JP2000200206A - Device and method for data management and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make easily performable design, construction, and alter of a data base. SOLUTION: The data structure of the data base is defined only with items and their contents. The same management number, i.e., a record number 1 in this case is assigned to a number '00', a name 'Yamada', an age '30', and height '165' and the same record number, i.e., a record number 2 is assigned for a next number '01', a name 'Satoh', an age '25', and height '160'. Items are retrieved with a retrieval key 1 and contents are retrieved with a retrieval key 2 respectively to obtain sets of management numbers. Data whose contents of the items specified with the retrieval key 1 match the retrieval key 2 can be retrieved from a set of the management numbers obtained by ANDing the respective management numbers.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data management device, a data management method, and a recording medium, and more particularly, to a database for an unspecified data item efficiently by eliminating a restriction on a data structure. The present invention relates to a data management device, a data management method, and a recording medium that can perform the operations.

[0002]

2. Description of the Related Art A conventional database management system (DB)
MS (database management sy)
) has execution functions such as a database definition language, a database operation language, and an inquiry language as basic functions for accessing data and schema between a user and an application program.

[0003] The database definition language is used to define the structure of data to be stored. For example, in a relational database, items such as numbers, names, genders, ages, and heights (see FIG. 7) are used. field)
Is defined as follows, for example.

Create table person (number integer, {number: integer type} name char (10), {name: character type (10)} sex char (2), {sex: character type (2)} age integer, { Age: integer type} height float {height: decimal type});

[0005] The database operation language is used to register, delete, and register data based on the data structure defined as described above.
And addition, or combining and dividing databases.

[0006] The query language is an interface language between the database and the user, and is used when the user searches for desired information from the database.

As described above, in the conventional database management system, it is necessary to previously set the framework (schema) of the data to be stored by designing the database.

[0008]

As described above, most of the existing database management systems perform operations such as registration, deletion, and retrieval of information based on a data structure defined in a database definition language. Therefore, when a change occurs in the data structure, it is necessary to change the application program developed based on the data structure before the change, which has been a troublesome problem.

In the past, when defining a data structure, there are restrictions on the storage capacity and processing capacity of the storage device of the database. Had to be written explicitly. However, nowadays, large-capacity storage devices are provided at low cost, and the processing capacity of computers has been dramatically improved. Physical limitations such as storage capacity and processing capacity have little effect on database specifications. There is no situation.

The present invention has been made in view of such circumstances, and has been made in view of the above circumstances. Thus, it is possible to eliminate the restriction on the data structure and efficiently design and build a database without complicated database design. Is what you do.

[0011]

According to a first aspect of the present invention, there is provided a data management apparatus comprising at least a first item data and a second item data, wherein the first item data and the second item data. Means for inputting second information composed of one or more pieces of first information whose data can be identified by a tag, and first storage means for storing the second information input by the input means And second storage means for storing third information obtained by removing the tag from the second information; and at least third information for the first item with respect to the third information stored in the second storage means. 1 search key and 2nd item for 2nd item
A search unit that performs a search based on the search key, and an output unit that outputs second information having the first information searched by the search unit as a component.
Further, it is possible to further provide a conversion means for converting the document data into tagged second information. Also,
The data of the first item indicates the type of the first information,
Can represent the contents of the first information. The data management method according to claim 4 includes at least data of a first item and data of a second item, and the data of the first item and the data of the second item can be identified by a tag. Inputting second information composed of one or a plurality of first information, storing the input second information, storing third information obtained by removing a tag from the second information; At least for the third information
A search is performed based on a first search key for a first item and a second search key for a second item, and second information having the searched first information as a component is output. And A recording medium according to a fifth aspect is characterized by recording a program capable of executing the data management method according to the fourth aspect. In the data management device, the data management method, and the recording medium according to the present invention,
At least, data of the first item and data of the second item are included, and the data of the first item and the data of the second item are composed of one or a plurality of pieces of first information that can be identified by a tag. Input second information, store the input second information, store third information obtained by removing a tag from the second information, and at least store the third information with respect to the stored third information. A search is performed based on a first search key for the first item and a second search key for the second item, and second information having the searched first information as a component is output.

