CN112181952B - Method, system, device and storage medium for constructing data model - Google Patents
Method, system, device and storage medium for constructing data model Download PDFInfo
- Publication number
- CN112181952B CN112181952B CN202011366878.9A CN202011366878A CN112181952B CN 112181952 B CN112181952 B CN 112181952B CN 202011366878 A CN202011366878 A CN 202011366878A CN 112181952 B CN112181952 B CN 112181952B
- Authority
- CN
- China
- Prior art keywords
- data model
- entity
- objects
- layer
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013499 data model Methods 0.000 title claims abstract description 110
- 238000003860 storage Methods 0.000 title claims abstract description 13
- 238000000034 method Methods 0.000 title claims description 31
- 238000005259 measurement Methods 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000004590 computer program Methods 0.000 claims description 16
- 230000006870 function Effects 0.000 claims description 11
- 238000005070 sampling Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 4
- 239000000126 substance Substances 0.000 claims description 2
- 238000010276 construction Methods 0.000 abstract description 9
- 230000004927 fusion Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 8
- 238000011161 development Methods 0.000 description 6
- 238000013461 design Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 2
- 238000007418 data mining Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013439 planning Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 230000008521 reorganization Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/211—Schema design and management
- G06F16/212—Schema design and management with details for data modelling support
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/288—Entity relationship models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a data model construction method, a system, equipment and a storage medium, comprising the following steps: 1) fusing each layer of objects in the data structure relation of the power data model by taking the SG-CIM model as a reference to form a big data model; 2) carrying out structured representation on each pair of object groups in the big data model, and then measuring the rationality of each pair of object groups in the big data model; 3) constructing a generator and a discriminator, performing counterstudy training on the generator based on the discriminator, the measurement result in the step 2) and the head entity and the tail entity of the real object group, and then generating a data structure relationship by using the trained generator to complete the construction of a data model based on the generation of the counterstudy and the knowledge fusion.
Description
Technical Field
The invention belongs to the field of power systems, and relates to a data model construction method, a data model construction system, data model construction equipment and a storage medium.
Background
Data are continuously generated along with development of an electric power system, acquisition and storage systems of different data models are established in various electric power professional fields according to professional requirements, and complex and huge electric power related data resources pose huge challenges to data management. Meanwhile, the technology of data application is increasingly enriched along with the rapid development of big data and artificial intelligence, the value of data analysis and mining is increasingly highlighted, the scheme based on the data-driven algorithm is very serious in dependence on data, data is required to be accumulated in a multi-dimension, high frequency, long accumulation and the like as much as possible, but in actual work, the model design of data acquisition is usually lagged behind the application, and the data analysis and mining are also limited by the available data condition.
The existing public data model (SG-CIM) of the national power grid company enterprise is a multi-level and full-coverage data model design for electric power data on the basis of a public information model (CIM) according to the expansion and reorganization of the CIM by the actual business of the national power grid company. However, the problems of incomplete coverage and description capability, insufficient consideration of the incidence relation between the model and the asset attribute and the like exist, so that a plurality of researches on SG-CIM-based extension or other data models exist, for example, a CIM-based power grid dispatching center application model, a CIM-based power grid graph maintenance and sharing scheme, a CIM-based power grid planning management decision auxiliary system and the like are provided, and the aim of designing a more complete data model is taken. However, it is difficult to get rid of the human experience, whether SG-CIM or optimization schemes rely on knowledge about individuals or small groups when designing data models. Therefore, a new approach is needed for designing a scientific, advanced, and efficient data model.
The SG-CIM model is a data model which is built by a national power grid company based on CIM standard extension, any original class of CIM is not changed in the extension process, and all modifications are carried out on the newly inherited class. The SG-CIM divides 12 primary theme domains and a plurality of secondary theme domains, and classifies and collects more than 1000 classes according to services. There are two main methodologies for data model design:
and designing from bottom to top, namely combing data entities accessed by an unstructured platform of a national network company and relations among the data entities, abstracting and refining the data entities, analyzing and merging data subject domains to which the data entities belong, analyzing relations among the subject domains and forming an unstructured data association model.
The top-down design is that the business requirements of unstructured data of each business line are sorted by analyzing and abstracting downwards from business targets and combining the existing business system, and according to the business process, key entities are abstracted, the relationship between the subject domain of the entity and the entity is analyzed, and the incidence relationship between the unstructured data entity and the structured data forms an unstructured data incidence model.
SC-CIM can not meet the requirements of a full-service data model at present, and the problems that the covering surface and the description capability are incomplete, the incidence relation between the model and the asset attribute is not fully considered and the like still exist, so that a plurality of SG-CIM-based expansions or other data model researches are provided, for example, a CIM-based power grid dispatching center application model, a CIM-based power grid graph maintenance and sharing scheme, a CIM-based power grid planning management decision auxiliary system and the like are provided, and the aim of designing a more complete data model is taken. However, no matter the SG-CIM or the optimization scheme is based on the knowledge of individuals or small groups when designing the data model, it is difficult to get rid of the constraint of human experience, so the comprehensiveness of the coverage object is poor, and no development margin is provided, and a new method is needed for designing an advanced data model.
Disclosure of Invention
The present invention is directed to overcoming the above-mentioned shortcomings in the prior art, and providing a method, a system, a device and a storage medium for constructing a data model, which can improve the comprehensiveness of a data model covering an object and have a development margin.
In order to achieve the above purpose, the method for constructing the data model comprises the following steps:
fusing each layer of objects in the data structure relation of the power data model by taking the SG-CIM model as a reference to form a big data model;
carrying out structural representation on each pair of object groups in the big data model, and measuring the rationality of each pair of object groups in the big data model according to the structural representation result to obtain a measurement result;
acquiring a head entity and a tail entity of a real object group, constructing a generator and a discriminator, performing counterstudy training on the generator based on the discriminator, the acquired measurement result and the head entity and the tail entity of the real object group, generating a data structure relationship by using the trained generator, and constructing a data model according to the data structure relationship generated by the generator.
Taking the SG-CIM model as a reference, fusing objects of each layer in the data structure relationship of the electric power data model to form a big data model, wherein the concrete operation is as follows:
collecting each power data model, extracting a data structure relation of each power data model, calculating the similarity between objects of each layer in the data structure relation layer by layer from bottom to top by taking the SG-CIM as a reference from a subtopic domain in the SG-CIM, and fusing two layers of objects with the similarity being more than or equal to a preset threshold value into a new object to form a big data model.
The similarity between two-layer objects is the sum of the similarities between the sub-layer objects in the two-layer object.
And calculating the similarity between the sub-layer objects based on the electric power word vector.
Carrying out structural representation on each pair of object groups in the big data model, and carrying out measurement on the rationality of each pair of object groups in the big data model according to the result of the structural representation comprises the following specific operations:
for each pair of object groups in big data model
Carrying out structured representation, wherein h is a head entity, r is a relation, t is a tail entity, and a projection matrix is defined for the relation r
Projecting the vector of the head entity and the vector of the tail entity from the entity space to a subspace of the relation r, wherein the head entity vector in the subspace
Tail entity vector in subspace
Relationships in subspaces
And calculating the distance between the two entities, and measuring the rationality of each object group according to the distance between the entities.
In the counterstudy training process, random sampling replacement is carried out on a head entity and a tail entity of a real object group to form disturbance, and then the disturbance is converted by a longitude quantity characteristic encoder and then input into a generator.
The loss function for the counterlearning training is:
wherein the content of the first and second substances,
in the interest of expectation,
as a feedback parameter for the discriminator,
representing a generating function consisting of a measurement sample and a random perturbation,
the weighted vector sum of the measures is prefaced, z represents the random perturbation, and x represents the true sample.
A construction system of a data model includes:
the generating module is used for fusing each layer of objects in the data structure relation of the power data model by taking the SG-CIM model as a reference to form a big data model;
the measurement module is used for carrying out structural representation on each pair of object groups in the big data model, measuring the rationality of each pair of object groups in the big data model and obtaining a measurement result;
the confrontation learning module is used for acquiring a head entity and a tail entity of the real object group, constructing a generator and a discriminator, carrying out confrontation learning training on the generator based on the measurement result of the discriminator and the measurement module and the head entity and the tail entity of the real object group, generating a data structure relationship by using the trained generator, and constructing a data model according to the data structure relationship generated by the generator.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method of constructing the data model when executing the computer program.
A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method of constructing the data model.
The invention has the following beneficial effects:
when the method, the system, the equipment and the storage medium for constructing the data model are specifically operated, firstly, objects of all layers in the data structure relationship of the electric power data model are fused to form a big data model, then, a generator and a discriminator are constructed, then, a counterstudy method is utilized, counterstudy training is carried out on the generator based on the discriminator, the combined rational measurement result of each pair of objects in the big data model and the head entity and the tail entity of the real object group, and in the process of the counterstudy training, objects which are possibly not used are generated to break through the limitation of human experience, so that the finally obtained data model covers the objects more comprehensively and has certain development margin.
Further, in the process of the countercheck learning training, random sampling replacement is carried out on head entities and tail entities of the real object group to form disturbance, so that objects which are not used by the data model are partially generated, and the comprehensiveness of the data structure model covering the objects is improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention. In the drawings:
FIG. 1 is a perspective view of a space;
FIG. 2 is a flow chart of generating a countermeasure;
FIG. 3 is a flow chart of the present invention.
Detailed Description
The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
The following detailed description is exemplary in nature and is intended to provide further details of the invention. Unless otherwise defined, all technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention.
Referring to fig. 3, the method for constructing a data model according to the present invention includes the following steps:
1) fusing each layer of objects in the data structure relation of the power data model by taking the SG-CIM model as a reference to form a big data model;
specifically, the electric power data models are collected, the data structure relation of the electric power data models is extracted, the similarity between the objects in each layer in the data structure relation is calculated layer by layer from bottom to top from the subtopic domain in the SG-CIM model by taking the SG-CIM model as a reference, the similarity between the objects in the two layers is the sum of the similarities between the objects in the sub-layers in the objects in the two layers, the similarity between the objects in the sub-layers is calculated based on the electric power word vector, and the objects in the two layers with the similarity larger than or equal to a preset threshold are fused into a new object to form the big data model.
2) Carrying out structural representation on each pair of object groups in the big data model, and measuring the rationality of each pair of object groups in the big data model according to the structural representation result to obtain a measurement result;
specifically, for each pair of object groups in the big data model
Carrying out structured representation, wherein h is a head entity, r is a relation, t is a tail entity, and a projection matrix is defined for the relation r
Projecting the vector of the head entity and the vector of the tail entity from the entity space to a subspace of the relation r, wherein the head entity vector in the subspace
Tail entity vector in subspace
Relationships in subspaces
And calculating the distance between the two entities, and measuring the rationality of each object group according to the distance between the entities.
3) Acquiring a head entity and a tail entity of the real object group, constructing a generator and a discriminator, performing counterstudy training on the generator based on the discriminator, the measurement result of the step 2) and the head entity and the tail entity of the real object group, generating a data structure relationship by using the trained generator, constructing a data model according to the data structure relationship, and completing the construction of the data model based on the generation of the counterstudy and the knowledge fusion.
The specific process is as follows:
31) generating a sample based on generating the countermeasure network delta;
inputting the entity pairs of the data structure relationship into a metric feature encoder to calculate an entity approximation representation (true data) of the data structure relationship, selecting the object group as a standard set
I.e. training, positive samples
Representing an approximation to a standard set, negative examples
Represents the score of a positive sample, far from the standard set
Setting a loss function for cosine similarity between the positive sample and the standard set and negative sample scoring similarity
Wherein, in the step (A),
in order for the parameters to be learned,
for the margin, the generator then replaces h or t in the real data samples with random sampling, generating the relationship embedding (dummy data). Randomly sampling h or t in the real data sample to replace to form disturbance, converting the disturbance by a longitude quantity characteristic encoder to be used as the input of a generator, and setting a loss function
Wherein, in the step (A),
as a feedback parameter for the discriminator,
representing a generating function formed by a measurement sample and random disturbance, finally, distinguishing real data from pseudo data by a discriminator, distributing correct relation types for the real data and the pseudo data, and setting a loss function
Wherein, in the step (A),
in the interest of expectation,
as a feedback parameter for the discriminator,
representing a generating function consisting of a measurement sample and a random perturbation,
the weighted vector sum of the measures is prefaced, z represents the random perturbation, and x represents the true sample.
And randomly sampling h or t in the real data sample, replacing and inputting the h or t into the generator, distinguishing a new object group output by the generator through artificial knowledge, reserving the forward object group, optimizing the sample set, substituting the sample set into the generation countermeasure learning iteration until a termination condition is met, wherein the termination condition is that no new retainable forward object group appears any more.
It should be noted that the data sample size is increased by generating the countermeasure method, meanwhile, with the addition of random disturbance, part of samples may be objects which are not used by the existing data model, positive samples are reserved, the process is repeated, iterative optimization is realized, human experience limitation is broken through to a certain extent, and the finally generated data model not only fuses the objects of the existing various data models, but also generates part of objects which are possibly useful and unused, so that the final data model covers the objects more comprehensively and has a certain development margin, and the science and the advancement of the design method are reflected.
The construction system of the data model comprises:
the generating module is used for fusing each layer of objects in the data structure relation of the power data model by taking the SG-CIM model as a reference to form a big data model;
the measurement module is used for carrying out structural representation on each pair of object groups in the big data model and then measuring the rationality of each pair of object groups in the big data model;
and the confrontation learning module is used for acquiring a head entity and a tail entity of the real object group, constructing a generator and a discriminator, carrying out confrontation learning training on the generator based on the measurement results of the discriminator and the measurement module and the head entity and the tail entity of the real object group, generating a data structure relationship by using the trained generator, constructing a data model according to the data structure relationship, and finishing the construction of the data model based on the generation of the confrontation learning and the knowledge fusion.
A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of the method of constructing the data model when executing the computer program.
A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method of constructing a data model.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (5)
1. A method for constructing a data model is characterized by comprising the following steps:
fusing each layer of objects in the data structure relation of the power data model by taking the SG-CIM model as a reference to form a big data model;
carrying out structural representation on each pair of object groups in the big data model, and measuring the rationality of each pair of object groups in the big data model according to the structural representation result to obtain a measurement result;
acquiring a head entity and a tail entity of a real object group, constructing a generator and a discriminator, carrying out countermeasure learning training on the generator based on the discriminator, the acquired measurement result and the head entity and the tail entity of the real object group, generating a data structure relationship by using the trained generator, and constructing a data model according to the data structure relationship generated by the generator;
taking the SG-CIM model as a reference, fusing objects of each layer in the data structure relationship of the electric power data model to form a big data model, wherein the concrete operation is as follows:
collecting each power data model, extracting a data structure relationship of each power data model, calculating similarity between objects of each layer in the data structure relationship layer by layer from bottom to top from a subtopic domain in the SG-CIM model by taking the SG-CIM model as a reference, and fusing two layers of objects with the similarity being more than or equal to a preset threshold value into a new object to form a big data model;
carrying out structural representation on each pair of object groups in the big data model, and carrying out measurement on the rationality of each pair of object groups in the big data model according to the result of the structural representation comprises the following specific operations:
carrying out structured representation on each pair of object groups (h, r, t) in the big data model, wherein h is a head entity, r is a relation, t is a tail entity, and a projection matrix M is defined for the relation rr∈Rd×kProjecting the vector of the head entity and the vector of the tail entity from the entity space to a subspace of the relation r, wherein the head entity vector h in the subspacer＝hMrTail entity vector t in subspacer＝tMrThe relation r in subspace is approximately equal to hr-trCalculating the distance between two entities, and measuring the rationality of each object group according to the distance between the entities;
the similarity between the two-layer objects is the sum of the similarities between the sub-layer objects in the two-layer objects;
calculating the similarity between sub-layer objects based on the electric power word vector;
in the counterstudy training process, random sampling replacement is carried out on a head entity and a tail entity of a real object group to form disturbance, and then the disturbance is converted by a longitude quantity characteristic encoder and then input into a generator.
2. The method of constructing a data model of claim 1, wherein the loss function for the counterlearning training is:
wherein the content of the first and second substances,
to expect, DφAs feedback parameter of the discriminator, Gθ(TrZ) represents a generating function consisting of a measurement sample and a random perturbation, TrThe weighted vector sum of the measures is prefaced, z represents the random perturbation, and x represents the true sample.
3. A system for building a data model, comprising:
the generating module is used for fusing each layer of objects in the data structure relation of the power data model by taking the SG-CIM model as a reference to form a big data model;
the measurement module is used for carrying out structural representation on each pair of object groups in the big data model, measuring the rationality of each pair of object groups in the big data model and obtaining a measurement result;
the confrontation learning module is used for acquiring a head entity and a tail entity of the real object group, constructing a generator and a discriminator, carrying out confrontation learning training on the generator based on the measurement results of the discriminator and the measurement module and the head entity and the tail entity of the real object group, generating a data structure relationship by using the trained generator, and constructing a data model according to the data structure relationship generated by the generator;
taking the SG-CIM model as a reference, fusing objects of each layer in the data structure relationship of the electric power data model to form a big data model, wherein the concrete operation is as follows:
collecting each power data model, extracting a data structure relationship of each power data model, calculating similarity between objects of each layer in the data structure relationship layer by layer from bottom to top from a subtopic domain in the SG-CIM model by taking the SG-CIM model as a reference, and fusing two layers of objects with the similarity being more than or equal to a preset threshold value into a new object to form a big data model;
carrying out structural representation on each pair of object groups in the big data model, and carrying out measurement on the rationality of each pair of object groups in the big data model according to the result of the structural representation comprises the following specific operations:
carrying out structured representation on each pair of object groups (h, r, t) in the big data model, wherein h is a head entity, r is a relation, t is a tail entity, and a projection matrix M is defined for the relation rr∈Rd×kProjecting the vector of the head entity and the vector of the tail entity from the entity space to the relationshipr, where the head entity vector h in the subspacer＝hMrTail entity vector t in subspacer＝tMrThe relation r in subspace is approximately equal to hr-trCalculating the distance between two entities, and measuring the rationality of each object group according to the distance between the entities;
the similarity between the two-layer objects is the sum of the similarities between the sub-layer objects in the two-layer objects;
calculating the similarity between sub-layer objects based on the electric power word vector;
in the counterstudy training process, random sampling replacement is carried out on a head entity and a tail entity of a real object group to form disturbance, and then the disturbance is converted by a longitude quantity characteristic encoder and then input into a generator.
4. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method of constructing a data model according to any one of claims 1 to 2 when executing the computer program.
5. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method of constructing a data model according to any one of claims 1 to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011366878.9A CN112181952B (en) | 2020-11-30 | 2020-11-30 | Method, system, device and storage medium for constructing data model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011366878.9A CN112181952B (en) | 2020-11-30 | 2020-11-30 | Method, system, device and storage medium for constructing data model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112181952A CN112181952A (en) | 2021-01-05 |
| CN112181952B true CN112181952B (en) | 2021-12-14 |
Family
ID=73918169
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011366878.9A Active CN112181952B (en) | 2020-11-30 | 2020-11-30 | Method, system, device and storage medium for constructing data model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112181952B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115952979A (en) * | 2022-12-13 | 2023-04-11 | 国家电网有限公司大数据中心 | Statistical big data logic model construction method and system based on data driving |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104008442A (en) * | 2014-05-07 | 2014-08-27 | 南京工程学院 | Power distribution network visual platform construction method based on SG-CIM |
| CN105631597A (en) * | 2015-12-31 | 2016-06-01 | 中国南方电网有限责任公司 | Construction method for transformer substation three-dimensional visual platform |
| KR102141784B1 (en) * | 2018-12-28 | 2020-08-06 | 주식회사 파워이십일 | System for managing ontology data of power grid |
| CN110674311A (en) * | 2019-09-05 | 2020-01-10 | 国家电网有限公司 | Knowledge graph-based power asset heterogeneous data fusion method |
| CN111046258B (en) * | 2019-11-25 | 2023-10-31 | 北京科东电力控制***有限责任公司 | Method, system and storage medium for extracting and integrally fusing multi-element basic information |
| CN111046189A (en) * | 2019-11-27 | 2020-04-21 | 广东电网有限责任公司 | Modeling method of power distribution network knowledge graph model |
| CN111241291B (en) * | 2020-04-24 | 2023-01-03 | 支付宝(杭州)信息技术有限公司 | Method and device for generating countermeasure sample by utilizing countermeasure generation network |
| CN111553587B (en) * | 2020-04-26 | 2023-04-18 | 中国电力科学研究院有限公司 | New energy scene generation method and system based on confrontation learning model |
-
2020
- 2020-11-30 CN CN202011366878.9A patent/CN112181952B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112181952A (en) | 2021-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114816909B (en) | Real-time log detection early warning method and system based on machine learning | |
| Chopade et al. | A framework for community detection in large networks using game-theoretic modeling | |
| CN112487033A (en) | Service visualization method and system for data flow and network topology construction | |
| CN110929951B (en) | Correlation analysis and prediction method for power grid alarm signal | |
| CN111967271A (en) | Analysis result generation method, device, equipment and readable storage medium | |
| CN112181952B (en) | Method, system, device and storage medium for constructing data model | |
| CN115375127A (en) | Intelligent refined management and control system for engineering construction | |
| CN116843083A (en) | Carbon emission prediction system and method based on hybrid neural network model | |
| CN110287237B (en) | Social network structure analysis based community data mining method | |
| Gong et al. | Evolutionary computation in China: A literature survey | |
| Cross et al. | Convergence of a hill-climbing genetic algorithm for graph matching | |
| CN111708919B (en) | Big data processing method and system | |
| CN112256884A (en) | Knowledge graph-based data asset library access method and device | |
| Caschera et al. | MONDE: a method for predicting social network dynamics and evolution | |
| Liu | Multidimensional nonlinear landscape planning based on parameter feature extraction and multimedia technology | |
| Aye et al. | Layman analytics system: A cloud-enabled system for data analytics workflow recommendation | |
| Narvaez-Teran et al. | Search trajectory networks applied to the cyclic bandwidth sum problem | |
| Mao et al. | Estimated time of arrival prediction via modeling the spatial-temporal interactions between links and crosses | |
| Daim et al. | Forecasting Technology Trends of The Gap Between Science and Technology: The Case of Software as a Service e-Commerce | |
| Cheng et al. | Using case-based reasoning to support Web service composition | |
| Rai et al. | Prediction of effort required to design software for smart city applications | |
| Cao et al. | Key technologies of big data and its development in intelligent ship | |
| Liu et al. | Web data mining algorithm based on cloud computing environment | |
| Illavenil et al. | Enhanced K-Means with Automated k Estimation and Outlier Detection Using Auto-Encoder Neural Network | |
| Heng et al. | Power Integrated Energy Collaborative Optimization System Based on PSO-PFCM Clustering Algorithm and Block Chain Technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |