RU2705456C1 - Method of creating topology of metamodel of information space of domain and system for its implementation - Google Patents

Abstract

FIELD: information technology.

SUBSTANCE: invention relates to the field of information technology. Disclosed is a method of searching information in dissimilar information media, comprising steps of: forming a metamodel of the subject area in the form of an information space topology, which is homogeneous by nature and in which root vertices of information objects are not determined in advance, wherein root vertices of information objects are assigned or automatically determined at the moment of request to information resources formation; receiving user request to information resources; determining at least one property in at least two entities, information on which must be sent in response to the received user request; determining the relationship between the property of the first entity and the property of the second entity, forming a tree query structure to information resources, in which the relationship between the property of the first entity and the property of the second entity is selected as the root node; filling the tree structure of the request to information resources with data and/or links to these data on objects of the domain; displaying the results of the information request to the user in response to the user's request in the form of a data set in accordance with the information request structure.

EFFECT: possibility of searching information in different information media.

4 cl, 7 dwg, 1 tbl

Description

FIELD OF TECHNOLOGY

The invention relates to the field of information technology and is intended to create a metamodel of the subject area, in particular, modeling the topology of the information space of the subject area.

BACKGROUND

It is known from the prior art that the main components of the information space of information systems are metamodels of objects of the subject area, as well as functional and structural relationships between them. Each information object is built as a separate structural unit, independent of others, belonging to any category or group of objects of the subject area, for example, groups: “people”, “transport units”, “buildings”, “machines”, etc. Each object is assigned a group of properties that uniquely defines it in the information system. For example, for a person, such properties are the person’s name, person’s middle name, person’s surname, person’s TIN, person’s birth date, gender, etc. For machines - this is the model of the machine, the power consumption of the machine, the performance of the machine, the overall dimensions of the machine, the weight of the machine, etc.

The traditional methods of decomposing and synthesizing the structures of information objects that describe the objects of the subject area are described in the works of Data K.J. [Date C. J. Introduction to Database Systems = IntroductiontoDatabaseSystems. - 8th ed. - M.: Williams, 2005 .-- 1328 p. - ISBN 5-8459-0788-8 (Russian) 0-321-19784-4 (English).], Kogalovsky M.R. [Kogalovsky M.R. Encyclopedia of database technologies. - M.: Finance and Statistics, 2002. - 800 p. - ISBN 5-279-02276-4], Kuznetsova S.D. [Kuznetsov S.D. Databases: a textbook for students. institutions of higher prof. education / S.D. Kuznetsov. - M.: Publishing Center "Academy", 2012. - 496 p. - (University textbook. Ser. Applied mathematics and computer science). - ISBN 978-5-7695-8430-5] and others.

The disadvantage of these methods is the presence of stable relational or object relationships, which leads to the creation of monolithic structures of information objects that do not allow to quickly modify the topology of the information space of information systems.

The proposed method allows you to build the topology of the meta-model of the subject area without dividing it into categories, classes, subgroups, etc., only indicating the relationship between the individual properties of the considered entities of the subject area. The structure of the information object is obtained automatically. For this, it is necessary to select its element in the topology of the information space, which becomes the root vertex of the information object. The root vertex of an information object uniquely determines its structure at the current time. The depth and number of connections of an information object is determined from the condition whether the information object will be analyzed in a human-machine system or will it be processed by means of intelligent data processing.

This feature allows you to:

- simplify the procedures for the formation of structures of information objects;

- expand the number of considered parameters of the topology of the information space of the subject area;

- simplify the procedure for extracting data about information objects and their relationships for analysis of the current state of the subject area;

- to analyze the interconnected data through the relationship between the properties of the objects of the topology of the information space of the subject area;

- carry out quick reconfiguration of requests to the data of information resources of information systems depending on the changing need for information when changing the target settings of their consumers;

- eliminate data processing anomalies.

The prior art also known application US 20090049422 A1 (https://patents.google.com/patent/US20090049422A1/), which describes a method of creating software applications based on object-oriented programming without writing program codes, by selecting the necessary objects, services etc., as well as data extraction based on the formation of the root vertex of the parent object, which, however, is not derived from any other type of entity except for a predetermined one.