[0012]

FIG. 1 is a block diagram showing a configuration example of an embodiment of a database management apparatus according to the present invention. As shown in FIG. 1, in the present embodiment, a document management server (WS (work station)) 1 for managing a database and a document management client (PC (personal computer)) including applications for accessing the document management server (WS (work station)) ) 10.

The document management server 1 has a document management database 2, a chart management database 3, and a full-text database 4. The document management database 2 and the chart management database 3 are for performing file management, and manage index information and the like using a relational database.

In the document management database 2, original documents (for example, various word processing documents) and structured documents (for example, SGML (Standard Genera) are stored.
There are two types of management formats based on a tagged document (compliant with a sized markup language), and the latter incorporates link information to a chart data file stored in the chart management database 3.

On the other hand, in the full-text database 4, all text information (excluding tags) of the structured document is registered and managed so that a free word search can be performed for all the text in the document.

A document management client 10 running on a PC uses a registration application 11 for registering data in each database on the document management server 1 and a search result searched by a search application 13 described later. The system includes an editing application 12 for creating a document, and a search application 13 for searching data in the document management database 2, the diagram management database 3, and the full-text database 4.

Each of the registration and search applications 11 and 13 is provided with a client corresponding to each database to be accessed, so that an application can be constructed for each database. ing.

Here, the original document file stored in the document management database 2 is a document file that can be reused (edited and printed) by using a corresponding application. In addition, for word processing documents that also have chart data files,
It can be combined into a single file to create an original document. In the document management database 2, index information such as a title and an author name, and storage information of an original document file are registered together with a management number automatically given by the system.
However, the entity of the original document file is distributed and stored on the WS defined for each user.

A chart data file is an image data file such as a figure or a photograph, or a Lotus (L
otus) (Lotus is Lotus Develo)
pment Corporation), Excel (Excel is a US Micro
(Trademark of Soft Corporation in the United States and other countries) is registered in a database as individual data independent of the original document file. When registering chart data, it is necessary to add index information to chart data as in the case of registering document data.

The structured document is an original document or O
It is created by extracting a text portion based on a print document read by a CR or the like and then performing an editing process such as adding tag information via a sentence structure analysis program. The created structured document is registered in the document management database 2 alone or together with the original document after the index information is added. Further, the text data of the registered structured document is automatically registered in the full-text database 4.

The management numbers assigned to the registered documents include a common document number for managing the same document having different formats (original document, structured document, full-text database document) and an individual document number unique to each document. In the document management database 2, a common document number is searched using each index information as a key. The original document and the structured document can be downloaded using the common document number as a key.

The chart management database 3 retrieves a graphic data number using index information as a key, and can download a chart data file based on the figure data number. Further, the chart data can be similarly searched using the link information to the chart data of the structured document searched in the document management database 2.

In the search of the registered document stored in the full-text database 4, a character string input by the user is searched from the registered document, and the common document number or the text of the context (or paragraph) is transmitted to the search client. Can be performed.

FIG. 2 shows an example of a data structure of data handled by the database management device of the present invention. As shown in the figure, data always has a format of "item" and "contents". In the case of this example, data 1 is the item “number”
Is “00”, and the content of the item “name” is “Yamada”. The content of the item “age” is “30”, and the content of the item “height” is “165”. In the data 2, the content of the item "number" is "01", the content of the item "name" is "Sato", the content of the item "age" is "25", and the content of the item "height" is "160".

This data structure corresponds to a conventional RDB (rel
It is the same in the case of the national database. For example, the content of the “name” item of data 1 is “Yamada”. Data 1 for RDB (number is 00, name is Yamada, gender is male, age is 30, height is 16
5) are the items “number” to “height” in the present embodiment.
Corresponds to.

As described above, in the present embodiment, the data structure is defined by the following format, that is, only items and contents. (Field 1 item character string), (field 2 item character string / binary), ..., (field N
Item string).

Since data is composed of only items and their contents, when creating data, it is possible to create tagged text information (structured document) as described below without regard to the definition of the database. Good.

