Associated party identification method and system in associated transaction audit based on knowledge graph

Abstract

The invention discloses a knowledge graph-based associated party identification method and system in associated transaction auditing, and the method comprises the steps: firstly setting an associated party identification condition in the associated transaction auditing, and then constructing an associated party knowledge graph mode layer based on the associated party identification condition; establishing an associated party knowledge graph data layer according to the associated party knowledge graph mode layer and obtaining an associated party knowledge graph, and identifying an audited associated party in the associated party knowledge graph by adopting path reasoning according to an associated party identification condition; according to the method and the system, the function of quickly identifying the associated parties in the associated transactions by utilizing the expression and reasoning capabilities of the knowledge graph is realized, the continuous management and continuous auditing of the associated transactions are facilitated, and the fair and credibility of the associated transactions are improved; the method for describing things and concepts through entities, attributes and relationships not only provides rich inference and query capabilities, but also can store and manage threat related information, and is suitable for wide popularization and use.

Description

Technical Field

[0001] The present invention relates to the technical field of financial auditing, and in particular to a method and system for identifying related parties in related transaction auditing based on a knowledge graph. Background Art

[0002] With the frequent cases of failed audits of related-party transactions in recent years, related-party identification and related-party transactions have become one of the difficulties and focuses in the audit field; related-party transactions refer to all legal acts of transferring resources or obligations between a company and its related parties. A related party refers to a party that controls, jointly controls, or exerts significant influence on another party, as well as situations where two or more parties are controlled, jointly controlled, or significantly influenced by one party. Related parties may conduct related-party transactions through complex organizational structures and implicit deep relationships, which will greatly increase the difficulty and risk of audits, making how to identify related parties one of the key issues that need to be addressed in related-party transaction audits.

[0003] At present, most related-party transaction audits include major risk assessment, internal control assessment, related-party identification, related-party transaction identification and transaction price fairness testing. Existing related-party identification methods generally find it difficult to identify invisible related parties, which poses a great challenge to auditors in carrying out their work. This makes it impossible for the audit results to accurately identify related-party transaction risks and fraudulent behaviors, and thus it is impossible to effectively take corresponding measures to improve and strengthen the legality and effectiveness of related-party transactions, and thus it is impossible for enterprises to reduce transaction costs through related-party transactions, which also reduces the credibility of corporate financial statements. Therefore, it is necessary to design a related-party identification method and system in related-party transaction audits based on knowledge graphs. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies of the prior art and to better and effectively solve the problem that the existing related party identification methods are generally difficult to identify invisible related parties, which poses a great challenge to auditors in carrying out their work. This makes it impossible for the audit results to accurately identify the risks and fraudulent behaviors of related transactions, and thus it is impossible to effectively take corresponding measures to improve and strengthen the legality and effectiveness of related transactions, and thus it is impossible to complete the enterprise's reduction of transaction costs through related transactions, and at the same time it also reduces the credibility of the enterprise's financial statements. A method and system for identifying related parties in related transaction audits based on a knowledge graph is provided, which realizes the function of quickly identifying related parties in related transactions using the representation and reasoning capabilities of the knowledge graph, facilitates the continuous management and continuous audit of related transactions, and improves the fairness and credibility of related transactions. This method of describing things and concepts in the form of entities, attributes and relationships not only provides rich reasoning and query capabilities, but also can store and manage threat-related information.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is: A method and system for identifying related parties in related transaction audit based on knowledge graph, comprising the following steps: Step A: Setting the conditions for identifying related parties in the audit of related-party transactions; Step B, constructing a related party knowledge graph model layer based on the related party identification conditions; Step C, establishing a related party knowledge graph data layer according to the related party knowledge graph model layer to obtain a related party knowledge graph; Step D: Use path reasoning in the related party knowledge graph to identify the audited related parties based on the related party identification conditions, and complete the related party identification task in the related party transaction audit.

[0006] The aforementioned method for identifying related parties in related party transaction audit based on knowledge graph, step A, setting related party identification conditions in related party transaction audit, wherein the related party identification conditions include related natural persons and related companies, and the related party identification conditions are specifically set from the relationship between related natural persons and related companies based on the audit standards, and the specific steps are as follows, Step A1: Set related natural persons, including natural persons who directly or indirectly hold more than 5% of the shares of the listed company, directors, supervisors and senior managers of the listed company and their corresponding family members. The related natural persons also include legal persons, directors, supervisors and senior managers who directly or indirectly control the listed company; Step A2, setting up affiliated companies, said affiliated companies include legal persons that directly or indirectly control the listed company, legal persons other than the listed company and its holding subsidiaries that are directly or indirectly controlled, and legal persons that hold more than 5% of the shares of the listed company. Said affiliated companies also include subsidiaries of the listed company, joint ventures of the listed company and associated enterprises of the listed company.