SUMMARY OF THE INVENTION

The technical problem to which the claimed solution is directed is to create a network structure for the topology of the information space of the subject area, the root vertices of which can be any elements of the network structure of the topology of the information space of the subject area, which allows you to:

- information search in local and distributed databases, Internet resources and other data sources;

- extracting structures of information objects;

- combining the structures of information objects into groups and their projections in the form of many tree structures.

The technical result is the creation of a metamodel topology for the information space of the subject area, the distinguishing feature of which is the ability to extract data from the information space of the information system, taking into account the multidimensionality of the network connections of the topology structure of the information space of the subject area, which are limited only by the computing capabilities of computer equipment, microprocessors, memory devices, bandwidth LAN and Internet channels with mouth copulating new (hidden) patterns of formal and / or non-formalized relationships domain objects.

The proposed method allows not to determine the root vertex of the request in advance, but to set it depending on the emerging information need.

The root of the request is:

- any named property of an entity;

- any previously established relationship between the properties of entities;

- any established relationship between the relationships of the properties of entities with respect to which a projection of the topology of the information space of the subject area is formed and its metamodel is created.

The selected root vertex becomes parent, and all other elements of the structure become children. If another property of an entity, or another relationship between properties of an entity, or another relationship between relationships of properties of an entity is selected as the root vertex of an information query, then the root vertex that was the parent becomes a child.

BRIEF DESCRIPTION OF THE DRAWINGS

For a better understanding of the essence of the invention, and to more clearly show how it can be implemented, hereinafter, reference will be made, only as an example, to the accompanying drawings, on which:

figure 1 - an Example of designing the structure of the information object;

figure 2 - an Example of a relationship diagram of the properties of entities;

figure 3 - An example of a scheme for extracting data on root vertices about the properties of entities based on established relationships between them;

figure 4 - an Example of the formation of the root top of the data request based on the establishment of relationships between relationships of properties of entities;

5 - An example of the organization of access to the information resources of the information system according to the selected root vertex of the data request;

6 - Algorithm for the formation of elements of the topology of the information space of the subject area: entities, entity properties, relationships between entity properties;

Fig.7 - the Algorithm for the formation of the topology of the information space of the subject area in accordance with a given tree of achieving goals.

DETAILED DESCRIPTION OF THE INVENTION

The topology of the information space of the subject area is a combination of structured and poorly structured information stored in the corresponding databases and knowledge bases, including:

- objects of the subject area, their metamodels, functional and structural relationships between them;

- a set of structured and weakly structured data describing the properties of objects of the subject area and the relationships between them;

- target settings for managing domain objects;

- methods of data processing and achievement of target settings for managing domain objects;

- as well as any other elements that carry out the creation, storage, modification, processing, transmission and deletion of data within a given information space.

The metamodel of the information space of the subject area consists of the elementary properties of the objects of the subject area, the objects of the subject area (projections of entities) and the relationships between them.

The basic elements of the metamodel of the information space of the subject area are the elementary properties of the objects of the subject area (projections of entities) and the relationship between the elementary properties of entities.

The elementary property of an object of an object domain (projection of an entity) is the attribute of an object of an object domain that is indivisible at the moment of time, which has its own meaning and which has its own measure, unit of measure, and data type. An example of an elementary property of an entity is a person’s name or the name of a settlement in a person’s place of residence.

The relationship between the elementary properties of the object of the subject area provides the conditions for connecting the elementary properties of the object / objects of the subject area, and is a named part of the topology of the information space, which has its own identification number (id), as well as addressing to two elementary properties of the entity, where addresses are their identification codes (id of elementary properties of an entity). Example: Passport belonging to a person.

A measure is a means of forming a measuring axis for assessing the qualitative and quantitative elementary properties of an object in an object domain (entity projection), which is described by different units of measurement and data types. Also, the system of measures allows you to correlate one entity of the subject area with another and evaluate the relationship between the individual elements of the subject area. Example: a measure of time whose units are seconds, minutes, hours, days, etc.

An entity is a named category of an object or objects of a subject area, united by one or a group of attributes, which is a stable independent structure. An entity may correspond to one or more of its projections - objects of the subject area, each of which may include one or more other objects of the subject area, a set of elementary properties and relationships between them.

The projection of the entity corresponds to the specific role that it performs in the subject area. Example: an entity is a car, a projection of an entity is a sports car, a passenger car, a cargo car, etc.

The set of properties of an entity is the set of all properties of objects that are projections of this entity. For example, the entity “Man” can be represented by such projections of the entity (objects) as “Personal data of a person”, “Education of a person”, “Family and tree of kinship relations of a person”, “Professional activity of a person”, etc., with each the projection of the entity has its own set of properties that characterize it.

The more complete the metamodel of the topology of the information space of the subject area, the more complex problems can be solved. As the subject area becomes more detailed and the range of tasks to be expanded, the formal description of the information space of the subject area will take on a new configuration, changing its topology by including in it:

- new entities;

- new objects of the subject area (projections of entities);

- new elementary properties;

- new connections between the elementary properties of domain objects (projections of entities);

- modernization of previously created structures and relations between them;

- inclusion in it of calls (links and access keys) to other information resources located outside the previously considered subject area (information systems, databases, knowledge bases, resources of the global Internet network).

An example of representing the structure of an information object (information model of the projection of an entity) is shown in figure 1.

The derived structure of the metamodel of the topology of the information space of the subject area is a set of information objects considered as interconnected tree structures, the root vertex of which can be any connection established in the metamodel of the topology of the information space of the subject area. The choice of the root vertex depends on the type of tasks being solved. The set root vertex of an object can be:

 any named property of a domain object (entity property),

- any previously established relationship between the properties of objects of the subject area (relationships between the properties of entities),

- any established relationship between the relationships of properties of objects of the subject area (the relationship between the relationships of properties of entities).

Based on this root vertex, any information query structure can be formed on the basis of properties available within a given set of properties and relationships of projection entities of the metamodel of the information space topology (see Figures 2-5).

Consider an example of a relationship scheme of entity properties (see figure 2). Let A, B, C be the essence of the subject area. Each entity is described by a set of elementary properties. The number of properties considered, as well as their types, depends on the type of task.

In the presented example:

- entity A is described by a set of properties a ₁ , a ₂ , a ₃ , and ₄

- entity B is described by a set of properties b ₁ , b ₂ , b ₃ , b ₄ , b ₅

- entity C is described by a set of properties from ₁ , s ₂ , s ₃ , s ₄ , s ₅ , s ₆

Within each individual entity, all its properties are interconnected, forming a graph with bidirectional connections.

Unidirectional or bidirectional communications can be established between some properties of different entities. Moreover, each property of an entity can have one or more unidirectional or bidirectional connections with another property of another entity / other properties of other entities.

In the presented example, bidirectional relationships are established between the following entity properties:

- property b _{5 of} entity B and property a _{2 of} entity A;

- property b _{2 of} entity B and property a _{3 of} entity A;

- property b _{4 of} entity B and property c _{6 of} entity C;

- property a _{3 of} entity A and property c of ₂ entities C;

- property a _{3 of} entity A and property c of ₄ entities C.

Example:

Essence A - SCHOOL SARAHFAN.

The considered properties of entity A:

- and ₁ - growth

- and ₂ - size

- a ₃ - fabric color

- a ₄ - article of fabric

Entity B is GIRL.

The considered properties of entity B:

- b ₁ - date and place of birth

- b ₂ - hair color

- b ₃ - eye color

- b ₄ - visual acuity

- b ₅ - waist circumference

Essence C - TRAINED.

The considered properties of entity C:

- from ₁ - year of enrollment in the first grade

- from ₂ - class number (1, 2, 3, ...)

- c ₃ - subclass number (A, B, C, D, D)

- from ₄ - school number

- from ₅ - desk number in the classroom

- from ₆ - the city in which the school is located

Considered relationships of entity properties:

- girth at the waist of the GIRL (b ₅ ) - the size of the SCHOOL SARAHFAN (a ₂ )

- GIRLS hair color (b ₂ ) - SCHOOL SARAHFAN fabric color (a ₃ )

- GIRL visual acuity (b ₄ ) - school desk number in the STUDENT class (from ₅ )

- date and place of birth of the GIRL (b ₁ ) - the city in which the student’s school is located (from ₆ )

- the color of the SCHOOL SARAHFAN fabric (a ₃ ) - the class number of the TRAINED (from ₂ )

- the color of the SCHOOL SARAHFAN fabric (a ₃ ) - the number of the TRAINED school (from ₄ )

An information request, and therefore the data retrieved from the information system, may relate to:

- one or more properties of one entity;

- one or more properties of several entities.

If a request is made for one property or several properties of one entity, then we use existing or designed at the time of formation of the structure of the entity of the connection (see figure 2).

