Intelligent supervision early warning method and system based on supervision knowledge graph big data

Abstract

The invention provides an intelligent supervision early warning method and system based on supervision knowledge graph big data, and belongs to the technical field of intelligent supervision. The intelligent supervision and early warning system based on supervision knowledge graph big data comprises a knowledge management module, a graph control module, a data exchange module and an intelligent early warning module. The method is realized based on an intelligent supervision early warning system based on supervision knowledge graph big data. According to the method, supervision knowledge graph big data in a'entity-attribute-attribute value 'or'entity-relation-entity' ternary structure composition form is created, knowledge of different data sources is integrated into the same knowledge base, potential risks and problems are accurately identified by utilizing rich entity relations and attribute information in the knowledge base, and the risk and the problem are accurately identified. The early warning accuracy and reliability are improved; and meanwhile, personalized supervision knowledge recommendation is carried out by utilizing big data of the supervision knowledge graph, so that supervision personnel can be assisted to quickly obtain information related to supervision and early warning, and a more intelligent and timely decision can be made in the face of complex conditions.

Description

Technical Field

[0001] The present invention relates to the field of intelligent supervision technology, and in particular to an intelligent supervision early warning method and system based on supervision knowledge graph big data. Background Art

[0002] With the rapid development of information technology, informatization has been gradually realized. The widespread use of advanced technologies such as the Internet, big data, and cloud computing has greatly improved the efficiency and accuracy of supervision work. By building a complete data collection, storage, processing, and analysis system, various types of business data can be obtained in real time, and automatic analysis and early warning can be performed based on the preset supervision model, which reduces manual intervention and false alarm rates, and can detect potential risks and problems in the first place, providing strong support for supervision work.

[0003] However, in the process of deepening supervision work, a large number of resources such as risk prevention and control lists and supervision cases have been formed, which are scattered in various systems and platforms. They lack effective integration and unified management, are difficult to fully utilize and share, and the value potential of the data has not been fully tapped. Summary of the invention

[0004] The present invention provides an intelligent supervision and early warning method and system based on supervision knowledge graph big data to solve the problem that the data resources formed by the existing supervision work are scattered in various systems and platforms, lack effective integration and unified management, are difficult to fully utilize and share, and the value potential of the data cannot be fully utilized. The present invention creates a supervision knowledge graph big data in the form of a ternary structure of "entity-attribute-attribute value" or "entity-relationship-entity", integrates knowledge from different data sources into the same knowledge base, and uses the rich entity relationship and attribute information in the supervision knowledge graph big data to accurately identify potential risks and problems, and improve the accuracy and reliability of early warnings; at the same time, the present invention uses the supervision knowledge graph big data to perform personalized supervision knowledge recommendations, assisting supervisors to quickly obtain comprehensive information related to supervision and early warning, and make more wise and timely decisions when facing complex situations.

[0005] The technical solution adopted by the present invention is:

[0006] An intelligent supervision and early warning system based on supervision knowledge graph big data, including:

[0007] A knowledge management module, which is used to obtain and store original supervisory knowledge data, including supervisory model information, risk prevention and control lists, supervisory cases, and historical supervisory warning information;

[0008] A graph control module, which is used to perform entity recognition, relationship extraction and attribute extraction on the original supervised knowledge data, and then perform knowledge fusion to construct and store supervised knowledge graph big data in the form of a "entity-attribute-attribute value" or "entity-relationship-entity" ternary structure;

[0009] A data exchange module, which is used to obtain business data, and perform data cleaning, data integration and data transformation on the business data to form and store standard business item data;

[0010] An intelligent early warning module is used to create a supervisory early warning model for each of the business fields, configure supervisory early warning rules and early warning conditions; apply the supervisory early warning model corresponding to the corresponding business field to the standard business item data, and form supervisory early warning information for the standard business item data that meets the early warning conditions; then select supervisory recommended knowledge from the supervisory knowledge graph big data based on the supervisory early warning information, and push the supervisory early warning information and the corresponding supervisory recommended knowledge to the supervisor; the supervisory early warning information includes relevant information of the supervisory early warning model triggered by the early warning, a detailed description of the early warning, and the responsible department.

[0011] Furthermore, the knowledge management module also supports "query", "add", "delete" and "modify".

[0012] Furthermore, the graph control module also supports "query".