[0007] The aforementioned method for identifying related parties in related party transaction audits based on knowledge graphs, step B, constructing a related party knowledge graph model layer based on related party identification conditions, wherein the related party knowledge graph model layer includes related party node types and related party relationship types, the related party node types include natural person nodes and company nodes, and the related party relationship types include personal investment relationships, close family relationships, key management relationships, parent company relationships, company control and influence relationships, personal control and influence relationships, and joint venture relationships.

[0008] In the aforementioned method for identifying related parties in related transaction audits based on knowledge graphs, the attributes of the natural person node include the person's name and identity card number, and the attributes of the company node include the company name, company registration address, business nature, company website and email address.

[0009] The aforementioned method for identifying related parties in related-party transaction audit based on knowledge graph, step C, establishes the related-party knowledge graph data layer according to the related-party knowledge graph model layer to obtain the related-party knowledge graph, and the specific steps are as follows: Step C1, obtaining information about the audited entity, wherein the audited entity information includes information about senior executives of the audited enterprise and investment holding information of the audited enterprise; Step C2, importing the audited object information into the related party knowledge graph data layer according to the related party knowledge graph model layer; Step C3, storing the associated party knowledge graph data layer into the knowledge graph database.

[0010] The aforementioned method for identifying related parties in related party transaction audit based on knowledge graph, step D, uses path reasoning in the related party knowledge graph to identify the audited related party according to the related party identification conditions, and completes the related party identification task in the related party transaction audit. The specific steps are as follows: Step D1, assume that there is a first natural person node that has a personal investment relationship with the audited object, then the second natural person node that has a close family relationship with the first natural person node is the related party of the audited object, and the first company node that has a personal control and influence relationship with the second natural person node is also the related party of the audited object; Step D2, assume that there is a second company node that has a parent company relationship with the audited object, then the third natural person node that has a personal control influence relationship with the second company node is the related party of the audited object, and the third company node that has a corporate control influence relationship with the second company node is also the related party of the audited object.

[0011] A system for identifying related parties in related transaction audits based on a knowledge graph, comprising an identification condition setting module, a knowledge graph pattern layer construction module, a knowledge graph data layer establishment module and a related party identification module, wherein the identification condition setting module is used to set related party identification conditions in related transaction audits; the knowledge graph pattern layer construction module is used to construct a related party knowledge graph pattern layer based on the related party identification conditions; the knowledge graph data layer establishment module is used to establish a related party knowledge graph data layer according to the related party knowledge graph pattern layer and obtain a related party knowledge graph; the related party identification module is used to identify audited related parties using path reasoning in the related party knowledge graph according to the related party identification conditions and complete the related party identification task in the related transaction audit.

[0012] The beneficial effects of the present invention are as follows: a method and system for identifying related parties in related transaction audits based on a knowledge graph of the present invention first sets related party identification conditions in related transaction audits, then constructs a related party knowledge graph model layer based on the related party identification conditions, then establishes a related party knowledge graph data layer according to the related party knowledge graph model layer and obtains a related party knowledge graph, and then uses path reasoning in the related party knowledge graph according to the related party identification conditions to identify the audited related parties, thereby completing the related party identification operation in the related transaction audit; effectively realizing that the method and system for identifying related parties in related transaction audits have the function of quickly identifying related parties in related transactions using the representation and reasoning capabilities of knowledge graphs, facilitating the continuous management and continuous auditing of related transactions, and improving the fairness and credibility of related transactions. This method of describing things and concepts in the form of entities, attributes and relationships not only provides rich reasoning and query capabilities, but also can store and manage threat-related information and provide more comprehensive information support for related transaction audits. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is an overall flow chart of a method for identifying related parties in related transaction audit based on knowledge graph of the present invention; Figure 2 It is a schematic diagram of the associated party knowledge graph model layer in an embodiment of the present invention. DETAILED DESCRIPTION

[0014] The present invention will be further described below in conjunction with the accompanying drawings.

[0015] like Figure 1 As shown, a method and system for identifying related parties in related transaction audit based on knowledge graph of the present invention includes the following steps: Step A, setting the related party identification conditions in the related party transaction audit, wherein the related party identification conditions include related natural persons and related companies, and the related party identification conditions are specifically set based on the relationship between related natural persons and related companies based on the audit standards. The specific steps are as follows: Step A1: Set related natural persons, including natural persons who directly or indirectly hold more than 5% of the shares of the listed company, directors, supervisors and senior managers of the listed company and their corresponding family members. The related natural persons also include legal persons, directors, supervisors and senior managers who directly or indirectly control the listed company; Step A2, setting up affiliated companies, said affiliated companies include legal persons that directly or indirectly control the listed company, legal persons other than the listed company and its holding subsidiaries that are directly or indirectly controlled, and legal persons that hold more than 5% of the shares of the listed company. Said affiliated companies also include subsidiaries of the listed company, joint ventures of the listed company and associated enterprises of the listed company.

[0016] Step B, constructing an associated party knowledge graph model layer based on the associated party identification conditions, wherein the associated party knowledge graph model layer includes associated party node types and associated party relationship types, the associated party node types include natural person nodes and company nodes, and the associated party relationship types include personal investment relationships, close family relationships, key management relationships, parent company relationships, company control and influence relationships, personal control and influence relationships, and joint venture relationships.

[0017] The attributes of the natural person node include the person's name and identity card number, and the attributes of the company node include the company name, company registration address, business nature, company website and email address.

[0018] Step C: Establish the associated party knowledge graph data layer according to the associated party knowledge graph model layer to obtain the associated party knowledge graph. The specific steps are as follows: Step C1, obtaining information about the audited entity, wherein the audited entity information includes information about senior executives of the audited enterprise and investment holding information of the audited enterprise; Step C2, importing the audited object information into the related party knowledge graph data layer according to the related party knowledge graph model layer; Step C3, storing the associated party knowledge graph data layer into the knowledge graph database.

[0019] Step D: According to the related party identification conditions, path reasoning is used in the related party knowledge graph to identify the audited related parties, and the related party identification task in the related party transaction audit is completed. The specific steps are as follows: Step D1, assume that there is a first natural person node that has a personal investment relationship with the audited object, then the second natural person node that has a close family relationship with the first natural person node is the related party of the audited object, and the first company node that has a personal control and influence relationship with the second natural person node is also the related party of the audited object; Step D2, assume that there is a second company node that has a parent company relationship with the audited object, then the third natural person node that has a personal control influence relationship with the second company node is the related party of the audited object, and the third company node that has a corporate control influence relationship with the second company node is also the related party of the audited object.

[0020] A system for identifying related parties in related transaction audits based on a knowledge graph, comprising an identification condition setting module, a knowledge graph pattern layer construction module, a knowledge graph data layer establishment module and a related party identification module, wherein the identification condition setting module is used to set related party identification conditions in related transaction audits; the knowledge graph pattern layer construction module is used to construct a related party knowledge graph pattern layer based on the related party identification conditions; the knowledge graph data layer establishment module is used to establish a related party knowledge graph data layer according to the related party knowledge graph pattern layer and obtain a related party knowledge graph; the related party identification module is used to identify audited related parties using path reasoning in the related party knowledge graph according to the related party identification conditions and complete the related party identification task in the related transaction audit.

[0021] In order to better illustrate the use effect of the present invention, a specific embodiment of the present invention is described below: like Figure 2 As shown, this embodiment first sets the related party identification conditions in the related party transaction audit according to step A, then builds the related party knowledge graph model layer according to step B, then establishes the related party knowledge graph data layer and obtains the related party knowledge graph according to step C, and then uses path reasoning in the related party knowledge graph according to the related party identification conditions in step D to identify the audited related party, thereby completing the related party identification operation in the related party transaction audit; This embodiment adopts the Cypher language used in the graph database in the knowledge graph of the method of the present invention to express path reasoning more naturally, and the built-in graph structure and relationship labels make reasoning complex paths more intuitive, while traditional relational databases need to explicitly process the connection table JOIN, which not only has lengthy code, but also relies on manual connection operations in multi-layer relationship reasoning, so that path reasoning needs to maintain multiple intermediate tables and conditions, which makes it more complicated; therefore, the method of the present invention has higher efficiency in identifying related parties.

[0022] In summary, the present invention provides a method and system for identifying related parties in related transaction audits based on a knowledge graph. First, related party identification conditions in related transaction audits are set, and then a related party knowledge graph model layer is constructed based on the related party identification conditions. Then, a related party knowledge graph data layer is established according to the related party knowledge graph model layer and a related party knowledge graph is obtained. Then, path reasoning is used in the related party knowledge graph according to the related party identification conditions to identify the audited related parties, and the related party identification operation in the related transaction audit is completed. The method and system for identifying related parties in related transaction audits effectively realize the function of quickly identifying related parties in related transactions by using the representation and reasoning capabilities of knowledge graphs, which facilitates the continuous management and continuous audit of related transactions, and improves the fairness and credibility of related transactions. This method of describing things and concepts in the form of entities, attributes and relationships not only provides rich reasoning and query capabilities, but also can store and manage threat-related information and provide more comprehensive information support for related transaction audits.

[0023] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for identifying related parties in related-party transaction audit based on knowledge graph, characterized by: The following steps are included: Step A: Setting the conditions for identifying related parties in the audit of related-party transactions; Step B, constructing a related party knowledge graph model layer based on the related party identification conditions; Step C, establishing a related party knowledge graph data layer according to the related party knowledge graph model layer to obtain a related party knowledge graph; Step D: Use path reasoning in the related party knowledge graph to identify the audited related parties based on the related party identification conditions, and complete the related party identification task in the related party transaction audit.

2. According to claim 1, a method for identifying related parties in related transaction audit based on knowledge graph, characterized in that: Step A, setting the related party identification conditions in the related party transaction audit, wherein the related party identification conditions include related natural persons and related companies, and the related party identification conditions are specifically set based on the relationship between related natural persons and related companies based on the audit standards. The specific steps are as follows: Step A1: Set related natural persons, including natural persons who directly or indirectly hold more than 5% of the shares of the listed company, directors, supervisors and senior managers of the listed company and their corresponding family members. The related natural persons also include legal persons, directors, supervisors and senior managers who directly or indirectly control the listed company; Step A2, setting up affiliated companies, said affiliated companies include legal persons that directly or indirectly control the listed company, legal persons other than the listed company and its holding subsidiaries that are directly or indirectly controlled, and legal persons that hold more than 5% of the shares of the listed company. Said affiliated companies also include subsidiaries of the listed company, joint ventures of the listed company and associated enterprises of the listed company.

3. According to claim 2, a method for identifying related parties in related transaction audit based on knowledge graph, characterized in that: Step B, constructing an associated party knowledge graph model layer based on the associated party identification conditions, wherein the associated party knowledge graph model layer includes associated party node types and associated party relationship types, the associated party node types include natural person nodes and company nodes, and the associated party relationship types include personal investment relationships, close family relationships, key management relationships, parent company relationships, company control and influence relationships, personal control and influence relationships, and joint venture relationships.

4. According to claim 3, a method for identifying related parties in related transaction audit based on knowledge graph is characterized in that: The attributes of the natural person node include the person's name and identity card number, and the attributes of the company node include the company name, company registration address, business nature, company website and email address.

5. According to claim 4, a method for identifying related parties in related transaction audit based on knowledge graph is characterized in that: Step C: Establish the associated party knowledge graph data layer according to the associated party knowledge graph model layer to obtain the associated party knowledge graph. The specific steps are as follows: Step C1, obtaining information about the audited entity, wherein the audited entity information includes information about senior executives of the audited enterprise and investment holding information of the audited enterprise; Step C2, importing the audited object information into the related party knowledge graph data layer according to the related party knowledge graph model layer; Step C3, storing the associated party knowledge graph data layer into the knowledge graph database.

6. According to claim 5, a method for identifying related parties in related transaction audit based on knowledge graph is characterized in that: Step D: According to the related party identification conditions, path reasoning is used in the related party knowledge graph to identify the audited related parties, and the related party identification task in the related party transaction audit is completed. The specific steps are as follows: Step D1, assume that there is a first natural person node that has a personal investment relationship with the audited object, then the second natural person node that has a close family relationship with the first natural person node is the related party of the audited object, and the first company node that has a personal control and influence relationship with the second natural person node is also the related party of the audited object; Step D2, assume that there is a second company node that has a parent company relationship with the audited object, then the third natural person node that has a personal control influence relationship with the second company node is the related party of the audited object, and the third company node that has a corporate control influence relationship with the second company node is also the related party of the audited object.

7. A system for identifying related parties in related-party transaction audits based on knowledge graphs, wherein the system for identifying related parties in related-party transaction audits is based on the method for identifying related parties in related-party transaction audits according to any one of claims 1 to 6, and is characterized in that: It includes an identification condition setting module, a knowledge graph model layer construction module, a knowledge graph data layer establishment module and a related party identification module. The identification condition setting module is used to set the related party identification conditions in the related transaction audit; The knowledge graph model layer construction module is used to construct the associated party knowledge graph model layer based on the associated party identification condition; The knowledge graph data layer establishment module is used to establish the associated party knowledge graph data layer according to the associated party knowledge graph model layer and obtain the associated party knowledge graph; The related party identification module is used to identify the audited related parties by using path reasoning in the related party knowledge graph according to the related party identification conditions and complete the related party identification task in the related party transaction audit.

Images