Example of an information request: Class number, school number, city in which the learner's school is located.

Each property of an entity can correspond to its own information resource, which is accessed either through an information request to a relational database, or reference information, or to a text or tabular form for recording information, or other structured or weakly structured data. At the same time, restrictions on the storage method, algorithms for extracting data from a resource, or its presentation format are not set in advance. Thus, the property of an entity, in addition to its name, identification code, measure, unit, and data type, includes the address of the access (access point) to the data, which can be a DNS address, IP address, page address of a web resource, access path to table / tables or text file / text files, as well as any other information recorded in the appropriate format, in a local computer system or the global Internet.

 If a request is made about one or several properties of several entities, then the root vertex of the request is formed, which is a relationship between the properties of entities (see figure 3).

Request example :

(b ₄ –c ₅ ) What places on the desks are GIRLS depending on the visual acuity in the student's school?

(b ₁ –c ₆ ) Do trainee girls born in 2000 in Moscow or St. Petersburg study at the school ?

Choosing the relationship between the properties of entities as the root vertex, the parent vertex of the analyzed fragment of the topology of the information space is determined, which includes the established relationships between the properties of entities that are part of the fragment of the topology of the information space and the relationship of the properties of the entity within each of them (see figure 4). Thus, when designating the root vertex of the query, a new (or is selected from among the previously established) relationship between the properties of entities is established, allowing gender learn access to data of all properties of the analyzed entities.

Depending on the selected root vertex of the information query (see Figure 3), the access path to the property data of the analyzed entities is determined, which can be represented as a graph with unidirectional and bidirectional connections.

By imposing restrictions on the requested data and the conditions for accessing it, as well as using well-known methods for determining the shortest way to build a chain of links for data extraction (for example, the method of calculating the shortest distance, etc.), the process of generating, processing an information request, and extracting data from information resources fragment of the topology of the information space of the information system.

The algorithm for generating a query for data extraction consists in imposing restrictions on data instances of each property of each entity, taking into account the existence of relationships between instances of entity properties.

The result of imposing restrictions gives a selection of data instances that satisfy the formulated query. When assigning the root vertex, the point and direction of the search for data instances are determined based on the established relationships between the properties of entities.

The root vertex of an information query can represent a single-level or multi-level system of relationships between entity properties.

Distinguish (see figure 5):

- the relationship of the properties of entities (for example, (b ₁ –c ₆ ), (f ₁ –e ₄ ));

- the relationship of the properties of the entity and the relationship of the properties of the entities (for example, (f ₂ - (b ₁ –c ₆ ));

- a relationship between the properties of entities (for example, (b ₁ –c ₆ ) - (f ₁ –e ₄ ).

To demonstrate an example of established relationships, we complement the set of entities described in the previous example with additional entities:

- entity D - ADDRESS

- essence E - FAMILY

- entity F - MEDICAL MAP

Considered properties of entity D:

- d ₁ - name and code of the settlement

- d ₂ - post office of the settlement

- d ₃ - street name, house and apartment number

Considered properties of entity E:

- e ₁ - name of mother, TIN, year and place of birth

- e ₂ - name of father, TIN, year and place of birth

- e ₃ - full name of brothers, TIN, year and place of birth

- e ₄ - full name of sisters, TIN, year and place of birth

The considered properties of entity F:

- f ₁ - full name, year and place of birth, No. of birth certificate

- f ₂ - name of the attending physician

- f ₃ - storage location of the medical record

Considered relationships of entity properties:

- (f ₁ - d ₃ ) - (name, year and place of birth, birth certificate number) MEDICAL CARD - (street name, house and apartment number) ADDRESS

- f ₃ - d ₂ - (storage location of the medical record) MEDICAL CARD - (post office of the settlement) ADDRESS

- f ₁ - e ₄ - (name, year and place of birth, birth certificate No.) MEDICAL CARD - (name of sisters, TIN, year and place of birth) FAMILY

Examples of forming the root vertex of a data request:

from communication (f ₁ - e ₄ ) - The number of sisters born after 2010 with the owner of the MEDICAL CARD Petrova Valentina Ivanovna, born in 2000, born in Perm.

from the relationship between the properties of the essence and the relationship between the properties of the entities (f ₂ - (b ₁ - c ₆ )) - Name of the doctors treating the girls, students in the school of TRAINED GIRLS, born in 2000 in Moscow or St. Petersburg

from the relationship of the properties of entities ((b ₁ - c ₆ ) - (f ₁ - e ₄ )) - Full name of the treating doctors GIRLS TRAINING in schools No. 5 and No. 7

The classification of data extraction by root vertices is presented in Table 1.