[0013] Based on the same inventive concept, the present invention also provides an intelligent supervision and early warning method based on supervision knowledge graph big data, which is implemented based on the aforementioned intelligent supervision and early warning system based on supervision knowledge graph big data, including the following steps:

[0014] Step S1, obtaining original data of supervisory knowledge, and performing entity recognition, relationship extraction and attribute extraction on the original data of supervisory knowledge, and then performing knowledge fusion to construct supervisory knowledge graph big data in the form of "entity-attribute-attribute value" or "entity-relationship-entity" ternary structure; the original data of supervisory knowledge includes supervisory model information, risk prevention and control list, supervisory cases, and historical supervisory warning information;

[0015] Step S2, acquiring business data, and performing data cleaning, data integration and data transformation on the business data to form standard business item data;

[0016] Step S3, creating a supervision warning model for each of the business fields, configuring supervision warning rules and warning conditions; applying the supervision warning model corresponding to the corresponding business field to the standard business item data, and forming supervision warning information for the standard business item data that meets the warning conditions; the supervision warning information includes relevant information of the supervision warning model triggered by the warning, a detailed warning description, and a responsible department;

[0017] Step S4, selecting supervisory recommended knowledge from the supervisory knowledge graph big data based on the supervisory warning information, and pushing the supervisory warning information and the corresponding supervisory recommended knowledge to the supervisor.

[0018] Furthermore, the specific process of knowledge fusion in step S1 includes reference resolution, entity disambiguation and entity alignment;

[0019] The reference resolution includes: summarizing different descriptions of the same thing into the same entity;

[0020] The entity disambiguation includes: distinguishing different meanings of entities with the same name;

[0021] The entity alignment includes: aligning the same entity to ensure that the same entity is identified as the same entity in different knowledge sources.

[0022] Furthermore, when the business data is acquired in step S2, the supervisory business data items conducted online are obtained from the business system and / or the data center through a data interface and / or a direct database connection, and the supervisory business data items conducted offline are obtained through manual entry or electronic spreadsheet summary submission.

[0023] Furthermore, the data cleaning in step S2 includes: deleting irrelevant data, duplicate data, and processing missing values ​​and outliers;

[0024] The data integration includes: merging data from different data sources into a consistent data storage, identifying and resolving homonymous and synonymous problems in different data sources, and identifying and reducing data redundancy;

[0025] The data transformation includes: data association, data splicing, and data calculation.

[0026] Furthermore, the specific process of selecting the supervisory recommended knowledge in step S4 includes: based on the supervisory warning information, screening out a warning-related information recommendation set from the supervisory knowledge graph big data; then, based on the warning-related information recommendation set, calculating the corresponding entity-related attributes in the supervisory knowledge graph big data to obtain the recommended information weight; finally, prioritizing the recommended information weights, and taking the corresponding supervisory knowledge part in the warning-related information recommendation set corresponding to the recommended information weights with the highest priority, the front ranking, or all the ranking as the supervisory recommended knowledge, and pushing the supervisory warning information and the supervisory recommended knowledge to the supervisor together to assist the supervisor in making supervisory judgments.

[0027] Furthermore, the weight of the recommended information in step S4 is calculated by analyzing the similarity between the entities in the supervisory warning information and the entities in the warning-related information recommendation set.

[0028] The beneficial effects of the present invention are:

[0029] The present invention provides an intelligent supervision and early warning method and system based on supervision knowledge graph big data, which creates a supervision knowledge graph big data in the form of "entity-attribute-attribute value" or "entity-relationship-entity" ternary structure by performing entity recognition, relationship extraction and attribute extraction on the original supervision knowledge data, and then performs knowledge fusion, and integrates knowledge from different data sources into the same knowledge base. At the same time, the present invention uses the rich entity relationship and attribute information in the supervision knowledge graph big data to accurately identify potential risks and problems, and improve the accuracy and reliability of early warning. Furthermore, the present invention uses the supervision knowledge graph big data to perform personalized supervision knowledge recommendation, and can provide deeper background information and solutions based on the supervision warning information association recommendation, which helps to better understand and respond to the warning situation, and through the structured knowledge representation framework, the supervisory personnel can quickly obtain comprehensive information related to the supervision warning, so as to make more wise and timely decisions when facing complex situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 This is a schematic diagram of the logical connection structure of the intelligent supervision and early warning system based on the supervision knowledge graph big data in Example 1.

[0032] Figure 2 Schematic diagram of the process of the intelligent supervision and early warning method based on the supervision knowledge graph big data in Example 2 DETAILED DESCRIPTION

[0033] The disclosure below provides many different embodiments or examples to realize different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention.

[0034] The embodiments of the invention are described in detail below with reference to the accompanying drawings.

[0035] Example 1

[0036] An intelligent supervision and early warning system based on supervision knowledge graph big data, its logical connection structure is as shown in the attached Figure 1 The intelligent supervision and early warning system based on the supervision knowledge graph big data includes: a knowledge management module, a graph control module, a data exchange module and an intelligent early warning module.

[0037] Among them, the knowledge management module is used to obtain and store original supervision knowledge data, and the original supervision knowledge data includes supervision model information, risk prevention and control list, supervision cases, and historical supervision warning information.

[0038] In this embodiment, the knowledge management module integrates the relevant information in the original supervisory knowledge data from different sources such as supervisory model information, risk prevention and control lists, supervisory cases, historical supervisory warning information, and different presentation forms such as text, attachments, and pictures to form a unified knowledge base. At the same time, users can dynamically update the information in the knowledge base by adding, deleting, modifying, etc. In addition, after the original supervisory knowledge data is classified by different categories and themes, users can quickly locate and query.

[0039] The graph control module is used to perform entity recognition, relationship extraction and attribute extraction on the supervised knowledge original data, and then perform knowledge fusion to construct and store supervised knowledge graph big data in the form of a "entity-attribute-attribute value" or "entity-relationship-entity" ternary structure.

[0040] In this embodiment, the graph control module is the core of the supervision knowledge graph big data management, and is responsible for controlling the generation, parameter configuration and associated query of the supervision knowledge graph big data. The graph control module first uses entity recognition, relationship extraction and attribute extraction, and then uses reference resolution, entity disambiguation and entity alignment and other technologies to fuse supervision knowledge information from different sources and different presentation forms, and dynamically controls the generation of supervision knowledge graph big data. At the same time, according to the parameters set by the user, the supervision knowledge graph big data is automatically constructed or updated to ensure the consistency, accuracy and timeliness of the graph. In addition, the graph control module can also provide complex graph query functions, support multi-hop queries and path discovery, and help users explore the deep connections between supervision knowledge. For example, view the same type of supervision warning information of the same position involved in a supervision case, view the supervision warning information involving major risks in a certain link in a certain field, and view the upstream and downstream supervision knowledge information of a certain link.

[0041] The data exchange module is used to obtain business data, and perform data cleaning, data integration and data transformation on the business data to form and store standard business item data.

[0042] In this embodiment, the data exchange module extracts the business data required for the supervision and early warning model analysis and comparison by configuring the data interface, data import, etc. When the specific business data is obtained, the online supervision business data items are obtained from the business system and / or the data center through the data interface and / or the database direct connection method, and the offline supervision business data items are obtained through manual entry or electronic form summary submission.

[0043] The intelligent early warning module is used to create a supervision early warning model for each of the business fields, configure supervision early warning rules and warning conditions; apply the supervision early warning model corresponding to the corresponding business field to the standard business matter data, form supervision early warning information for the standard business matter data that meets the warning conditions, and then use the supervision knowledge graph big data to recommend the supervision early warning information; the supervision early warning information includes relevant information of the supervision early warning model triggered by the warning, a detailed description of the warning, and the responsible department.

[0044] In this embodiment, the intelligent early warning module can configure the supervision early warning model around the key links of the key business areas of supervision, access the supervision-related business data, and generate early warning information for the business data that meets the early warning conditions through analysis and comparison. At the same time, the supervision early warning model, supervision field, responsible department, position and other information corresponding to the early warning information are identified, and the supervision knowledge query is performed in the supervision knowledge graph big data, and the supervision recommended knowledge of the same field, same position, same risk, same supervision model and other dimensions is selected, and pushed to the supervisor for processing together with the early warning information.

[0045] In this embodiment, by performing entity recognition, relationship extraction and attribute extraction on the original data of supervisory knowledge, and then performing knowledge fusion, a supervisory knowledge graph big data in the form of a ternary structure of "entity-attribute-attribute value" or "entity-relationship-entity" is created, and knowledge from different data sources is integrated into the same knowledge base. At the same time, the present invention uses the rich entity relationship and attribute information in the supervisory knowledge graph big data to accurately identify potential risks and problems, and improve the accuracy and reliability of early warnings. Furthermore, the present invention uses the supervisory knowledge graph big data to perform personalized supervisory knowledge recommendations, and can provide deeper background information and solutions based on supervisory early warning information association recommendations, which helps to better understand and respond to early warning situations, and through a structured knowledge representation framework, supervisors can quickly obtain comprehensive information related to supervisory early warnings, so as to make more wise and timely decisions when faced with complex situations.

[0046] Example 2

[0047] An intelligent supervision and early warning method based on supervision knowledge graph big data is implemented based on the intelligent supervision and early warning system based on supervision knowledge graph big data in Example 1, comprising the following steps:

[0048] Step S1, obtain the original data of supervisory knowledge, and perform entity recognition, relationship extraction and attribute extraction on the original data of supervisory knowledge, and then perform knowledge fusion to construct a supervisory knowledge graph big data in the form of a "entity-attribute-attribute value" or "entity-relationship-entity" ternary structure; the original data of supervisory knowledge includes supervisory model information, risk prevention and control list, supervisory cases, and historical supervisory warning information.

[0049] The specific process of knowledge fusion includes reference resolution, entity disambiguation and entity alignment. Reference resolution includes: summarizing different descriptions of the same thing as the same entity. Entity disambiguation includes: distinguishing different meanings of entities with the same name; entity alignment includes: aligning the same entity to ensure that the same entity is recognized as the same entity in different knowledge sources.

[0050] For example, when performing entity recognition on the supervision model information, entities such as model information, business fields, links, and positions can be obtained; when performing relationship extraction on the supervision model information, relationships or attributes such as the model belongs to a certain business link, belongs to a certain business field, and the model involves certain positions can be obtained. The entity relationship in the supervision model information can be expressed as: supervision model (name, level, model description, warning conditions)-(belongs to)->business link-(belongs to)->supervision field.

[0051] For example, when performing entity recognition on the risk prevention and control list, entities such as the list name, business field, and workflow can be obtained; when performing relationship extraction on the risk prevention and control list, relationships or attributes such as the risk belongs to a certain business field, involves a certain workflow, and involves certain positions can be obtained. The entity relationship in the risk prevention and control list can be expressed as: business field-(included)->exercise matters-(included)->risk (name, level, control measures)-(involved)->workflow-(involved)->position.

[0052] In this embodiment, the original data of the supervised knowledge is obtained, processed and converted into structured data, and then entity recognition, relationship extraction, and attribute extraction are performed. Then, data from different sources and in different presentation forms are integrated through steps such as reference resolution, entity disambiguation, and entity alignment to form a supervised knowledge graph big data with consistency and accuracy. Subsequently, the supervised knowledge graph big data is optimized continuously or regularly, and the supervised knowledge graph big data is updated according to the newly added data and information to maintain the timeliness and accuracy of the supervised knowledge graph big data.

[0053] Step S2, acquiring business data, and performing data cleaning, data integration and data transformation on the business data to form standard business item data.

[0054] Wherein, when the business data is acquired, the supervisory business data items conducted online are obtained from the business system and / or the data center through the data interface and / or the database direct connection method, and the supervisory business data items conducted offline are obtained through manual entry or electronic form summary submission. The data cleaning includes: deleting irrelevant data, duplicate data, and processing missing values ​​and abnormal values. The data integration includes: merging data from different data sources into a consistent data storage, and identifying and solving the problems of synonyms and synonyms in different data sources, and identifying and reducing data redundancy. The data transformation includes: data association, data splicing, and data calculation.

[0055] Step S3, create a supervision and early warning model for each of the business fields, configure supervision and early warning rules and early warning conditions; apply the supervision and early warning model corresponding to the corresponding business field to the standard business matter data, and form supervision and early warning information for the standard business matter data that meets the early warning conditions; the supervision and early warning information includes relevant information of the supervision and early warning model triggered by the early warning, a detailed description of the early warning, and the responsible department.

[0056] Step S4, selecting supervisory recommended knowledge from the supervisory knowledge graph big data based on the supervisory warning information, and pushing the supervisory warning information and the corresponding supervisory recommended knowledge to the supervisor.

[0057] Among them, the specific process of selecting the supervisory recommended knowledge includes: based on the supervisory warning information, filtering out a warning-related information recommendation set from the supervisory knowledge graph big data; then, based on the warning-related information recommendation set, calculating the entity-related attributes in the corresponding supervisory knowledge graph big data to obtain the recommended information weight; finally, prioritizing the recommended information weights, and taking the corresponding supervisory knowledge part in the warning-related information recommendation set corresponding to the highest priority, the front ranking, or all the recommended information weights after sorting as supervisory recommended knowledge, and pushing the supervisory warning information and the supervisory recommended knowledge to the supervisor together to assist the supervisor in supervision and judgment. Preferably, the recommended information weight is calculated by analyzing the similarity between the entities in the supervisory warning information and the entities in the warning-related information recommendation set.

[0058] For example, taking the supervision and early warning of the contract payment process as an example, the specific process of intelligent supervision and early warning includes:

[0059] (1) The warning condition of the supervision warning model is: the contract payment is not made according to the agreed time and proportion in the contract. The relevant business data such as the basic contract information, agreed payment information, and financial payment order information are directly obtained from the business system. After calculation and analysis, Contract A did not pay 20% of the agreed payment terms after contract signing, and the actual payment was 40%, which triggered the warning condition and generated supervision warning information; the supervision warning information includes: information on the contract management department, financial management department, relevant personnel, payment process nodes, etc.

[0060] (2) Based on the supervision warning information, the relevant entity information in the supervision knowledge graph big data is automatically selected through the relevant entity association, such as supervision model, supervision field, business process, job position, etc., to form a warning related information recommendation set;

[0061] (3) According to the similarity analysis weights of the entities in the supervision warning information and the entities in the warning-related information recommendation set, the weight of the recommended information is obtained; the recommended information is sorted by priority weight, and the supervision warning information and the corresponding supervision knowledge part of the warning-related information recommendation set corresponding to the highest ranking are pushed to the supervisor as the supervision recommendation knowledge. For example, the supervision warning information involves the Finance Department-Contract Payment Matters-Integrity Risk A, and the risk level is a major risk. What are the specific risk prevention and control measures, what are the work processes, and what positions and responsibilities are involved; the supervision warning information is similar to the supervision case in the Finance Department: when, what position, because of what, what discipline was violated, what punishment was received, etc.

[0062] In this embodiment, by performing entity recognition, relationship extraction and attribute extraction on the original data of supervisory knowledge, and then performing knowledge fusion, a supervisory knowledge graph big data in the form of a ternary structure of "entity-attribute-attribute value" or "entity-relationship-entity" is created, and knowledge from different data sources is integrated into the same knowledge base. At the same time, the present invention uses the rich entity relationship and attribute information in the supervisory knowledge graph big data to accurately identify potential risks and problems, and improve the accuracy and reliability of early warnings. Furthermore, the present invention uses the supervisory knowledge graph big data to perform personalized supervisory knowledge recommendations, and can provide deeper background information and solutions based on supervisory early warning information association recommendations, which helps to better understand and respond to early warning situations, and through a structured knowledge representation framework, supervisors can quickly obtain comprehensive information related to supervisory early warnings, so as to make more wise and timely decisions when faced with complex situations.

Claims

1. An intelligent supervision and early warning system based on supervision knowledge graph big data, characterized in that: include: A knowledge management module, which is used to obtain and store original supervisory knowledge data, including supervisory model information, risk prevention and control lists, supervisory cases, and historical supervisory warning information; A graph control module, which is used to perform entity recognition, relationship extraction and attribute extraction on the original supervisory knowledge data, and then perform knowledge fusion to construct and store supervisory knowledge graph big data in the form of a "entity-attribute-attribute value" or "entity-relationship-entity" ternary structure; A data exchange module, which is used to obtain business data, and perform data cleaning, data integration and data transformation on the business data to form and store standard business item data; An intelligent early warning module, the intelligent early warning module is used to create a supervision early warning model for each of the business fields, configure supervision early warning rules and early warning conditions; apply the supervision early warning model corresponding to the corresponding business field to the standard business item data, and form supervision early warning information for the standard business item data that meets the early warning conditions; Then, based on the supervision warning information, supervision recommendation knowledge is selected from the supervision knowledge graph big data, and the supervision warning information and the corresponding supervision recommendation knowledge are pushed to the supervisor; the supervision warning information includes relevant information of the supervision warning model triggered by the warning, a detailed description of the warning, and the responsible department.

2. According to claim 1, the intelligent supervision and early warning system based on supervision knowledge graph big data is characterized in that: The knowledge management module also supports "query", "add", "delete" and "modify".

3. According to claim 1, the intelligent supervision and early warning system based on supervision knowledge graph big data is characterized in that: The graph control module also supports "query".

4. An intelligent supervision and early warning method based on supervision knowledge graph big data, which is implemented based on the intelligent supervision and early warning system based on supervision knowledge graph big data as described in any one of claims 1 to 3, characterized in that: The following steps are involved: Step S1, obtaining original data of supervisory knowledge, and performing entity recognition, relationship extraction and attribute extraction on the original data of supervisory knowledge, and then performing knowledge fusion to construct supervisory knowledge graph big data in the form of "entity-attribute-attribute value" or "entity-relationship-entity" ternary structure; the original data of supervisory knowledge includes supervisory model information, risk prevention and control list, supervisory cases, and historical supervisory warning information; Step S2, obtaining business data, and performing data cleaning, data integration and data transformation on the business data to form standard business item data; Step S3, creating a supervision warning model for each of the business fields, configuring supervision warning rules and warning conditions; applying the supervision warning model corresponding to the corresponding business field to the standard business item data, and forming supervision warning information for the standard business item data that meets the warning conditions; The supervisory warning information includes relevant information of the supervisory warning model triggered by the warning, a detailed description of the warning, and the responsible department; Step S4, selecting supervisory recommended knowledge from the supervisory knowledge graph big data based on the supervisory warning information, and pushing the supervisory warning information and the corresponding supervisory recommended knowledge to the supervisor.

5. According to claim 4, the intelligent supervision and early warning method based on supervision knowledge graph big data is characterized in that: The specific process of knowledge fusion in step S1 includes reference resolution, entity disambiguation and entity alignment; The reference resolution includes: summarizing different descriptions of the same thing into the same entity; The entity disambiguation includes: distinguishing different meanings of entities with the same name; The entity alignment includes: aligning the same entity to ensure that the same entity is identified as the same entity in different knowledge sources.

6. The intelligent supervision and early warning method based on supervision knowledge graph big data according to claim 4 is characterized in that: When the business data is acquired in step S2, the supervisory business data items conducted online are obtained from the business system and / or the data center through the data interface and / or direct database connection, and the supervisory business data items conducted offline are obtained through manual entry or electronic spreadsheet summary submission.

7. The intelligent supervision and early warning method based on supervision knowledge graph big data according to claim 4 or 6 is characterized in that: The data cleaning in step S2 includes: deleting irrelevant data, duplicate data, and processing missing values ​​and outliers; The data integration includes: merging data from different data sources into a consistent data storage, identifying and resolving homonymous and synonymous problems in different data sources, and identifying and reducing data redundancy; The data transformation includes: data association, data splicing, and data calculation.

8. The intelligent supervision and early warning method based on supervision knowledge graph big data according to any one of claims 4 to 6 is characterized in that: The specific process of selecting the supervisory recommended knowledge in step S4 includes: based on the supervisory warning information, screening from the supervisory knowledge graph big data to obtain a warning-related information recommendation set; then, based on the warning-related information recommendation set, calculating the corresponding entity-related attributes in the supervisory knowledge graph big data to obtain the recommended information weight; finally, prioritizing the recommended information weights, and taking the corresponding supervisory knowledge part in the warning-related information recommendation set corresponding to the recommended information weights with the highest priority, the front ranking, or all the recommended information weights after the ranking as the supervisory recommended knowledge, and pushing the supervisory warning information and the supervisory recommended knowledge to the supervisor to assist the supervisor in making supervisory judgments.

9. The intelligent supervision and early warning method based on supervision knowledge graph big data according to claim 8 is characterized in that: The weight of the recommended information in step S4 is calculated by analyzing the similarity between the entities in the supervision warning information and the entities in the warning-related information recommendation set.

Images