[0028]

Here, the character string "<data>" is a tag indicating the start of data. Character string "<item>"
Is a tag indicating that the following character string is a character string representing the type of item. Character string "<content>"
Is a tag indicating that the following character string is a character string representing the content of the immediately preceding item. In addition, the character string “</ item
m>"is a tag indicating the end of a character string representing the type of item. The character string “</ content>” is a tag indicating the end of the character string representing the content of the item. The string "<
/ Data>"is a tag indicating the end of data.

The created tagged text information (structured document) is divided into items and their contents by a structured document interpretation program, as shown in FIG. 3, and then management information (identification information and the like) and full text Stored separately in DB (database). That is, the items and their contents are stored in the full-text database 4, and management information such as a record number corresponding to each item is stored in a storage unit (not shown).

When retrieving data stored in this manner, the user first designates "OO" as an item and "XX" as a content as a retrieval key. The search client of the search application 13 uses the specified search keys to search the full-text database 4 for relevant data. The search method can be an exact match or a partial match. Then, management information corresponding to the searched data, in this case, a set of record numbers is obtained.

For example, assuming that the search key “OO” for an item is “age” and the search key for content is “30”, a set of record numbers {1, 25, 44, 75, 89, 111,...
Set of record numbers of data whose content is "30" {1, 33, 44, 57, 76, 89, 93, ...}
Is obtained.

Here, by performing a logical product (AND) operation on each record number searched by each search key, data having an item of “age” and a content of “30”, that is, Set of record numbers $ 1,44,89,
...｝ can be obtained. In this manner, data whose age is 30 can be searched.

As described above, in the case of the present embodiment, since the full-text database 4 is searched, high-speed processing is possible. If the same processing is performed on a relational database in which data composed of two fields of items and their contents are registered, it takes a considerable amount of time to search a large amount of data. Generally, a full-text database system is provided with a mechanism for performing pre-processing called learning (processing for converting document data into a unique data structure and storing it) and performing high-speed search by an application called a search engine. In the present embodiment, this is used. A detailed description of the learning method and the search engine is omitted here.

Next, taking an employee information management database as an example,
This embodiment will be described in more detail. For example,
In a certain company, a personal information card as shown in FIG.
Is created using word processing software such as Microsoft Corporation in the United States and other countries), and a management database of employee information is created based on this. And this card,
Completed by all employees and updated annually in April.

First, in order to register data, a structured document (here, an SGML document) is created. Word document file (Microsoft) submitted by each employee
Document file created using Word), for example,
Using the employee number, the file name is 7198235. D
Tag information is added based on the file such as OC, and a file having a file name “7198235.SGM” is created.

The management information varies depending on the assumed application. In the case of the above example, the information to be given at the time of registration may be the registration date and the employee number or name of the recipient (manager). However, when the information for managing the database is added, the following management information is required.

Management number (automatically assigned by the system at the time of registration, corresponding to the record number) Registration date Recipient (administrator's employee number) or name Original file name (for example, 7198235.D)
OC) • SGML file name (for example, 7198235.SG
M)-Access authority of each file-Data deletion flag

The management information of the system can be as follows.

Directory for storing original files (including machine (WS, PC, etc.) names, etc.) Directory for storing SGML files (including machine names, etc.)

Next, the structure of the management information will be described. The management information includes system management information and database management information.

The system management information is information indicating the storage location of the file. The management method may be in any format. For example, the system management information may be a management file (SYSTEM.IN).
There is a method of creating I) and managing it as a global variable as follows.

＄ ORG_FILE = WS1: / home / ORG ＄ SGM_FILE = WS2: / home / SGML

The management information of the database is managed in a table format as shown in FIG. 5, as in the case of a normal database. That is, management information is managed using a table including items such as a management number, a registration date, an administrator, an original file name, an SGML file name, authority, and whether or not there is a deletion. For the management of the management information, a conventional database may be used, or the management information may be read from a management file into the server program of the document management server 1 and managed by the server program.