Table 1 - Classification of the root nodes of the query

Type of root vertex Root vertex designation Root vertex - property of an entity (zero-level relationship) X _i Root vertex - the relationship between the properties of entities (first-level relationship) X _i -Y _j Root peak - the relationship between the properties of an entity and the relationship between the properties of entities (second-level relationship) Z _k - (X _i - Y _j ) Root vertex - a relationship between relationships of entity properties (third-level relationship) (X _i - Y _j ) - (Z _k -V _m ) The root peak is the connection between the bonds of the bonds of the properties of entities - the bonds of the fourth, fifth and further levels. They are used to build relationships between the properties of objects (projections of entities) of different subject areas based on mathematical, logical or intuitively assigned dependencies of structured and / or weakly structured data.

An example, the root vertex (a property of an entity) is “a person’s belonging to the classmates”:

connection

(entity properties = object properties) - Name, TIN

(projection of entity = object) - PERSONAL DATA

(essence) - MAN

and

(entity properties) - CLASS NUMBER, YEAR OF EDUCATION, SCHOOL NUMBER, CITY, STUDENT LIST

(projection of entity = object) - EDUCATION

(essence) - MAN

By assigning as the root node of the information query the connection relationship “person belongs to the number of classmates” and the property “place of work” of the projection of the entity “professional activity” of the entity “person”, one can obtain data on the history of professional activity of all classmates of the person in question.

The established root vertex (an entity property, an entity property relationship, an entity property relationship relationship, an entity property relationship relationship, an entity property relationship relationship) acts as the parent element of the tree structure of a fragment of the information space topology, the construction depth and level of detail of which is determined by the way data is presented and processed: automated (human-machine system) or automatic (means of intelligent data processing).

When implementing human-machine data processing, a fragment of the topology of the information space should be represented on the plane (screen form) as a set of interrelated properties of entities, measures, units of measurement and types of data representation in a computer system for the convenience of human perception.

In automatic data processing using artificial intelligence methods, such as artificial neural networks, genetic algorithms, etc., the dimensionality of the representation of an information object is limited by the computing capabilities of the information system. So, for example, regression analysis allows you to obtain predicted values in flat and linear systems with a communication depth of not more than one. And when using artificial neural networks, multidimensionality is limited, for example, by the number of layers of the neural network, the accuracy of the forecast) and the like.

The following properties are inherent in the obtained information structures of a fragment of the topology of the information space:

- property of polymorphism (variety of forms) of bond structures;

- the property of collinearity of structures (the ability to use the generated structure of the data request in different subject areas);

- the property of the accumulation of relationships between the properties of entities and relationships of relationships between the properties of entities, for subsequent reference to previously accumulated experience, with newly formed information requests.

The method of forming entities, entity properties, relationships of entity properties of the topology of the information space of the subject area can be described by the following sequence of actions (see Fig.6). As technical means providing the formation of entities, entity properties, relationships of entity properties of the topology of the information space of the subject area, for example, a data processing device implemented on the basis of at least one processor or microcontroller made in the hardware-software part in this way to carry out the functions attributed to them below.

1. Accordingly, at the first stage, the initial structure of the topology of the information space of the subject area is formed. At this stage, the data processing device receives information about the subject area, in particular, information about the totality of real processes and objects (entities). Information about the subject area can come directly from sensors that record this information, or from the corresponding databases designed to store this information.

1.1 Based on the received information about the subject area, the data processing device generates a set of entities describing the subject area, and each entity is assigned an entity name, a unique code (id) (see block 1.1 of figure 6). The name of the entities and their id can be obtained from the corresponding containers of entities located in the database of entities by generating the appropriate queries.

1.2 Next, the data processing device performs the formation of a set of elementary properties of the entity, where each elementary property is assigned a name and a unique code (id) after checking the availability of this property in the database of elementary properties of entities (see block 1.2 of figure 6). Before starting the formation of a set of elementary properties of an entity, the data processing device performs a query to the database of entity properties (containers of properties of the properties database) containing an indication of the name and / or code of the entity. The data processing device is also configured to provide information about the elementary properties assigned to the entity at the request of the user. For each property of an entity, information is given on the name of the property, its identification code, measure, unit of measure, data type, and a link to the posted information resource.

If an entity in the topology of the information space is only being formed and has not been previously presented, then upon request a message will be issued indicating the absence of the previously generated properties of this entity. After that, information about the characteristics of the properties assigned to the entity is entered into the database of entity properties. If necessary, the structure of the properties of the entity is adjusted.

After completion of data entry or adjustment, confirmation is made that the properties of the entity are reflected in the information system, and the transition to the next stage is carried out (see block 1.3 of figure 6).

1.3 Next, the data processing device in the automatic mode creates links between the elementary properties of entities, where each link is assigned a link name and a link code (id) after checking if this link is in the database of links of elementary properties of entities (see block 1.3 of figure 6). Information about the links in the links database can be stored in the respective links containers.

If the relationship between the properties of entities is only being established, the data processing device makes a request to the database of relationships of properties of entities for the presence of already existing (established) relationships, and then new relationships are formed. At the same time, for each connection, the names and identification codes of the properties of entities being connected among themselves, the direction of the connection (from property 1 to property 2; from property 2 to property 1; bidirectional communications) are indicated; communication conditions; the period of the beginning and completion of communication). If necessary, the structure of relationships between the properties of entities is adjusted.

After completion of data input or adjustment, the data processing device additionally checks that all the relationships between the properties of the entities are reflected for the formation of the information system, and the transition to the next stage is carried out (see block 1.4 of figure 6).

1.4 At the next stage, the data processing device automatically creates the structures of information objects based on the data obtained in the previous stages (paragraphs 1.1 - 1.3) and assigns a unique code (id) to the information object (see block 1.4 of figure 6).

1.5. In accordance with the obtained structure of information objects, the data processing device then automatically generates entity containers, property containers, and connection containers (see block 1.5 of figure 6).

Each entity container that has an original id-code corresponds to two types of container: groups of containers for entity properties and groups of containers for relationships between entity properties. Each property of the entity has its own data container that has an original id-code. Each relationship of the properties of the entity also corresponds to its own container, which has an original id-code and a number of other parameters.

The handle to the entity container contains the entity id, the procedure for generating a surrogate key (name of the entity instance), the procedures for recording, updating, and deleting records. The data structure of the entity container is invariant and consists of the id of the record in the container, the name of the entity instance, the date of its creation, the date of its adjustment, the date of deletion, as well as a data set characterizing the history of the change of the entity instance and a number of other parameters.

The handle to the container of the projection of the entity (object of the subject area) contains the id of the projection of the entity (object of the subject area), the procedure for generating a surrogate key (the name of the instance of the projection of the entity), the procedures for recording, updating, and deleting records. The data structure of the container of the projection of the entity (object of the subject area) is invariant and consists of the record id in the container, the name of the instance of the projection of the entity (object of the subject area), the date of its creation, the date of its adjustment, the date of deletion, as well as the data set characterizing the history of the change in the instance projections of the entity (object of the subject area) and a number of other parameters.

The handle to the container of entity properties contains the entity property code (id), the name of the entity property, measure, unit of measure, data type. The container of entity properties also contains typical procedures (methods for working with data) located in the container: creating a record, updating a record, deleting a record, and a number of other parameters.

The structure of the data record in the container of entity properties is invariant and contains the code (id) of the data instance in the container of entity properties, the value of the data instance in the container of entity properties, the time it was written to the container of entity properties, the time for adjusting data in the container of entity properties, time deleting an entry from the container of entity properties, as well as the history of changes in the value of the entity property and a number of other parameters.

An entity property relationship container contains information about relationships between entity properties. The handle to the entity property relationship container contains the association code (id), the name of the association, and two id codes of the properties being joined. It also contains typical procedures for establishing a relationship between the properties of entities and the conditions for extracting data according to relevant requests (restrictions on the selection of data from containers of entity properties). The container for linking properties of entities is an invariant structure and contains the internal code id for linking properties of entities, id of properties of entities from containers of properties of entities, time to establish a connection, time to remove a connection, and a number of other parameters.

