JP2001067251A - Electronic data integration data management system and data base system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a new system capable of realizing the integral management of data, and to provide a data base system capable of realizing mutual operation between different systems. SOLUTION: The electronic data integration data management system for integrating and managing various data is provided with a means 1 for generating association information with source data for generating information for executing association with original data, a cognitive data extracting means 2 for extracting information necessary for the cognition of the data, and for generating it as cognitive data, a structural data analyzing and subdividing means 3 for analyzing the data contents and for generating structural data in a state that the data are subdivided, an attribute data analyzing and specifying means 4 for analyzing data contents and for generating attribute data in a state that the data are normalized, a data adding means 5 for adding one or plural cognitive data, structural data and attribute data, an object generating means 6 for generating an object by arranging the generated cognitive data and structural data as layers, an object storing means 7 for storing the objects, and an attribute data adding means 8 for adding attributes to the objects.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention belongs to the technical field related to integration of electronic data. In particular, it relates to integration of a data operation environment by integrating different types of data. More specifically, the present invention relates to a database system that flexibly supports different types of data.

[0002]

2. Description of the Related Art There are various systems for the purpose of uniformly managing electronic data. Generally, a method of realizing a unified operation of data by converting a data format into a unified format defined by the system is used. In other words, this is a method of realizing data unification by converting into a structure of the same value (or an approximate value) according to the standard of the data structure of different data.

For example, there is a type for unifying data in a format developed from a printing system. In this type, any format that can be printed is imported to make it a unified format.

[0004] A database system is an important factor in performing unified management of data. Database systems are built to enable shared use of data, centralized management of data, and independence of data. As a main data model, a hierarchical type, a network type, a relational type, an object-oriented type, and the like have been proposed.

[0005] In the hierarchical or network type data model, the relationship between records is described by a method depending on the physical system, so that there is a problem that it is difficult to maintain data independence. The relational data model, on the other hand, is the most widely used because it has the advantages that data operations can be written without considering the physical structure, application programs can be easily changed, and the data structure is easy to understand. ing.

In a relational database, all data is represented as a data model by a relation which is a two-dimensional table format. A relation is composed of a plurality of attributes, and a set of individual values belonging to each relation is called a tuple.

In a normal relational data model,
There is a restriction that only one value in a relational field can be entered. Such a relation is also called a normal relation. A relation in which a plurality of values are associated with one column is called a non-normal type, and is rarely used because the arithmetic processing becomes complicated.

In the normal type relational data model, set operations such as sum, common, difference, direct product, projection, selection, combination, and quotient can be performed as data operations, but the operation time is inferior to the hierarchical type and network type. It is said that.

As the application of databases has shifted from simple business processing to complex information processing, the shortcomings of the relational data model have become apparent. For example, in the case of a complex data structure, it is difficult to intuitively grasp the data structure, the number of join operations increases, the processing efficiency deteriorates, the meaning of the original entity becomes unclear, the query becomes complicated, etc. Problems are raised.

[0010] The cause of this problem is that the object is not modeled as it is, but is broken down into relational fragments. When a relation becomes tens to hundreds, it becomes impossible to memorize the relation. Therefore, the basic idea of an object-oriented model is to make a model correspond to an object as it is.

[0011]

Generally, electronic data is composed of three elements: a "structure" defined by the application that created the data, an "attribute" representing the characteristics of the value, and the data itself. Is done. FIG.
Shows a general method of data conversion in the related art. A, B, C, and Z represent the format of the data structure.
α, β, γ, and δ represent the contents of the attribute.

In such a conventional method, an unknown data structure such as C cannot be handled in principle because it is necessary to convert the data into a structure having the same value (or an approximate value) as the structure of the original data. If the data structure before the conversion and the data structure after the conversion are significantly different, the conversion is practically impossible. It is not applicable to systems where the original data structure is frequently changed.

The unification of the data format in the prior art involves the following problems. It is possible to unify only data (known data) that complies with rules pre-defined by the system. Since the data structure is converted according to the rules defined by the system, a part of the information held by the original data is lost. Unless the same data structure is adopted in all operation environments, data integration in the operation environment cannot be realized.

For this reason, in the conventional data management technology, the system operates data with respect to data for which the attribute is not specified in advance, and in an operating environment in which a plurality of data structures are mixed, the data operation is not performed. It is difficult to operate data.

When trying to perform unified management of data,
There are also problems with database systems. In a conventional database system, its structure must first be defined in order to integrally manage data. However, in the data structure of the core system on the host constructed by the old design method, sufficient normalization is not performed, and there may be data redundancy or lack due to inconsistency between data. Originally, these defects should be corrected during the system renewal period, but in many cases, the costs and risks involved in changing the data structure have been taken into account, and they have been taken over as they are.