In preparing the full-text database 4,
Using the management number and the SGML file, index information as shown in FIG. 6 is internally constructed. That is, the items (staff number, date and time, name,
Index information composed of phonetic characters, affiliations, etc.) and their contents is created, and this is used as the full-text database 4.

As can be seen from FIGS. 5 and 6, the management information and the full-text database 4 are linked by a management number. That is, in this example, the link information is a management number. Therefore, the full-text database 4 is searched for by item and content, and from the set of obtained management numbers,
The target data can be searched by the D processing or the like. In addition, the original file name or the like corresponding to the searched management number can be known from the management information, so that the original file can be displayed.

As described above, in the above embodiment, since the full-text database is used for the search, the design of the database can be facilitated. Further, in a conventional database management system (DBMS), it is necessary to modify a program in accordance with a change in the database, but in the case of the present embodiment, the data structure itself is not changed (always, Therefore, a new item can be freely added to the database or an item can be freely deleted.

It should be noted that the <
data>, </ data>, <item>, </ it
The tags such as <em>, <content>, and </ content> are examples, and the present invention is not limited thereto.

In the above embodiment, the document management server is constituted by the workstation (WS). However, the present invention is not limited to this, and the document management server may be constituted by a personal computer (PC). Good.

In the above embodiment, the structured document described in SGML is stored in the document management database. However, SGML is an example.
HTML (Hyper Text Markup La)
nstructure), and other structured documents.

In the above embodiment, the case where the present invention is applied to a client / server type system has been described. However, the present invention is not limited to this, and may be applied to a stand-alone type system.

Further, each application software in the server program and the client in the above embodiment is a compact disc (CD-ROM).
read only memory), DVD (dig)
ital video disc), FD (flopp
y disc) or other recording medium.

[0053]

As described above, according to the data management apparatus, the data management method, and the recording medium of the present invention, at least the data of the first item and the data of the second item are provided. Inputting second information composed of one or more pieces of first information in which the data of the item and the data of the second item are identifiable by a tag, storing the input second information, The third information obtained by removing the tag from the information of is stored, and at least the stored third information is
A search is performed based on a first search key for a first item and a second search key for a second item, and second information having the searched first information as a component is output. Therefore, it is possible to efficiently design and construct a database without restricting the data structure and without complicated database design. Therefore, it is possible to easily add or delete items from the database. Also,
Since there is no restriction on the data structure, a database for unspecified data items can be efficiently created.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration example of a data management device according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of a data structure used in the data management device of the present invention.

FIG. 3 is a diagram showing a procedure for storing text information separately in management information and a full-text database.

FIG. 4 is a diagram showing an example of a personal information card.

FIG. 5 is a diagram showing a table for managing management information of a database.

FIG. 6 is a diagram illustrating an example of a full-text database.

FIG. 7 is a diagram showing an example of a data structure of a conventional relational database.

[Explanation of symbols]

 1 Document Management Server 2 Document Management Database 3 Chart Management Database 4 Full Text Database 11 Registration Application 12 Editing Application 13 Search Application

Claims (5)

[Claims] At least a first item data and a second item data
, The data of the first item and the data of the second item can be identified by a tag.
Or input means for inputting second information composed of a plurality of first information; first storage means for storing the second information input by the input means; and A second storage unit for storing third information excluding the tag; and a first search for at least the first item with respect to the third information stored in the second storage unit. Search means for performing a search based on a key and a second search key for the second item; and output means for outputting the second information having the first information searched by the search means as a component. A data management device comprising: 2. The data management apparatus according to claim 1, further comprising a conversion unit configured to convert document data into the tagged second information. 3. The data of the first item indicates a type of the first information, and the data of the second item indicates a content of the first information. A data management device according to claim 1. 4. At least the data of the first item and the second item
, The data of the first item and the data of the second item can be identified by a tag.
Or inputting second information composed of a plurality of first information; storing the input second information; storing third information obtained by removing the tag from the second information; The stored third information is searched based on at least a first search key for the first item and a second search key for the second item. And outputting the second information having the above information as a constituent element. 5. A recording medium on which a program capable of executing the data management method according to claim 4 is recorded.