The communication container for data instance communications contains information about the relationships between the properties of instances of information objects. The handle to the data instance communication container contains the data instance communication code (id), the name of the data instance communication, and two id codes of the connected properties of the entity data instances, including the Internet address. It also contains typical procedures for establishing a relationship between instances of entity properties and conditions for extracting data from relevant information requests (restrictions on data sampling). The communication container for data instance communications is an invariant structure and contains the internal code for the communication instance id of the data, the id code or data address on the Internet, the time it took to establish the connection, the time the connection was removed, and a number of other parameters.

1.6. Further, the data processing device automatically records in the structure of each information object the id-codes of entity containers, projections of the entity, properties and relationships between entity properties (see block 1.6 of figure 6) and filling information resources with data (see block 1.7 of figure 6), forming Thus, the structure of the topology of the metamodel of the information space of the subject area. Since root vertices are not defined in the mentioned structure, the topology of the meta-model of the information space of the subject area is homogeneous.

Stage 2. Expansion of the structure of the topology of the information space of the subject area:

2.1. The data processing device can also be programmed to perform automatic analysis and determine the need for complicating the metamodel of the topology of the information space using intelligent data processing methods that reveal new relationships and properties of entities (see block 2.1 of figure 6). The choice of methods for intelligent data processing depends on the objectives of the study of entities, projections of entities, properties of entities and relationships of the subject area, as well as selected information resources for its implementation. Downloading the intellectual data processing method is carried out from the corresponding knowledge base of the information system.

If the need for complicating the metamodel of the topology of the information space is established, the data processing device proceeds to step 1.1. If not, the data processing device proceeds to paragraph 2.2 .

At step 2.2, the data processing device clarifies the need to complicate the structure of information objects (entity projections) by adding new properties and relationships to the metamodel of the information space topology, after which it proceeds to 1.1. When forming a new entity or a new entity property, data is automatically processed for the possibility of establishing new relationships in the metamodel of the topology of the information space of the subject area, and writing them to the corresponding relationship containers.

At step 2.3, the data processing device fills the information resources with data.

The sequence diagram of the data processing device for forming the topology of the information space of the subject area in accordance with a given tree of achieving goals is presented in Fig.7.

Stage 1 consists in the formation of a tree to achieve goals or its adjustment, if it was previously created and its structure is stored in the appropriate database (block 1 of figure 7).

Stage 2. In accordance with the methods for achieving the goal and the conditions for its selection, laid down in the topology of the information space, the tree of business processes for achieving goals is constructed in an automated or automatic mode. Moreover, each business process includes at least one stage of the business process. At each stage of the business process, data is handled. Data is recorded in the database in accordance with the structure of information objects, which are divided into three groups according to the relationship to data: control objects, resource objects, control entities (block 2 of figure 7).

Stage 3. In accordance with the methods laid down for detailing the structure of objects and identifying a property with a process, each information object is represented as a set of property values. Moreover, each projection of an information object corresponds to its own set of properties, which may differ or partially coincide with another projection of this information object (block 3 of figure 7).

Stage 4. In accordance with the methods laid down for assessing the properties of an object, each property of an information object can be represented by at least one indicator having one or more units of measure (block 4 of figure 7).

Step 5. Writing data to the database (block 5 of figure 7).

Step 6. Forming data requests to information resources and extracting data (block 6 of figure 7) from the corresponding entity containers, containers of projections of entities (objects of the domain) and containers of properties of entities using communication containers of entities and communication containers of data instances, taking into account the restrictions imposed to sample data.

The query processing device performs the formation of queries as follows. The request for data search from the user is received by the data processing device, which determines at least one property in at least two entities (for example, property b1 of entity B and property c6 of entity C, see Fig. 5), information about which must be sent in response to a request received. The data processing device accesses, for example, a database with information resources stored in accordance with the previously defined structure of the topology of the information space of the subject area and determines the connection (for example, connection b1-c6) stored between the property of the first entity and the property of the second entity. If information about the relationship between the property of the first entity and the property of the second entity is missing, then the data processing device generates such a connection by methods known from the prior art. The resulting connection can be stored in a database with information resources.

Further, the data processing device generates a tree structure of the request for information resources, in which the relationship between the property of the first entity and the property of the second entity is selected as the root vertex, and the properties of the first and second entities, on the basis of which the mentioned relationship, the first and the second entity, all the properties of entities and their relationships. The data processing device fills the generated tree structure of the request with data and / or links to this data about the objects of the domain to create a metamodel topology for the information space of the domain and sends the received data to the user for display as a data set in accordance with the structure of the information request. The data processing device determines the number of nodes of the tree structure of the request for information resources based on the analysis of formalized relationships between entity properties and relationships of entity property relationships in accordance with the topology of the information space metamodel stored in the database.

