification indexes to a specified one.

(C) Return the classification structure of the classification index as required.

(D) The contents are changed in response to a request from the classification index creating unit 26 described below.

(E) According to the search formula stored in the classification index, a search is requested to the database processing unit 27 described later.

(Iii) Classification index creation unit 26: It has a classification table 28 and a working memory 29, and performs the following operations.

(B) Receive designation from the user for creating a category index and retain it.

(B) The contents of the classification table 28 are extracted and presented to the user. Alternatively, the contents are held in the working memory 29.

(C) According to the contents of the working memory 29, a request for creating a classification index is sent to the classification index processing unit 27.

(Iv) Database processing unit 27: (a) The database processing unit 27 searches the database stored in the database main body 31 for the value of the field and the search by the record number. Here, the record is, for example, an area for holding data on each individual when the personal information of the employee is used as a database.

(B) Addition, deletion, and editing of data in the database are performed according to a user's request.

(V) File edit processing unit 32: The file edit processing unit 32 is connected to the classification index file 33 and the classification method file 34 together with the classification table 28, and performs the following operation.

(A) The file editing unit 32, that is, the editing of a text file, is processed in response to a user request. It is a normal text editor.

(B) The classification method file 34 is used to specify whether to classify data by field for each field in the database. Here, the field refers to an area in a record that holds data of classification items such as a name and an age, using the above-described example of the employee's personal information.

(Vi) Display control circuit 35: a circuit for displaying various data on the display 36.

(Vii) Keyboard 37: An input device for performing various operations of the classification index creation device. Pointing on keyboard 37
A mouse 38 as a device is connected. The cursor displayed on the display 36 is operated using the mouse 38.

FIG. 2 is a diagram in which an example of records stored in the database in the database main body 31 shown in FIG. 1 is hierarchically classified. In this example, as shown in a table format in FIG. 3, the first tier is “year-end adjustment”, and the second tier is related to departments such as the personnel department and the general affairs department. The third hierarchy as the lowest hierarchy relates to ages such as those in their twenties and thirties. In the second hierarchy, a search formula for accessing the database body 31 is stored. Access is performed by selecting one classification index file and tracing the hierarchy in order, so that desired data or records can be reached. The classification index processing unit can hold a plurality of such hierarchies. Each of these hierarchies can be defined by the user.

FIG. 4 shows the structure of the classification table 28 shown in FIG. The classification table is a table of the names of the fields of the "age" and "annual income" windows and the file names of files specifying the classification method such as "age 1" and "age 2".

FIG. 5 shows the details of the file “age 1” shown in FIG. 4 as an example. In this example, for example, a search formula of “age ≧ 20AND age <30” is paired with the name of the category “20s”. The file specifying the classification method in this way is a text file in which pairs of classification names and data retrieval formulas included in the classification are arranged.

FIG. 6 shows the structure of the working memory 29 shown in FIG. Working memory
Reference numeral 29 denotes a table format in which a hierarchy and a classification method are stored as a pair of data. In this case, the hierarchy is lower as the number is larger in this figure, and the classification method is the same as the contents of the text file shown in FIG.

When defining a classification index, first, a classification table is displayed on a display 36 (FIG. 1) used by the user. If the classification table cannot be displayed at once, the entire screen can be viewed by scrolling the screen.

FIG. 7 shows an example of the display contents of the display. In the left half of the display 36, a classification table column 41 is displayed. In the right half of the display 36, an area 42 for displaying the contents of the file and an area 43 for displaying a specified result in the middle are arranged. The user specifies the file name of the file storing the classification method from the classification table column 41 in the order of the hierarchy to be created. Specifically, by using the mouse 38 to move the cursor to a desired file name and clicking a button of the mouse 38, the file name is specified.

Of course, the contents of the classification method file can be displayed on the display 36. When a target file name is selected from the classification table column 41 as in this operation example, the content is moved into the working memory 29. This completes the designation work for creating the classification index.

When the above operation is completed, the data is sequentially traced from the working memory 29 and written into the classification index file 33. That is, a request for writing is made to the classification index processing unit 24.

FIG. 8 is a diagram for explaining the operation for creating the series of search formulas shown in FIG.

First, three memory areas “N”, “A”, and “B” are prepared on the working memory 29, a numerical value “0” is written in the memory area “N”, and the other memory areas “A”, “B” are written. Write an empty string in "" (step). Here, “A” is the name of the category to be searched, and “B” is a search formula corresponding to this category.

After the initial setting of the three memory areas “N”, “A”, and “B” is completed, the search formula is created by tracing to the respective lowest layers shown in FIG. So, first, the numerical value “N”
It is determined whether or not a larger value exists as a value indicating the hierarchy in the working memory 29 (step). In this case, since N = 1, the numerical value "N" is set to the numerical value "1" (step), and the classification method for the numerical value "1" is extracted from the working memory 29 (step).
Then, the following replacement is performed for each of the classification name and the search formula in the classification method (step).

A ← A + (classification name) B ← B + (AND) + (search formula) In this case, since an empty character string has been written in both “A” and “B”, the classification of the first hierarchy The name is written to the memory area “A” as it is, and the search formula of the first hierarchy is written to the memory area “B” as it is.

When the above operation is completed, the self is recursively called (step), and the process returns to the step. That is, it is determined whether or not a value larger than the numerical value “N” exists as a value indicating the hierarchy in the working memory 29. At this time, if the second hierarchy exists (step; Y), N = 2
And one layer deeper. Explaining with reference to FIG. 2, when creating a search expression that is “HR department” and “20s”, the user reaches the hierarchy of “20s” with N = 2. In this state, the classification method corresponding to N = 2 is extracted from the working memory 29 (step). Then, a replacement process is performed in steps for each of the classification name and the search formula therein. In the case of this example, this would be:

HR department ← HR department + (classification name) = HR department + 20s 'affiliation = HR department ←' affiliation = HR department + (AND) + (search formula) = 'affiliation = HR department' AND (age ≥ 20AND Age <30) When the above processing is completed, when returning to the step, it is determined whether or not the third hierarchy exists. Second
In the case of this example shown in the figure, since there is no file (N), the final pair of “A” and “B” is stored in the classification index file 25 (step), and this “HR section” is “20”. Teens ”
The creation of the search expression that is is terminated. Then, in this body, the recursion level is returned to the “HR department” (step), and the process proceeds to the next “HR department”, where “30”
The creation of a search expression that is “alternative” is started. When the creation of this search expression is completed, the search expression is written to the classification index file 25 (step). Are created and written into the classification index file 25.

When the creation of each search expression in which the first hierarchy is "HR department" is completed, the level of recursion is returned, and this time the "General affairs department"
To the bottom layer in order, “20s”, “30”
Each of the search formulas for "" is generated. When this is completed, a search formula for "Finance Department" and the like is automatically generated.

In this way, when the classification of each classification method is completed,
Returning to the next higher level, the processing is performed recursively. In the search formula, the logical product of the search formulas of the nodes of each layer is generated in the process of tracing the hierarchy from the highest layer to the lowest layer, and when these arrive at the lowest layer, these are written to the classification index file 25. Will be.

When a category method is added, the following is performed.

(1) Create a classification method file 34 as a text file.

(2) Register the file name in the category table. By doing so, it can be used for creating a classification index.

[Effects of the Invention] As described above, according to the present invention, various classification indexes can be defined without having to type their search formulas one by one, and even a beginner can easily create them. Further, since the classification method once created is reused, there is an effect that waste of labor can be eliminated. Furthermore, since all the nodes of the same hierarchy create a hierarchical classification index having a group of lower nodes having the same structure, the hierarchical structure of the classification is defined simply by specifying the classification method for each hierarchy. There is an effect that can be.

[Brief description of the drawings]

1 to 8 are for explaining an embodiment of the present invention, in which FIG. 1 is a block diagram showing the outline of the structure of a classification index creating apparatus, and FIG. 2 is stored in a database. Explanatory diagram classifying examples of records hierarchically,
FIG. 3 is an explanatory diagram showing the relationship between each hierarchy and their retrieval formulas, FIG. 4 is an explanatory diagram showing the structure of the classification table shown in FIG. 1, and FIG. 5 is a diagram of a file named 'age 1'. FIG. 6 is an explanatory view showing details of the working memory shown in FIG. 1, FIG. 7 is a plan view showing an example of display contents of a display, and FIG. 8 is FIG. 5 is a flowchart showing the operation for creating the series of search formulas shown in FIG. 15: Classification index creation unit, 24: Classification index processing unit, 26: Classification index creation unit, 27: Database processing unit, 28: Classification table, 29: Working memory, 32: File editing processing unit , 33 ... Classification index file, 34 ... Classification method file.

Continuation of the front page (56) References Ueno, Hiyoshi, Nagai, "Summary Database-A Consideration on Query Languages-" Proc. Of the 30th Annual Convention of IPSJ, pp. 1757-1758. (Showa 60-3-13) Chimori and Ueno, "Structural Editor of Summary Information Service System SIMPLEX", Proc. Of the 35th IPSJ Annual Conference (late 1987), p. 309-310 (Showa 62-9-28) (58) Fields investigated (Int. Cl. ⁶ , DB name) G06F 17/30 JICST scientific and technical literature file

Claims (1)

(57) [Claims] 1. A database device for holding a plurality of sets of records in which data consisting of a plurality of classification items is a set of records, wherein a classification of these classification items in correspondence with the plurality of classification items is provided. Means for storing a plurality of types of classification methods storing classification methods for performing the method, and a classification method specifying means for specifying a classification method corresponding to a hierarchy to be created from among the plurality of classification methods stored in the classification method storage means Classification method reading means for reading a corresponding classification method from the classification method holding means in accordance with the specification of the classification method by the classification method designating means; Combination means for combining the classification methods read by the classification method reading means; Using the classification method combined by this combination means, all nodes having a hierarchical structure and having the same hierarchy are identical to each other. Means for creating a classification index having a group of nodes at the lower level of the structure, and searching the classification index created by the classification index creation means for data of a desired classification item from a plurality of classification items in the record. A classification index creation device, comprising: