JPH04124774A - Data storage method for hierarchical construction in related data base - Google Patents

Abstract

PURPOSE:To enable the high-speed retrieval of information with meaningly hierarchical construction by expressing the position of information to be recorded in the hierarchical construction for the retrieval with the kind of characters and the length of a character string. CONSTITUTION:A keyword is inputted, and a document ID with the keyword included in conception lower than a keyword 'digital computer' is retrieved. In this case, the keyword which belongs to the low rank of the keyword 'digital computer' becomes the one adding the character string in addition to the classification of the keyword. Therefore, forward coincidence retrieval on keyword classification is performed when the document ID with the keyword which belongs to the low rank of the keyword 'digital computer' is retrieved. Next, it is ended by performing the display of the result of retrieval. In this case, the necessary retrieval number is only two times. Thus, the information for retrieval can be rapidly retrieved when all the information which belongs to the lower rank corresponding to a certain paragraph.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a method for storing data in a hierarchical structure in a relational database, and in particular, the present invention relates to a method for storing data in a hierarchical structure in a relational database. This invention relates to a technique that is effective when applied to increase search speed when recording.

[Prior art]

Conventionally, hierarchical databases and dual-type databases have been used to store information in which search information has a hierarchical structure. Due to their structure, these databases are suitable for storing information that has a semantically hierarchical structure, and can quickly search for necessary information.

In addition, when storing such information using a relational database, for example, as described in C. J. Dito [Introduction to Database Systems J (Maru IP) PPI 91-194] The hierarchical structure was recorded by expressing information corresponding to a clause in a group and information corresponding to a clause above or below it as one set.

[Problem to be solved by the invention]

However, when using a hierarchical database or a dual-type database, for example, rComputer T.

day” (Science Publishing) 1989/3 (No, 30)
pp85-87 and [Nikkei Computer J (Nikkei BP)
1989.11.20 (No, 213) PP88-10
As described in Section 8, there are problems in that it takes time to create application software and a large amount of effort is required to maintain and manage the database.

In addition, when using a relational database, the above problem does not occur, but if information corresponding to a certain node in a hierarchical structure and information corresponding to a node above or below it are expressed as a set, then the information below a certain node in the hierarchical structure There has been a problem in that the search speed when searching for all the information that belongs to it becomes extremely slow.

For example, if the keywords for searching are the information shown in Figure 5 and stored in a relational database using the conventional method,
It is expressed as 68th and 6b, and the search is performed as follows.

6a and 6b are intended for a search system for one document, and it is assumed that each document is assigned a plurality of keywords.

Figure 6a shows the semantic relationship of these keywords, and the keywords "digital computer" and "analog computer" are considered to be conceptually included in the keyword "computer", and the keywords "computer" and "analog computer" are considered to be conceptually included in the keyword "computer". ” is ranked lower than that. In this way, the tree-like structure shown in FIG. 5 is constructed.

It is assumed that a document ID having a keyword belonging to a concept lower than the keyword "digital computer" is searched. It is assumed that the document ID is a number uniquely assigned to each document.

Such searches are often used because they can be performed with a single command without knowing the subconcepts of a certain concept.

In addition to literature search databases such as this example, in employee databases, for example, position information has a tree-like structure such as president - managing director, manager, section manager, etc., so in such a database, you would like to know all the subordinates of a certain department manager. The scope of application of this type of search is wide, as it can also be applied to searches such as .

In this example, conventional methods using relational databases store information using the charts shown in Figures 6a and 6b.

When the charts shown in FIGS. 6a and 6b are used, a keyword "digital computer" is searched for a narrow term in FIG. 6a. In this case, the hyponym is “computer”.
and “minicomputer” is searched.

Next, in FIG. 6b, the previously searched keyword “
Search for the set of required material IDs by performing a search using "PC" and performing a logical OR between the set of material IDs obtained through the search and the set of material IDs obtained by searching with the keyword "minicomputer" In this method,
The greater the number of keywords that are subordinate to the specified keyword, the longer the search will take. The reason is that Section 6b
The reason why the table in the figure is searched many times is because it is necessary to perform a logical OR on all the sets obtained by each search and to remove duplicate document IDs.

The present invention has been made to solve the above-mentioned problems, and an object of the present invention is to use a relational database that does not require much time to create application software and does not require effort to maintain and manage the database. It is an object of the present invention to provide a method for storing hierarchically structured data in a relational database that enables high-speed retrieval of information in which search information has a semantically hierarchical structure.

The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings.

[Means to solve the problem]

In order to achieve the above object, the method for storing data in a hierarchical structure in a relational database of the present invention stores information in which search information accompanying information to be recorded forms a hierarchical structure.
When recording using a relational database, the most important feature is that the position of the information to be recorded in the hierarchical structure of the search database is expressed by the type of character and the length of the character string.

[For production]

According to the above-mentioned means, when information in which the information for searching associated with the information to be recorded forms a hierarchical structure is recorded using a relational database, it is difficult to determine where the information to be recorded is positioned in the hierarchical structure of the search database. Since the information is expressed by the type of character and the length of the character string, it is possible to quickly search semantically hierarchical information without the time required to create application software and the effort required to maintain and manage the database. can be done.

[Embodiments of the invention]

Hereinafter, one embodiment of the present invention will be specifically described using the drawings.

Figure 1], Figure 1A, and Figure 1B are diagrams for explaining the relational database of the embodiment used in the hierarchical data storage method in the relational database of the present invention.
Figure a is a thesaurus classification table, and Figure 15 is a thesaurus classification comparison table.

Keyword classifications in the thesaurus classification table and thesaurus classification comparison table of this embodiment are expressed by types of characters (including numbers) and lengths of character strings. In the example of FIGS. 1a and 1b, it is expressed by a string of numbers.

FIG. 2 is a diagram showing a tree-structured database according to an embodiment of the present invention, and FIG. 3 is a block diagram showing a schematic configuration of a system that implements the hierarchical data storage method in a relational database of the present invention. be.

FIG. 4 is a flowchart of a method for storing data in a hierarchical structure in a relational database according to an embodiment of the present invention.

As shown in FIG. 3, the system of this embodiment includes an input means 1, an arithmetic processing unit (comprising a CPU and a memory) 2 for operating the relational database management system, and an information processing unit (comprising a CPU and a memory) for inputting information having a dendritic structure. It consists of a storage device 3 for filing and an output means 4.

In the method for storing data in a hierarchical structure in the relational database of this embodiment, as shown in FIG. 4, keywords are input in step 101. Next, similar to the conventional technique described above, a document ID having a keyword included in a lower level concept than the keyword ``digital computer'' is searched. (step 102).In this case, the keyword classification is “1”.
-1' is searched. At this time, as shown in Figure 1a,
Keywords belonging to the lower order of the keyword "digital computer" are obtained by adding a character string to the keyword classification "1-1", such as "1-1-2", for example. Therefore, in order to search for a document ID with a keyword belonging to the keyword "digital computer", in Figure 1-b, keyword classification II 1 11
It is sufficient to perform a prefix match search for 1 (103). Note that the prefix search function is always included in general relational database management systems.

Next, the search results are displayed (step 104) and the process ends.

According to the method of this embodiment, no matter how many keywords exist below the specified keyword, the required number of searches is 1a.
1b, ie, only twice. This makes it possible to perform a high-speed search when searching for all information belonging to information below the information corresponding to a certain section.

For example, when search information (keywords, etc.) specifically indicating a certain section is designated, first, location information regarding that section is searched based on that information.

The location information for this section consists of several characters. The length of the character string or the number of punctuation marks such as hyphens in the character string indicates the depth of this node in the hierarchical structure, and the type of character indicates the horizontal position within a cylinder of the same depth. For example, '1-2' indicates a node that exists at a position below the second line from the left of a node that is below the leftmost line from the top (root) in the hierarchical structure. The desired information can be obtained by performing a prefix match search on a table containing this position information and the information to be recorded, using the position information as a key.

The method of the present invention is particularly effective as the number of nodes to which a certain node belongs increases.

The present invention has been specifically described above based on examples, but 1.
It goes without saying that the present invention is not limited to the embodiments described above, and can be modified in various ways without departing from the spirit thereof.

〔Effect of the invention〕

As described above, according to the present invention, it is possible to search for semantically hierarchical information at high speed.

[Brief explanation of drawings]

Figures 1a and 1b are diagrams for explaining a relational database of an embodiment used in the hierarchical data storage method in a relational database of the present invention, and Figure 2 is a tree diagram of an embodiment of the present invention. A diagram showing the semantic structure of information in a structured database. FIG. 3 is a block diagram showing a schematic configuration of a system that implements a method for storing hierarchical data in a relational database according to the present invention, and FIG. 4 shows a method for storing hierarchical data in a relational database according to an embodiment of the present invention. Flowchart of FIG. 5 is a diagram showing the semantic structure of the illustrated information, and FIG.
Figures 6b and 6b are diagrams expressed in a conventional manner using a relational database. In the figure, 1... input means, 2... arithmetic processing unit, 3... storage device 3. 4... Output means.

Claims (1)

[Claims] (1) When using a relational database to record information in which the search information that accompanies the information you want to record forms a hierarchical structure, you can specify in writing where the information you want to record is positioned in the search hierarchy. A method for storing data in a hierarchical structure in a relational database, which is characterized by representing the type of data and the length of a character string.