In a conventional database system,
Since the structural definition was a pre-determined matter, the original importance of the data was neglected, and interoperability between different systems was not established only by the difference in the structural definition, that is, the difference in the accompanying attributes. .

An object of the present invention is to provide a new system for unified management of data which solves the above-mentioned disadvantages.
Interoperability between different systems (integer
The aim is to obtain a database system with roperability).

[0018]

According to the present invention, there is provided an electronic data integrated data management system for integrating and managing various data in order to solve the above-mentioned problems. Generating information (data location, source application, etc.), means for generating information associated with source data, (2) Extracting information necessary to recognize data and generating cognitive data, extracting cognitive data Means, (3) Analyze data contents and generate structural data as it is in a subdivided state, Analyze and subdivide means of structural data, (4) Analyze data contents and attribute data in a standardized state (5) One or more cognitive data, structural data,
Add attribute data, data addition means, (6) arrange generated cognitive data, structural data as a layer,
Object generation means, object conversion means, (7) storing the object, object storage means,
(8) An electronic data integrated data management system comprising: an attribute data adding unit for adding an attribute to the object.

In the system of the present invention, the original data is managed as a plurality of elements of cognitive data, structural data, and attribute data, and the data is managed as embodied objects.

For a user who operates data, data for expressing the contents of the data is defined as cognitive data. Holds information necessary for generating cognitive data, for example, information on the program name required for reproducing cognitive data, without preliminarily defining an electronic data format for physically expressing the cognitive data. However, only the operation method for this information is specified.

One or more different types of recognition data can be set for the same source data, and a means is provided for adding and changing recognition data even after an object is generated from source data held by an object. Provided is a means for selecting which cognitive data to use from among a plurality of cognitive data held by an object in a data operation environment.

Among the constituent elements included in the data, data decomposed into the minimum unit having meaning as data is defined as structural data. The data is decomposed as structural data in accordance with a predetermined rule. Structural data that does not match any of the rules is classified as structural data having a special state such as "unknown".

Rules for decomposing structural data are classified into rules to be incorporated into the system and rules created as programs outside the system. The rules for decomposing structural data include a data pattern for specifying data to be decomposed when decomposing data, and a program for extracting structural data from data to be decomposed.
It consists of information necessary for reproducing the structure data, for example, a program name. The entity of the structure data is not specified as long as it complies with the rules.

Among the structural elements included in the data, for the structural data which is the minimum unit having a meaning as the data, additional information of the structural data relating to the name, position, state, expression method and the like is defined as attribute data. A means for adding, changing, and deleting attribute data to structural data and objects is provided. The attribute data is composed of an identifier indicating the type of the attribute data and an entity that is the content of the attribute data itself. This provides a means for setting one or more pieces of attribute data for one piece of structural data. The entity of the attribute data is a hexadecimal AS in 4-bit units
Defined as CII text.

The database system used in the present invention is a database system composed of a set of entity objects in which a plurality of attributes are attached to a single entity data. The database system used in the system of the present invention is a relational attribute collection / class, an item attribute collection / class, an edit of which an instance of a data management class that manages the data body can have a plurality of relational attributes, item attributes, edit attributes, and the like. By defining attribute collections / classes, it is possible to manipulate each attribute from the data body, and by extracting common terms of each attribute,
By giving meaning to attributed data related to them, that is, a subset of a set of entity objects, it is possible to reproduce a real model.

This is a database system comprising a set of entity objects in which a plurality of attributes are added to a single entity data, wherein the entity data is a hexadecimal number A in 4-bit units.
It can be a variable length text by SCII text conversion.

The attributes may be classified into a relationship attribute that defines a relationship with another object, an item attribute that defines the object itself, an edit attribute that defines the modification of the entity data of the object, and a classification into other attributes. it can. It may include the structure of each class that defines the entity data, the relation attribute, the item attribute, and the editing attribute, and their mutual relation structure. In addition, a mechanism is provided that specifies that the modification based on the attribute character string is started by the offset from the beginning of the actual data in the edit attribute and the attribute character string, and can cope with various word processing documents.

According to the present invention, a database (databas) is a set of entity objects in which a plurality of attributes are attached to a single instance data.
e). Here, the attributes are classified into relation attributes, item attributes, and editing attributes. FIG. 2 shows this relationship.

The relationship attribute is provided with an ID of a master object in order to define a subordinate relationship between the real objects. It is possible to define a plurality of relation attributes in one entity data. Conversely, a set of entity objects having one entity object as a master object corresponds to a row (tuple) in a relational database.

Since an entity object can be subordinate to a plurality of master objects, it can belong to rows composed of different members. Thereby, a multilayer database is realized. FIG. 3 shows this relationship.

Also, a row by a composite key in a relational database can be expressed by a logical product (AND) of the entity objects subordinate to a plurality of master objects. Thereby, a multidimensional database is realized. FIG. 4 shows this relationship.

FIG. 5 shows the structure of the database management classes, data management collections / classes, relation attribute management collections / classes, item attribute management collections / classes, and edit attribute management collections / classes, which are the frameworks of the present invention, and the respective collections and classes It is explanatory drawing of mutual relationship.

The item attributes are given the ID of the entity object, the ID of the column (attribute), and various other attributes. Selection by giving a conditional expression to a specific column ID
Is performed, a set of the corresponding real objects is selected. Then, a set of master objects is obtained from these entity objects, and a set of slave objects is obtained for each master object. By selecting the slave object to be acquired in advance by the column ID, a projection in the relational database is obtained.

By performing a logical operation between the columns with respect to the result of the selection for a plurality of columns, it is possible to select rows and project columns based on complicated conditions. The modification information in the editing attribute is defined by the offset from the head of the entity data to be modified and the attribute character string, in association with the entity data that the modification is started based on the attribute character string.

FIG. 6 shows the processing for each stage of the system of the present invention. The description of the processing at each stage is as follows. In the data operation environment, data (data location, creation source application, etc.) for associating with the original data is generated. In the data operation environment, since the means for data users to recognize data must be unified in order to utilize the data, extract the information necessary to recognize the data. , Generated as cognitive data. Since the recognition data has a property defined by the usage environment, the technology for generating the recognition data can be set individually, and can be selected according to the usage environment. Analyze the data contents to subdivide the information on the data structure, and generate the structural data in the subdivided state. Analyze data contents to unify information related to data attributes and unify them.In data operation in different environments, add multiple cognitive data, structural data, and attribute data if necessary. . The generated cognitive data and structural data are arranged as layers to generate objects.

FIG. 7 shows data conversion in the system of the present invention. A, B, and C represent the format of the data structure. A ′, A ″, A ′ ″ represent a state in which the data structure A is subdivided, and B ′, B ″ represent a state in which the data structure B is subdivided. X () indicates an operation means for the content of the attribute.

FIG. 8 shows hierarchical management of data by layers. That is, the shaded surface indicates the recognition data, and the plain surface indicates the structure data.

[0038]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An example of actual use of the system of the present invention will be described. In the database system used in the system of the present invention, since it is not necessary to define the database structure in advance, it is possible to first import data on the host computer and then add attribute information according to the file layout diagram. it can. The data is analyzed / evaluated and used for decision making based on the data.
Hereinafter, an operation example of the system of the present invention will be described.

(Example 1) In a certain company, data different for each department is unified, and data as various assets has already been constructed. For example, in department A, data format A is used. , B department adopts data format B. There was a need in this company to share data company-wide.

When the system of the present invention is introduced, it is possible to unifyly manage data in a form including the data format A and the data format B.
The department B can share data while keeping the data format B in operation.

(Embodiment 1) The data also includes information which is usually added due to functional reasons of the application that created the data. In the process of decomposing and generating the data structure, by removing essentially unnecessary information as data contained in the original data, it is possible to regenerate the data with the quality of the data raised as a result It becomes possible.

(Embodiment 3) New data is created by fusing a plurality of data. However, a need has arisen to replace some of the newly created data with different data for reuse. In the present invention, since the data constituting the data is held and managed in a subdivided form, the replacement of the constituted data can be very flexibly handled.

(Embodiment 4) A need has arisen for the holding company to collect the data of the core system of the investment target company and always keep track of the business state of the investment target company. It is assumed that the core system of the investment target company is constructed with data in completely different formats according to the individuality of each company, and that the target is assumed to be replaced. In the present invention, the data extracted from the core systems of each investment target company is integrated and can be managed in a unified manner, so that unified analysis / evaluation can be performed without changing the system in the core systems of the investment target companies. And make decisions as a holding company.

Embodiment 5 In a company in which an ERP system has been introduced, it is necessary to obtain input data from an existing core system in order to obtain a meaningful output result of the ERP. In a company, since the data of the core system that can be input data is stored in various formats, it is necessary to create converters corresponding to each data format, but the more input data is increased, the more conversion required The number of combinations to be performed increases geometrically, and it is not possible to cope with the situation.

In the system of the present invention, since multi-polar data is subdivided and attributes are standardized, in order to meet this need, output data (input data as viewed from the ERP) is required. It is only necessary to add a function for converting to a format. Even when input data is increased, the number of combinations only increases proportionally.

(Embodiment 6) It is necessary to represent the same data in different media formats due to differences in the use environment. Specifically, there is a need to represent the use environment A with pictures and the use environment B with characters. In the present invention,
Since it is possible to include expressions in a plurality of media formats, it is possible to perform expressions using different media while maintaining data in a unified manner.

(Embodiment 7) In the field of civil engineering and construction, the construction period is long, and even after delivery, maintenance and extension / remodeling by another contractor occur. Therefore, different multiple CAD
It is desired to use system data by centralized management. Even in such a case, the database system of the present invention is capable of centrally managing data of different CAD systems, which is effective for process management.

(Embodiment 8) Text information and modified editing information are separated from rich text (text information and modified editing information are mixed in a document file) included in a word processor or spreadsheet software. This makes it possible to centrally manage documents and tables of different creation sources together with image data and sound data, and also to create a multi-media document.

[0049]

According to the present invention, as shown in the above examples, it is possible to integrate data in an operating environment without being aware of the difference in the original data structure and to share data. .

In the database system of the present invention, before the data storage structure (schema) is determined, a large amount of data is fetched from another environment. Can be added after being arranged.

Of course, in the accompanying conversion tool, the old attribute can be separated and held, so that the new attribute is manually added only at the changed part, and the other attributes are referred to, and the trouble involved in the conversion is saved. It is also possible to omit it. In addition, since the data retention management does not depend on the storage structure, even if the attribute of the data changes, the situation can be dealt with without any inconvenience.

Since entity data exists independently of attributes, it is represented by a compound document by separating old attributes such as multimedia data, word processors, and spreadsheets having different attributes and adding a new common attribute. Information can be centrally managed.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of a general method of data conversion in the related art.

FIG. 2 is an explanatory diagram of a relationship among entity data, a relationship attribute, an item attribute, and an edit attribute that constitute an entity object according to the present invention.

FIG. 3 is an explanatory diagram of a state of a multi-layer database according to the present invention.

FIG. 4 is an explanatory diagram of a state that becomes a multidimensional database in the present invention.

FIG. 5 is an explanatory diagram of a mutual relationship between classes in the database according to the present invention.

FIG. 6 is an explanatory diagram of a step-by-step process of the system of the present invention.

FIG. 7 is an explanatory diagram of data conversion of the system of the present invention.

FIG. 8 is an explanatory diagram of hierarchical management of data by layers in the system of the present invention.

Claims (10)

[Claims] In an electronic data integrated data management system for integrating and managing various data, (1) a source for generating information (data location, creation source application, etc.) for associating with original data; Means for generating information relating to data, (2) Cognitive data extraction means for extracting information necessary for recognizing that it is data and generating it as cognitive data, (3) Analyzing data content and segmenting it (4) Structural data analysis and segmentation means to generate structural data as it is, (4) attribute data analysis and normalization means to analyze data contents and generate attribute data in a standardized state, (5) ) One or more cognitive data, structural data,
Add attribute data, data addition means, (6) arrange generated cognitive data, structural data as a layer,
Object generation means, object conversion means, (7) storing the object, object storage means,
(8) An electronic data integrated data management system comprising: an attribute data adding unit for adding an attribute to the object. 2. The cognitive data extracting means reproduces information and cognitive data necessary for generating cognitive data without previously defining an electronic data format for physically expressing the cognitive data. 2. An electronic data integrated data management system according to claim 1, wherein said means is a means for holding information relating to information necessary for said information and defining only an operation method for said information. 3. The structure data analyzing / subdividing means,
2. The electronic data integrated data management system according to claim 1, further comprising means for decomposing according to a rule defined in advance and classifying structural data that does not match any rule as structural data having a special state. 4. The attribute data analyzing and standardizing means includes:
Among the structural elements included in the data, for the structural data that is the minimum unit having meaning as data, attribute data defined as additional information of the structural data related to name, position, state, representation method, etc. 2. The electronic data integrated data management system according to claim 1, further comprising means for adding, changing, and deleting objects. 5. The electronic data integration according to claim 1, wherein the original data is managed as a plurality of elements of cognitive data, structural data, and attribute data, and the data is managed as a embodied object. Data management system 6. A database system comprising a set of entity objects having a plurality of attributes attached to a single entity data. 7. The method according to claim 6, wherein the entity data is 16 bits in 4-bit units.
7. The database system according to claim 6, wherein the text is a variable-length text converted to hexadecimal ASCII text. 8. The attribute may be a relation attribute that defines a relationship with another object, an item attribute that defines the object itself, an edit attribute that defines modification of the entity data of the object, and a classification into other attributes. 7. The database system according to claim 6, wherein: 9. The database system according to claim 6, further comprising: a structure of a class that defines entity data, a relation attribute, an item attribute, and an editing attribute, and a mutual relation structure thereof. 10. A mechanism according to claim 8, further comprising a mechanism for defining that the modification based on the attribute character string is started by an offset from the head of the actual data in the edit attribute and the attribute character string. Database system.