In the process of working with the displayed data, the user may need to request additional information from the database with information resources. Accordingly, a refined or additional data request also arrives at the data processing device, which determines at least one additional property in at least two entities (for example, property e4 of entity E and property f1 of entity F), information about which must be sent in response to the received inquiry. In a similar way, the data processing device extracts stored information about the connection (communication f1-d3) between the additional property of the first entity and the additional property of the second entity from the database with information resources, or forms such a connection using methods known from the prior art.

Next, the data processing device generates a relationship (for example, a relationship (b1 - c6) - (f1 - e4)) between a relationship (b1 - c6) selected as the root vertex and a connection (f1 - e4) extracted at the previous stage between additional properties entities. The formed connection ((b1 - c6) - (f1 - e4)) between the mentioned connections is selected as the root vertex, and the connection (b1 - c6), previously selected as the root vertex, and the connection (f1 - e4) extracted at the previous stage are assigned by child nodes.

At the next stage, the data processing device refines the tree structure of the request for information resources in accordance with the selected root (parent) vertex and child vertices. Additionally, the tree structure of the request can be refined by nodes about entities and properties of entities associated with the additional property defined in the user’s qualifying request. The data processing device fills the refined tree structure of the request with data and / or links to data on objects of the subject area, thus forming the transformed topology of the metamodel of the information space of the subject area due to the emergence of new relationships between the properties of entities and properties of properties of entities, and sends the data in response to the user to an updated user request for display in accordance with the updated tree structure of the information request.

Additionally, the user’s request or the updated user’s request may contain information about the size of the tree structure of the request data to information resources, which is formed or refined by the data processing device. Based on the information about the size of the tree structure, the data processing device determines the number of nodes of the tree structure and, taking into account a certain number of nodes, generates or refines the tree structure of the request for information resources in the manner described earlier, which is further filled with data from the subject area.

The available working methods allow you to create new instances, generate surrogate keys, configure relationships between surrogate keys and properties of entities (domain objects), generate historical records for each property and surrogate entity keys, and also search within a given metamodel structure for the topology of the information space of the subject area.

Requests to information resources according to the proposed method of organizing the topology of the information space of the subject area allow the formation of multidimensional data structures, the dimension of which is limited only by the computing capabilities of computer technology.

Claims (15)

1. A method of searching for information in heterogeneous information environments, comprising stages in which: form a metamodel of the subject area in the form of a topology of the information space, which is homogeneous in nature and in which the root vertices of information objects are not determined in advance, and the root vertices of information objects are assigned or automatically determined at the time of generating a request for information resources; receive a user request for information resources; determining at least one property in at least two entities, information about which must be sent in response to a received user request; determine the relationship between the property of the first entity and the property of the second entity, form a tree structure of the request for information resources, in which the connection between the property of the first entity and the property of the second entity is selected as the root vertex; fill the tree structure of the request to information resources with data and / or links to this data about objects of the subject area; display the results of the information request to the user in response to the user request in the form of a data set in accordance with the structure of the information request. 2. The method according to claim 1, characterized in that the number of nodes of the tree structure of the request for information resources is determined based on the analysis of formalized relationships between entity properties and entity property relationships in accordance with the topology of the information space metamodel. 3. The method according to p. 1, characterized in that it further comprises stages in which: receive an updated request for data to information resources; determine at least one additional property in at least two entities, information about which must be sent in response to the received qualifying request; determine the relationship between an additional property of entities; form a relationship between a relationship previously selected as a root vertex and a relationship between additional entity properties, and select a formed relationship between said relationships as a root vertex, and a relationship previously selected as a root vertex, and a relationship between an additional entity property become child vertices ; specify the tree structure of the request for information resources, taking into account the assigned root vertex; Fill the refined tree structure of the request to information resources with data and / or links to data on objects of the subject area to create a transformed topology metamodel of the information space of the subject area due to the emergence of new relationships between the properties of entities and properties of properties of entities and send to the user the data displayed in accordance with refined structure of the information request. 4. The information retrieval system in heterogeneous information environments, containing at least one data processing device, configured to implement the method according to any one of paragraphs. 1-3.

Images