Association relation determination method and device of associated persons

Abstract

The invention discloses a method and a device for determining an association relationship of associated persons. The method comprises the following steps of: determining an association relationship hierarchy graph of the associated persons; performing character recognition processing on the received related party supervision information to obtain a specified association relation element; identifying existing association elements of an existing association relationship corresponding to the supervision organization identifier of the associated person; the existing association elements are matched with the specified association relation elements; taking the existing associated elements which are not successfully matched as to-be-corrected associated elements which do not conform to a specified associated party affirmation rule; according to the to-be-corrected associated elements, correcting one or more supervision organization identifiers of the associated person; and determining a hierarchical path step length between the associated person and the associated party corresponding to each corrected associated party supervision identifier, and modifying to obtain an association relationship hierarchical graph of the associated person. The method is used for improving the recognition accuracy and recognition efficiency of the related person supervision identifier and avoiding data redundancy.

Description

Technical Field

[0001] The present invention relates to the field of financial supervision technology, and in particular to a method and device for determining the association relationship of an associated person. Background Art

[0002] This section is intended to provide a background or context for the presented embodiments of the invention. No description herein is admitted to be prior art by virtue of its inclusion in this section.

[0003] With the increasing complexity of business operations and the continuous improvement of regulatory requirements, it is crucial to accurately identify related parties and their regulatory identification. The accurate identification of related parties and the correct calculation of regulatory identification are of great significance to the compliance operation of enterprises and the accurate reporting of regulatory data. The current method of determining the relationship between related parties requires the identification of the regulatory agency of the identified related party, and then a related party map can be constructed. The related party of a node in the map can obtain the relationship between the related parties at each level in the wire and their superior related parties through a virtual wire. For related parties with specific relationship combinations that are not within the scope of the regulatory related party identification rules, it is necessary to manually search the relationship between each related party on the transmission wire one by one and layer by layer in the related party map, make a judgment, and manually modify the regulatory agency identification of the related party.

[0004] The existing technical solutions have the following problems:

[0005] 1. The problem of regulatory identification transmission errors: The regulatory identification of the superior related party can be infinitely transmitted downward through the combination of related relationships, but because the specific combination of related relationships is not within the scope of the regulatory related party identification rules, it will lead to errors in the calculation results of the regulatory identification of related party A. For example, in some cases, some entities that should not fall within the definition of related parties may be mistakenly identified as related parties based on the existing transmission model.

[0006] 2. Difficulty in manual investigation: For related parties with specific association relationships that are not within the scope of the regulatory related party identification rules, there are many difficulties in manual investigation and correction. It is necessary to manually search for the association relationships of each related party on the transmission line one by one and layer by layer in the related party map for identification and modification. The workload is huge, time-consuming, difficult and easy to miss, resulting in errors in the calculation of related party supervision marks and failure to correct them in time, which in turn leads to misreporting of related parties and transaction data involving related parties, which is easy to trigger regulatory penalties. Summary of the invention

[0007] The embodiment of the present invention provides a method for determining the association relationship of an associated person, which is used to improve the recognition accuracy and recognition efficiency of the associated person supervision identification, simplify the association relationship of the associated person, avoid data redundancy, and ensure the accuracy and timeliness of the supervision reporting data. The method includes:

[0008] Determine a hierarchical map of the associated relationships of the associated person; the hierarchical map of the associated relationships is marked with one or more regulatory agency identifiers of the associated person; the regulatory agency identifiers are used to characterize the associated elements of the associated person; the associated elements include each associated party of the associated person, and the hierarchy and associated relationship between the associated person and each of its associated parties;

[0009] Performing text recognition processing on the received related party supervision information to obtain designated related relationship elements; the related party supervision information is used to describe information that complies with designated related party identification rules;

[0010] Identify the existing association elements of the existing association relationship corresponding to the regulatory agency identification of the associated person; match the existing association elements with the designated association relationship elements; and treat the existing association elements that have not been successfully matched as association elements to be corrected that do not comply with the designated associated party identification rules;

[0011] According to the association elements to be amended, one or more regulatory agency identifiers of the associated person are amended to obtain amended regulatory identifiers of the associated party;

[0012] Determine the hierarchical path step between the related party and each related party corresponding to the revised related party regulatory identification;

[0013] The association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical map are deleted to obtain a revised association relationship hierarchical map of the associated persons.

[0014] The embodiment of the present invention also provides a device for determining the association relationship of an associated person, which is used to improve the recognition accuracy and recognition efficiency of the associated person supervision identification, simplify the association relationship of the associated person, avoid data redundancy, and ensure the accuracy and timeliness of the supervision reporting data. The device includes:

[0015] A graph determination module is used to determine a hierarchical graph of association relationships of an associated person; the hierarchical graph of association relationships is marked with one or more regulatory agency identifiers of the associated person; the regulatory agency identifiers are used to characterize association elements of the associated person; the association elements include each associated party of the associated person, and the hierarchy and association relationship between the associated person and each of its associated parties;

[0016] A text recognition processing module is used to perform text recognition processing on the received related party supervision information to obtain the specified association relationship elements; the related party supervision information is used to describe the information that meets the specified related party identification rules;

[0017] An existing association element determination module is used to identify the existing association elements of the existing association relationship corresponding to the regulatory agency identifier of the associated person; match the existing association elements with the specified association relationship elements; and treat the existing association elements that have not been successfully matched as association elements to be corrected that do not comply with the specified associated party identification rules;

[0018] A regulatory agency identification correction module, used to correct one or more regulatory agency identifications of the associated person according to the associated elements to be corrected, to obtain a corrected regulatory identification of the associated party;

[0019] A hierarchical path step length determination module, used to determine the hierarchical path step length between the associated person and each associated party corresponding to the revised associated party supervision identifier;

[0020] The graph correction module is used to delete the association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical graph, so as to obtain a corrected association relationship hierarchical graph of the associated persons.

[0021] An embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for determining the association relationship of associated persons when executing the computer program.

[0022] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for determining the association relationship of associated persons is implemented.

[0023] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the method for determining the association relationship of associated persons is implemented.

[0024] In an embodiment of the present invention, a hierarchical map of association relationships of an associated person is determined; one or more regulatory agency identifiers of the associated person are marked in the hierarchical map of association relationships; the regulatory agency identifier is used to characterize the association elements of the associated person; the association elements include each associated party of the associated person, and the hierarchy and association relationship between the associated person and each of its associated parties; the received associated party regulatory information is subjected to text recognition processing to obtain a specified association relationship element; the associated party regulatory information is used to describe information that complies with the specified associated party identification rule; the existing association elements of the existing association relationship corresponding to the regulatory agency identifier of the associated person are identified; the existing association elements are matched with the specified association relationship elements; the existing association elements that are not successfully matched are used as the association elements to be corrected that do not comply with the specified associated party identification rule; according to the association elements to be corrected, one or more regulatory agency identifiers of the associated person are corrected to obtain the corrected associated party regulatory identifier; the hierarchical path step between the associated person and the associated party corresponding to each corrected associated party regulatory identifier is determined; the association corresponding to the hierarchical path step that exceeds a preset value in the hierarchical map of association relationships is deleted to obtain the corrected hierarchical map of association relationships of the associated person, thereby The system can identify the related elements to be corrected that do not meet the specified related party identification rules and correct them to ensure the accuracy of the regulatory identification, solving the problem that the specific combination of related relationships is not within the scope of the regulatory identification rules under the existing technology, resulting in incorrect calculation of the regulatory identification; the received related party regulatory information is processed by text recognition in an automated manner, the existing related elements are automatically matched, and the elements that have not been successfully matched are corrected, reducing manual intervention, improving efficiency and accuracy, and solving the problem that the existing technology requires manual search and modification of the related relationships in the related party map one by one and layer by layer, which is labor-intensive, time-consuming and easy to miss; by determining the hierarchical path step and deleting the path that exceeds the preset value, reducing redundant paths, simplifying the relationship map, improving the performance and maintainability of the relationship system, solving the problem that the data is redundant due to the geometric growth of data and the complexity of the relationship as the transmission path becomes longer and the requirements of the regulatory mark change under the existing technology; by dynamically adjusting and correcting the regulatory mark, it can quickly respond to changes in regulatory rules to ensure the accuracy and timeliness of regulatory reporting data, solving the problem that it is difficult to quickly adapt to changes in regulatory rules under the existing technology, resulting in the inability to update regulatory marks in a timely manner and affecting the quality of reported data. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0026] Figure 1 A schematic diagram of a flow chart of a method for determining an association relationship of an associated person in an embodiment of the present invention;

[0027] Figure 2 A specific example diagram of a method for determining an associated relationship of associated persons in an embodiment of the present invention;

[0028] Figure 3 A specific example diagram of a method for determining an associated relationship of associated persons in an embodiment of the present invention;

[0029] Figure 4 A specific example diagram of a method for determining an associated relationship of associated persons in an embodiment of the present invention;

[0030] Figure 5 A specific example diagram of a method for determining an associated relationship of associated persons in an embodiment of the present invention;

[0031] Figure 6 It is a structural schematic diagram of a device for determining an association relationship of associated persons in an embodiment of the present invention;

[0032] Figure 7 A schematic diagram of a computer device for determining an association relationship of associated persons in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0034] The term "and / or" herein only describes an association relationship, indicating that three relationships may exist. For example, A and / or B may represent the following three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set consisting of A, B, and C.

[0035] In the description of this specification, the terms "include", "including", "have", "contain", etc. are all open terms, which mean including but not limited to. The descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", etc. mean that the specific features, structures or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. The order of steps involved in each embodiment is used to schematically illustrate the implementation of the present application, and the order of steps is not limited and can be appropriately adjusted as needed.

[0036] The acquisition, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of national laws and regulations. The information collected in this application is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure, and application of relevant data are in compliance with the relevant laws, regulations, and standards of relevant countries and regions, and necessary confidentiality measures are taken, which do not violate public order and good customs, and provide corresponding operation entrances for users to choose to authorize or refuse. In addition, this application provides users with corresponding operation entrances for users to choose to agree or refuse the results of automated decision-making. If the user chooses to refuse, they can enter the expert decision-making process.

[0037] It should be noted that in the embodiments of the present application, some existing solutions in the industry such as certain software, components, models, etc. may be mentioned. For example, some existing software tools, components, algorithm models or other well-known solutions in other technical fields may be cited. They should be considered as exemplary, and their purpose is only to illustrate the feasibility of the implementation of the technical solution of the present application. These references should be understood as typical examples, and their core purpose is to explain and verify the rationality and feasibility of the implementation of the technical solution proposed in the present application. However, it does not mean that the applicant has or will inevitably use the solution. Such references do not imply that the applicant has actually adopted these existing solutions, or will inevitably adopt these methods in the future during its technical implementation. In other words, these references only serve to illustrate the nature of the description, to help understand the connection and transcendence between the innovation of the present application and the prior art, and do not constitute an endorsement or reliance statement on a specific prior art product.

[0038] The existing methods for identifying related parties are generally as follows: the identification of the regulatory agency to which the pre-identified related party A belongs needs to be determined based on the relationship between A and the superior related party and the identification of the regulatory agency to which A's superior related party belongs. Therefore, a related party graph can be constructed through the above model. A related party A at a certain node in the graph can obtain the relationship between the related parties at each level in the wire and its superior related party through a virtual wire. These relationships form an association combination in sequence with the relationship of related party A. A certain regulatory mark of related party A is transmitted downward step by step by the first-level related party of the association combination through the virtual wire formed by the association combination. In other words, it can be said that the identification of the regulatory agency to which related party A belongs is obtained by the model through the association combination calculation. In the above existing model, the regulatory identification of the superior related party can be infinitely transmitted downward through the association combination, but because the specific association combination is not within the scope of the regulatory related party identification rules, it will cause errors in the calculation results of the regulatory identification of related party A.

[0039] At present, for the related parties with specific association combinations that are not within the scope of the regulatory related party identification rules, it is necessary to manually search for the associations of each related party on the transmission line one by one and layer by layer in the related party map to make judgments. For example, the child B of the direct related person A needs to be identified as a related party. According to the existing regulatory mark transmission model, A's grandchild C can also be identified as a related party by choosing the association with B (child) to inherit B's regulatory mark. However, under a certain regulatory rule definition, the grandchild C of the direct related person A does not fall within the scope of the definition of related parties. This method requires a large amount of related party data to be checked, and the associations are complicated. It is difficult to manually enumerate all specific association combinations; this results in errors in the calculation of related party regulatory marks and the inability to correct them in a timely manner, which in turn leads to misreporting of related parties and transaction data involving related parties, which can easily lead to regulatory penalties.

[0040] In order to solve the above problems, the embodiment of the present invention provides a method for determining the association relationship of an associated person, which is used to improve the recognition accuracy and efficiency of the associated person supervision identification, simplify the association relationship of the associated person, avoid data redundancy, and ensure the accuracy and timeliness of the supervision reporting data, see Figure 1 A flowchart of a method for determining an associated relationship of an associated person is shown, and the method may include:

[0041] Step 101: Determine a hierarchical map of association relationships of an associated person; the hierarchical map of association relationships is marked with one or more regulatory agency identifiers of the associated person; the regulatory agency identifiers are used to characterize association elements of the associated person; the association elements include each associated party of the associated person, and the hierarchy and association relationship between the associated person and each of its associated parties;

[0042] Step 102: Perform text recognition processing on the received related party supervision information to obtain designated related relationship elements; the related party supervision information is used to describe information that meets the designated related party identification rules;

[0043] Step 103: Identify the existing association elements of the existing association relationship corresponding to the regulatory agency identifier of the associated person; match the existing association elements with the designated association relationship elements; and treat the existing association elements that have not been successfully matched as association elements to be corrected that do not comply with the designated associated party identification rules;

[0044] Step 104: According to the association elements to be amended, one or more regulatory agency identifiers of the associated person are amended to obtain amended regulatory identifiers of the associated party;

[0045] Step 105: Determine the hierarchical path step between the associated person and each associated party corresponding to the revised associated party supervision identifier;

[0046] Step 106: Delete the association relationships corresponding to the hierarchical path steps exceeding the preset value in the association relationship hierarchical map, and obtain a revised association relationship hierarchical map of the associated persons.

[0047] In an embodiment of the present invention, a hierarchical map of association relationships of an associated person is determined; one or more regulatory agency identifiers of the associated person are marked in the hierarchical map of association relationships; the regulatory agency identifier is used to characterize the association elements of the associated person; the association elements include each associated party of the associated person, and the hierarchy and association relationship between the associated person and each of its associated parties; the received associated party regulatory information is subjected to text recognition processing to obtain a specified association relationship element; the associated party regulatory information is used to describe information that complies with the specified associated party identification rule; the existing association elements of the existing association relationship corresponding to the regulatory agency identifier of the associated person are identified; the existing association elements are matched with the specified association relationship elements; the existing association elements that are not successfully matched are used as the association elements to be corrected that do not comply with the specified associated party identification rule; according to the association elements to be corrected, one or more regulatory agency identifiers of the associated person are corrected to obtain the corrected associated party regulatory identifier; the hierarchical path step between the associated person and the associated party corresponding to each corrected associated party regulatory identifier is determined; the association corresponding to the hierarchical path step that exceeds a preset value in the hierarchical map of association relationships is deleted to obtain the corrected hierarchical map of association relationships of the associated person, thereby The system can identify the related elements to be corrected that do not meet the specified related party identification rules and correct them to ensure the accuracy of the regulatory identification, solving the problem that the specific combination of related relationships is not within the scope of the regulatory identification rules under the existing technology, resulting in incorrect calculation of the regulatory identification; the received related party regulatory information is processed by text recognition in an automated manner, the existing related elements are automatically matched, and the elements that have not been successfully matched are corrected, reducing manual intervention, improving efficiency and accuracy, and solving the problem that the existing technology requires manual search and modification of the related relationships in the related party map one by one and layer by layer, which is labor-intensive, time-consuming and easy to miss; by determining the hierarchical path step and deleting the path that exceeds the preset value, reducing redundant paths, simplifying the relationship map, improving the performance and maintainability of the relationship system, solving the problem that the data is redundant due to the geometric growth of data and the complexity of the relationship as the transmission path becomes longer and the requirements of the regulatory mark change under the existing technology; by dynamically adjusting and correcting the regulatory mark, it can quickly respond to changes in regulatory rules to ensure the accuracy and timeliness of regulatory reporting data, solving the problem that it is difficult to quickly adapt to changes in regulatory rules under the existing technology, resulting in the inability to update regulatory marks in a timely manner and affecting the quality of reported data.

[0048] The present invention solidifies specific association relationship combinations that are not within the scope of the regulatory related party identification rules by establishing a correction model. The system automatically completes the retrieval of association relationships at all levels, corrects the corresponding regulatory identification for eligible related parties, supplements and improves the existing related party regulatory identification calculation model, ensures the accuracy of related party regulatory identification calculation, reduces manual workload, improves the accuracy of related party regulatory identification identification, reduces management costs, and thereby improves the quality of submitted regulatory data.

[0049] In the specific implementation, step 101 is first performed: determining a hierarchical map of associated relationships of an associated person; the hierarchical map of associated relationships is marked with one or more regulatory agency identifiers of the associated person; the regulatory agency identifiers are used to characterize associated elements of the associated person; the associated elements include each associated party of the associated person, and the hierarchy and associated relationship between the associated person and each of its associated parties.

[0050] In the embodiment, the construction of the relationship hierarchy map is based on multiple data sources. First, the basic related party data of the related party is obtained from the related party registration information within the enterprise, including the name and type of the related party (such as legal person, natural person, etc.) and the preliminary recorded relationship information. Secondly, the transaction record data of the enterprise is collected, and potential relationships are mined by analyzing the subject information of the two parties to the transaction and the nature, amount, frequency and other characteristics of the transaction. In addition, it is necessary to refer to the relevant regulatory information released by the regulatory authorities, such as the standards and special requirements for the identification of related parties in the industry regulatory rules, to ensure that the map construction complies with regulatory specifications.

[0051] Integrate and process all types of collected data. Use data cleaning technology to remove noise and erroneous information in the data, for example, correct spelling errors in the names of related parties, and unify data formats (such as date format, amount units, etc.). At the same time, classify and annotate the data for subsequent map construction. For example, classify and annotate the relationship by type (such as equity relationship, transaction relationship, kinship relationship, etc.).

[0052] With the associated person as the central node, the graph is gradually constructed according to the hierarchical relationship of its associated parties. First, the direct associated party is determined, that is, the subject with a direct associated relationship with the associated person, and it is used as the first-level node of the graph, and the specific type of relationship between it and the associated person is clarified (for example, if the associated person holds the equity of a legal person's direct associated party, it is marked as an equity relationship). Then, for each directly associated party, further search for its own associated parties as the second-level nodes of the graph, and so on, gradually construct a multi-level hierarchical graph of associated relationships.

[0053] In the process of building the graph, graph theory-related algorithms and techniques are used to optimize the graph structure. For example, the minimum spanning tree algorithm is used to ensure that the connection relationship between nodes in the graph is the simplest and can fully reflect the association relationship, avoiding redundant connection paths. At the same time, graph database technology is used to store and manage graph data, so as to efficiently query and update node information and connection relationships in the graph.

[0054] The one or more regulatory agency identifiers marked in the relationship hierarchy map come from many sources. On the one hand, the regulatory agency applicable to different types of related parties is determined based on the official documents and regulations issued by the regulatory agency. For example, certain related parties in the financial industry may be supervised by specific financial regulatory agencies. On the other hand, based on the industry and business characteristics of the enterprise itself, combined with internal compliance management requirements, the special regulatory requirements and corresponding regulatory agencies that some of the related parties of the related person may be subject to are determined.

[0055] For each regulatory agency identification, there is a clear basis for identification. These bases include, but are not limited to, the nature of the related party (such as whether it is a financial institution related party), the type of related relationship (such as equity control relationship may lead to the involvement of different regulatory agencies), and the scope and nature of the company's business activities (such as the related parties of a global company may be regulated by different regulatory agencies).

[0056] The regulator identification plays an important role in the relationship hierarchy map. It is used to characterize the relationship elements of the related party. Specifically, the regulator identification can clarify whether each related party of the related party is regulated by a specific regulator and the degree and scope of supervision. For example, a related party with a specific regulator identification may need to submit relevant financial information and business activity reports regularly in accordance with the requirements of the regulator.

[0057] Regulatory agency identification also helps to classify and manage related relationships. Related parties identified by different regulatory agencies may have different regulatory priorities and requirements. By marking different regulatory agency identifications on nodes in the map, we can more clearly understand the regulatory status of different parts of the entire related relationship network, and provide accurate guidance for corporate compliance operations and regulatory data reporting.

[0058] The related elements include each related party of the related party. The related parties here cover a wide range, including but not limited to legal entities (such as subsidiaries, parent companies, joint ventures, etc.), natural persons (such as corporate executives and their relatives, etc.) and other non-legal organizations (such as partnerships, etc.). Whether it is a direct related party or an indirect related party, as long as it meets the definition of a related party in the regulatory rules, it should be included in the relationship hierarchy map and considered as a related element.

[0059] When determining related parties, it is necessary to strictly follow the regulatory rules and the relevant requirements of the enterprise itself. For example, when determining equity relationships, it is necessary to determine whether they are related parties based on whether the shareholding ratio meets the standards stipulated in the regulatory rules or the internal regulations of the enterprise; when determining transaction relationships, it is necessary to consider whether factors such as transaction amount and frequency meet the relevant standards.

[0060] The association elements also include the hierarchy and association relationship between the associated person and each of his / her associated parties. The hierarchy refers to the distance measurement between the associated person and his / her associated parties in the association hierarchy map. With the associated person as the center, the direct associated parties are at the first level, and the levels of indirect associated parties increase in sequence. The association relationship refers to the specific type of relationship between the associated person and his / her associated parties, such as equity relationship, transaction relationship, kinship relationship, etc.

[0061] The determination of hierarchy and relationship must be based on accurate data and reasonable analysis methods. For example, the equity relationship can be determined by analyzing the equity structure documents of the enterprise, the transaction relationship can be determined by transaction records, and the kinship relationship can be determined by kinship certification documents. At the same time, the hierarchy and relationship information should be accurately marked in the map to facilitate the subsequent analysis and processing of the relationship.

[0062] In one embodiment, Figure 2 The specific example diagram of a method for determining the association relationship of an associated person shown in the figure, determining the association relationship level map of the associated person, includes:

[0063] Step 201: constructing a hierarchical graph of the associated relationships of the associated person according to the associated party input information of the associated person; the associated party input information includes the associated person's superior associated party and the type of the associated relationship between the associated person and its superior associated party;

[0064] Step 202: Based on the regulatory identification configuration information, generate one or more regulatory agency identifications of the associated persons in the association relationship hierarchy map.

[0065] In the embodiment, the level refers to the distance measurement between the associated person and his / her associated parties in the relationship level map. It reflects the closeness of the relationship, with the associated person as the center, the direct associated party at the first level, and the associated parties of the associated party (i.e., indirect associated parties) at increasing levels. The number of levels should be limited and clearly defined. For example, a maximum number of levels can be set, such as no more than 10 levels (the specific value can be determined based on actual regulatory needs and business scenarios) to prevent the relationship from being too complex and difficult to track. This helps to control the scale of the relationship level map and improve analysis efficiency and accuracy.

[0066] In the relationship hierarchy map, each level should have a clear identification method. Numerical numbers (such as level 1, level 2, etc.) or specific symbol systems can be used to distinguish different levels. At the same time, for the associated parties on each level, their level information should be recorded so that their position in the map can be accurately identified when analyzing and processing the relationship.

[0067] Related relationships include, but are not limited to, equity relationships, transaction relationships, kinship relationships, control relationships, etc. These relationships should have clear definitions and criteria for definition. For example, equity relationships can be determined based on the shareholding ratio. If the shareholding exceeds a certain ratio (such as 20%), it is considered that there is an equity related relationship; transaction relationships can be defined based on indicators such as transaction amount and transaction frequency to determine whether they constitute an important related relationship. Related relationships should be directional, that is, it is clearly stated whether it is from the related person to the related party or vice versa. For example, in an equity relationship, if the related person holds shares in the related party, the direction of the related relationship is from the related person to the related party, indicating that the related person has a certain influence on the related party. This directionality helps to accurately describe the nature of the relationship and the direction of flow in the map.

[0068] In addition to type and direction, the relationship should also have a strength description. Strength can be measured in many ways. For example, in equity relationships, the higher the shareholding ratio, the stronger the relationship strength; in transaction relationships, the larger the transaction amount and the higher the transaction frequency, the stronger the relationship strength.

[0069] In one embodiment, a graph is constructed based on the information entered by the related parties, including:

[0070] The information entered by the related parties of the related party is an important basis for constructing the hierarchical map of related relationships. First of all, it is necessary to ensure the accuracy and completeness of the information entered by the related parties. This requires strict compliance with relevant data entry specifications and standards during the information collection process. For the information of the superior related parties of the related party, its name, nature (such as legal person or natural person) and other relevant identification characteristics must be clarified. At the same time, the type of relationship between the related party and its superior related party must be recorded in detail, such as whether it is an equity relationship, a control relationship or other specific business relationship.

[0071] After collecting the input information of related parties, organize and classify it. Group different types of related party information according to certain rules so that the information can be used more efficiently in the subsequent map construction process. For example, you can make a preliminary classification based on the nature of the related parties (legal person or natural person), and then subdivide each category based on the type of relationship.

[0072] In one embodiment, the graph construction process includes:

[0073] With the associated person as the core node, we start to build a hierarchical graph of associated relationships. We take the associated person's superior associated party as the first-level node, and connect the associated person and its superior associated party with directed edges according to the type of relationship between the associated person and the associated person. For example, if the associated person is a company, its superior associated party is the parent company, and the type of relationship is an equity relationship, then there will be a directed edge from the associated person to the parent company in the graph, marked as an equity relationship.

[0074] For the first-layer node (i.e., the superior associated party of the associated person), continue to search for its own superior associated party as the second-layer node of the graph, and determine its type of association relationship with the first-layer node in the same way, and connect them with directed edges. In this way, a multi-level hierarchical graph of association relationships is gradually constructed. During the construction process, it is necessary to ensure that the structure of the graph can accurately reflect the hierarchical relationship and type of association relationship between the associated parties.

[0075] In one embodiment, generating a regulatory agency identifier based on regulatory agency configuration information includes:

[0076] Regulatory identification configuration information is the basis for generating one or more regulatory agency identifications of associated persons in the relationship hierarchy map. This information usually comes from relevant regulations and documents issued by regulatory agencies, as well as relevant requirements formulated by enterprises based on regulatory requirements. It is necessary to study and understand these identification configuration information in depth to clarify what regulatory agency identifications should be assigned to different types of associated parties under what circumstances.

[0077] For example, for enterprises in the financial industry, regulators may prescribe different rules for configuring regulatory agency identification based on factors such as the equity structure, business scope, and risk level of related parties. Enterprises need to accurately obtain and apply regulatory identification configuration information based on these rules and their own actual conditions.

[0078] In the process of constructing the relationship hierarchy map, the corresponding regulatory agency identification is generated for the associated persons and their associated parties in the map based on the determined associated party information and regulatory identification configuration information. For each node (i.e., the associated person or its associated party), the corresponding regulatory identification configuration rules are searched based on its own nature, relationship type, and level, and the regulatory agency identification that should be assigned to it is determined.

[0079] For example, if an associated party is a financial subsidiary, it should be assigned the logo of a specific financial regulatory agency based on its equity structure and business scope, as well as the relevant regulatory identification configuration rules. In the graph, the regulatory agency logo is marked on the corresponding node to clearly show the regulatory status of each node.

[0080] In specific implementation, after performing step 101: determining the relationship hierarchy map of the associated person, step 102 is performed: performing text recognition processing on the received associated party supervision information to obtain the specified relationship elements; the associated party supervision information is used to describe the information that complies with the specified associated party identification rules.

[0081] In the embodiment, related party regulatory information refers to information used to describe compliance with the rules for identifying designated related parties. This information comes from a wide range of sources, covering various documents and records within the enterprise and relevant regulations and guidelines issued by regulatory agencies. Its content includes but is not limited to basic information of related parties (such as name, nature, registered address, etc.), description of related relationships (such as equity relationships, transaction relationships, family relationships, etc.), and regulatory requirements and compliance standards. This information plays a vital role in accurately identifying related parties and their relationships, and ensuring that enterprises operate in compliance.

[0082] The purpose of text recognition processing on the received related party supervision information is to extract the key specified relationship elements from the complex text information. Since related party supervision information may exist in various forms, such as paper documents, electronic documents, etc., and its content may contain a large amount of text descriptions and professional terms, it is necessary to convert it into a form that can be analyzed and processed through text recognition processing in order to accurately obtain the specified relationship elements.

[0083] Through the above-mentioned text recognition processing method, the designated relationship elements are obtained from the related party supervision information. These elements include but are not limited to the name, nature, type of relationship, hierarchical relationship between related parties, etc. of the related party. For example, the subject "Company A" and the object "Company B" in the sentence "Company A holds 30% of the equity of Company B" are identified through the semantic analysis algorithm, and the type of relationship represented by the predicate "holding" is an equity relationship; through the semantic role labeling technology, Company A is determined as the subject and Company B is determined as the object, thereby accurately obtaining the designated relationship element of the equity relationship. These designated relationship elements will be used for subsequent operations such as related party identification and regulatory identification.

[0084] In one embodiment, the received related party supervision information is subjected to text recognition processing to obtain the specified association relationship elements, including:

[0085] The semantic analysis algorithm and semantic role labeling technology in natural language processing technology are used to perform text recognition processing on the received related party supervision information to obtain specified association relationship elements; the semantic analysis algorithm is used to parse the semantic structure in the supervision information; the semantic role labeling technology is used to analyze the semantic roles in the supervision information to determine the related person entity, related party entity and association relationship entity.

[0086] In one embodiment, the detailed process of obtaining the specified association relationship element through text recognition processing is as follows:

[0087] When processing related-party regulatory information, the semantic analysis algorithm parses the text according to the rules and principles of linguistics. It regards the text as a hierarchical structure consisting of words, phrases, and sentences, and parses the semantic structure in the regulatory information by identifying the grammatical components and semantic relationships therein. For example, for the sentence "Company A and Company B formed an association due to equity transactions", the semantic analysis algorithm can identify "Company A" and "Company B" as noun phrases, respectively serving as the subject and object in the sentence, "equity transaction" as a verb phrase, indicating the behavior between the two, and "formed an association" as a predicate verb phrase, describing the overall semantic relationship.

[0088] First, the regulatory information is lexically analyzed to decompose the text into individual words and annotate their parts of speech. For example, "Enterprise A and Enterprise B form an associated relationship due to equity transactions" is decomposed into "Enterprise / A", "A / conjunction", "Enterprise / B", "cause / preposition", "Equity / noun", "Transaction / noun", "Forming / verb", "Association / noun". Then, based on part-of-speech tagging and grammatical rules, a grammatical tree of the sentence is constructed. The grammatical tree takes the core verb of the sentence as the root node and other components as child nodes, and displays the grammatical structure of the sentence through branching relationships. For example, for the above sentence, the root node of the grammatical tree is "formation", and its child nodes include "Enterprise A", "Enterprise B", "Equity transaction", etc. Finally, through the analysis of the grammatical tree, the semantic structure in the regulatory information is parsed, including the relationship between subject, object and behavior.

[0089] Semantic role labeling technology is based on semantic theory and aims to analyze the semantic roles played by each component in a sentence. It further clarifies the semantic connotation of a sentence by identifying semantic roles such as subject, object, tool, and place in the sentence. For related party regulatory information, it can accurately determine the related person entity, related party entity, and related relationship entity. For example, in the sentence "Enterprise A controls Enterprise B through equity acquisition", "Enterprise A" is labeled as the subject role because it is the subject that performs the "control" behavior; "Enterprise B" is labeled as the object role because it is the object of "control"; "Equity acquisition" is labeled as the tool role because it is the means by which Enterprise A achieves control.

[0090] First, based on the semantic structure parsed by the semantic analysis algorithm, the core verb in the sentence is determined. For example, in "Enterprise A controls Enterprise B through equity acquisition", the core verb is "control". Then, the semantic roles of other components in the sentence are annotated according to the pre-defined semantic role template and semantic theory. For each component, its relationship with the core verb is judged to determine its semantic role. For example, the relationship between "Enterprise A" and "control" is the execution subject, so it is marked as the subject role; the relationship between "Enterprise B" and "control" is the controlled object, so it is marked as the object role; the relationship between "equity acquisition" and "control" is the means of implementation, so it is marked as the tool role. Through this process, the semantic roles in the regulatory information can be accurately analyzed to determine the associated person entity, the associated party entity and the associated relationship entity.

[0091] The semantic structure analyzed by the semantic analysis algorithm and the semantic roles determined by the semantic role labeling technology are used to integrate relevant information. For example, the entities corresponding to the subject role (such as enterprise A), the entities corresponding to the object role (such as enterprise B), and the relationship between them (such as equity relationship) are integrated to form a complete description of the relationship elements.

[0092] For complex regulatory information, which may contain multiple sentences and multiple relationships, it is necessary to perform the above analysis on each sentence and summarize and integrate all analysis results to form a comprehensive set of specified relationship elements.

[0093] Extract key designated relationship elements from the integrated information. These elements include the name, nature, type of relationship, hierarchical relationship between related parties, etc. For example, from regulatory information such as "Enterprise A controls Enterprise B through equity acquisition, and Enterprise B has a transaction relationship with Enterprise C", extract the names of related parties such as "Enterprise A", "Enterprise B", "Enterprise C", "equity relationship", "transaction relationship" and other types of relationships, as well as the hierarchical relationship reflected by the control relationship of Enterprise A over Enterprise B and other designated relationship elements. These elements will be used for subsequent operations such as identification of related parties and determination of regulatory identification.

[0094] During specific implementation, after performing step 102: performing text recognition processing on the received related party regulatory information to obtain the designated related party relationship elements, perform step 103: identifying the existing related party elements of the existing relationship corresponding to the regulatory agency identifier of the related person; matching the existing related party elements with the designated related party elements; and treating the existing related party elements that have not been successfully matched as related party elements to be corrected that do not comply with the designated related party identification rules.

[0095] In the embodiment, in the process of identifying and correcting the association relationship, the following steps can ensure that the association elements to be corrected that do not comply with the specified association party identification rules are accurately found.

[0096] First, it is necessary to identify the existing association elements of the existing association relationships corresponding to the regulatory agency identification of the related persons. These existing association elements include, but are not limited to, the specific types of association relationships between related parties (such as kinship, equity relationships, etc.), hierarchical relationships between related parties, attribute information of related parties, etc. Through in-depth analysis of existing association relationships, these key elements can be extracted to provide a basis for subsequent matching and correction. For example, for an enterprise's related party system, existing association elements may include equity relationships between the enterprise and shareholders, control relationships between the enterprise and subsidiaries, business cooperation relationships between the enterprise and partners, etc. These elements can be obtained by querying and analyzing the enterprise's related party database.

[0097] In order to accurately identify existing related elements, it is necessary to collect and organize relevant data. This can include the company's internal database, industrial and commercial registration information, financial statements, etc. By integrating and analyzing these data, the existing related elements between related parties can be extracted and classified and organized for subsequent processing. For example, the equity ratio information between the company and its shareholders can be extracted from the company's financial statements, and the control relationship information between the company and its subsidiaries can be obtained from the industrial and commercial registration information. These data can provide strong support for identifying existing related elements.

[0098] Clearly specifying the elements of the relationship is a prerequisite for matching. The specified elements of the relationship are determined based on the requirements of the regulatory agency and the actual situation of the enterprise, and are used to determine whether the relationship complies with the specified related party identification rules. These elements can include specific relationship types, hierarchical relationship requirements, regulatory identification, etc. For example, according to the regulations of the regulatory agency, certain types of relationships may not be recognized, or the hierarchical relationship between related parties cannot exceed a certain level. These requirements can be converted into specified relationship elements for matching with existing relationship elements.

[0099] Match the existing association elements with the specified association elements. This can be achieved by comparing the key attributes and features of the two. For example, you can compare information such as the type of association, hierarchical relationship, and regulatory identification to determine whether the existing association elements meet the requirements of the specified association elements. For example, you can use an automated matching algorithm to compare the existing association elements with the specified association elements one by one to find out the mismatched parts. Alternatively, you can check and judge the key associations one by one through manual review.

[0100] Existing association elements that have not been successfully matched are treated as association elements to be corrected that do not comply with the designated related party identification rules. These association elements to be corrected need to be further processed to ensure that the association relationship meets regulatory requirements. For example, for association relationships that have not been successfully matched, they can be marked as association elements to be corrected, and the specific reasons and information for the mismatch can be recorded. This can provide clear goals and directions for subsequent correction work. Classify and organize the association elements to be corrected to better carry out subsequent correction work. Classification can be based on factors such as the reason for the mismatch, the type of related party, and the requirements of the regulatory agency. For example, the association elements to be corrected can be divided into different categories, such as mismatched association relationship types, mismatched hierarchical relationships, and mismatched regulatory identification. For each category, corresponding correction strategies and methods can be formulated to improve the efficiency and accuracy of correction work.

[0101] Through the above steps, the related elements to be corrected that do not comply with the designated related party identification rules can be accurately identified, providing strong support for subsequent related relationship correction work.

[0102] In specific implementation, after performing step 103: identifying the existing association elements of the existing association relationship corresponding to the regulatory agency identifier of the associated person; matching the existing association elements with the designated association relationship elements; treating the existing association elements that have not been successfully matched as association elements to be corrected that do not comply with the designated associated party identification rules, perform step 104: correct one or more regulatory agency identifiers of the associated person according to the association elements to be corrected to obtain a corrected associated party regulatory identifier.

[0103] In an embodiment, identifying an existing association element of an existing association relationship includes:

[0104] First, it is necessary to clarify the mapping mechanism between the regulator identification and the existing relationship. The regulator identification is a way to classify and identify the associated persons, which reflects the regulatory environment and requirements to which the associated persons are subject. Each regulator identification corresponds to a specific set of existing relationships. For example, in the financial industry, a specific regulator identification may correspond to existing relationships such as equity relationships and transaction relationships with financial institutions. By establishing this mapping relationship, relevant existing relationships can be quickly located based on the regulator identification.

[0105] In order to establish an accurate mapping relationship, it is necessary to conduct in-depth research on the coding rules and meanings of the regulatory agency logo. The regulatory agency logo may contain multiple parts, each of which represents different information. For example, one part may represent the industry type, another part may represent the nature of the relationship, etc. By analyzing these coding rules, we can better understand the connection between the regulatory agency logo and the existing relationship.

[0106] Once the existing associations corresponding to the regulatory agency identifiers are determined, it is necessary to extract the existing association elements from these existing associations. The existing association elements include various attribute information of the related parties, such as the name, type (legal person or natural person), registered address, etc. of the related parties, as well as the nature of the association (such as equity relationship, transaction relationship, etc.) and hierarchical information. For example, in an equity relationship, the existing association elements may include the name of the shareholder, the shareholding ratio, the registered address of the company, etc.

[0107] Extracting existing related elements requires the use of a variety of methods. Basic information about related parties can be obtained from the company's registration information, internal databases, and other sources. The nature and level of related relationships can be determined by analyzing the company's equity structure documents, transaction records, and other information.

[0108] Before matching, it is necessary to determine the matching criteria. The matching criteria should be based on regulatory requirements and the relevant requirements of the enterprise itself. For example, for the names of related parties, the matching criteria may require complete consistency; for the nature of the relationship, the matching criteria may require compliance with specific industry specifications and definitions. The determination of matching criteria requires comprehensive consideration of various factors to ensure the accuracy and reliability of the matching results.

[0109] In order to facilitate the matching operation, the matching criteria can be quantified and standardized. For example, the nature of the association relationship can be divided into different categories, and each category can be assigned a specific code. In this way, when matching, the codes can be compared to determine whether the match is successful.

[0110] According to the determined matching criteria, the existing association elements are matched with the specified association relationship elements. For the names of the related parties, they are directly compared for consistency; for the nature of the association relationship, their category codes are compared for the sameness; for the hierarchical information of the association relationship, their hierarchical values ​​are compared for equality. For example, if the nature of the association relationship in the existing association element is equity relationship, the category code is "01", and the nature of the association relationship in the specified association relationship element is also equity relationship, and the category code is "01", then it is considered that the match is successful in terms of the nature of the association relationship.

[0111] In the matching process, some complex situations may be encountered. For example, the relationship hierarchy information in the existing relationship elements may be ambiguous, or there may be multiple ways to express the name of the associated party in the specified relationship element. In this case, some auxiliary methods are needed to solve these problems. For example, the accurate expression of the relationship hierarchy information and the associated party name can be further determined by querying the internal database of the enterprise and referring to relevant industry standards.

[0112] After the matching process, the existing association elements that have not been successfully matched are screened out. During the screening process, it is necessary to distinguish between the unmatched situations due to inaccurate or incomplete data and the unmatched situations due to non-compliance with regulatory requirements. The unmatched situations due to inaccurate or incomplete data can be resolved by further verifying the data source and supplementing the missing information. The unmatched situations due to non-compliance with regulatory requirements are treated as association elements to be corrected.

[0113] For example, if the name of the related party in the existing relationship element is inconsistent with the name of the related party in the specified relationship element, but it is found by querying the internal database of the enterprise that it is caused by a data entry error, the problem can be solved by correcting the data entry error and it will not be treated as a relationship element to be corrected. However, if the nature of the relationship in the existing relationship element is inconsistent with the nature of the relationship in the specified relationship element, and it meets the judgment conditions for non-compliance with regulatory requirements, it will be treated as a relationship element to be corrected.

[0114] Classify and manage the association elements to be corrected. According to the type of association elements to be corrected (such as errors in the names of related parties, errors in the nature of the association relationship, etc.) and the degree of impact on the regulatory results, they are divided into different categories. For example, the association elements to be corrected that have a greater impact on the regulatory results can be placed in one category and corrected first. This can improve the efficiency of correction and ensure that the corrected association relationship meets regulatory requirements.

[0115] In specific implementation, after performing step 104: correcting one or more regulatory agency identifiers of the associated person according to the associated elements to be corrected and obtaining the corrected associated party regulatory identifiers, perform step 105: determine the hierarchical path step between the associated person and the associated party corresponding to each corrected associated party regulatory identifier.

[0116] In the embodiment, the hierarchical path step refers to the hierarchical distance measurement between the associated person and the specific associated party with the modified regulatory identification in the associated party relationship network. It reflects the relative position relationship between the associated parties in the hierarchical structure, which is of great significance for accurately understanding the closeness of the relationship between the associated parties, the transfer of regulatory responsibilities and risk transmission.

[0117] For example, in an enterprise's related party map, the related party may be the core controller of the enterprise, while the related parties with different regulatory identification may be subsidiaries, shareholders or partners of the enterprise at all levels. The hierarchical path step can help determine the hierarchical span from the related party to each related party, so as to better evaluate the status and role of different related parties in the regulatory system.

[0118] Accurately determining the step length of the hierarchical path helps to achieve effective supervision. By understanding the hierarchical relationship between related parties, it is possible to formulate more targeted supervision strategies, clarify the allocation of supervision responsibilities, and timely discover potential risk transmission paths. For example, if the step length of the hierarchical path between a related party and its related persons is shorter, it means that its relationship with the core related persons is closer and may require more stringent supervision measures; while the supervision focus of related parties with longer hierarchical path steps may be different.

[0119] Here is an example of determining the step length of a hierarchical path:

[0120] The related party relationship network is constructed as a directed graph, where nodes represent related parties and edges represent the relationships between related parties. Through the algorithm in graph theory, the shortest path length between the related person and the related party corresponding to each revised related party regulatory identification, i.e., the hierarchical path step length, can be calculated.

[0121] For example, the Dijkstra algorithm or other shortest path algorithms can be used to traverse the entire related party relationship graph starting from the associated person, and calculate the shortest path length to each associated party with a revised regulatory identifier. During the calculation process, different weights can be assigned to the edges based on factors such as the strength of the relationship between the associated parties and the importance of supervision, so as to more accurately reflect the actual hierarchical path step length.

[0122] In addition to pure graph theory analysis, specific regulatory rules are also needed to determine the hierarchical path step length. Different regulatory rules may have different definitions and requirements for the hierarchical relationship between related parties.

[0123] For example, some regulatory rules may stipulate hierarchical restrictions between certain types of related parties, or set different regulatory standards for related parties at different levels. When determining the hierarchical path step length, these regulatory rules need to be taken into account and the calculated graph path length needs to be adjusted and corrected to ensure compliance with regulatory requirements.

[0124] As the relationship between related parties changes and regulatory requirements are adjusted, the hierarchical path step length also needs to be dynamically updated and adjusted. For example, when a new related party joins or the relationship between an existing related party changes, the hierarchical path step length needs to be recalculated to ensure the accuracy and timeliness of supervision.

[0125] At the same time, changes in regulatory rules may also lead to a reassessment of the step length of the hierarchical path. Enterprises need to establish corresponding monitoring mechanisms to promptly detect changes in related party relationships and regulatory rules, and make corresponding adjustments to the step length of the hierarchical path.

[0126] Through the above method, the hierarchical path step between the related party and each related party corresponding to the revised related party supervision identification can be accurately determined, providing strong support for the company's related party management and supervision.

[0127] In one embodiment, determining the hierarchical path step between the associated person and each associated party corresponding to the modified associated party supervision identifier includes:

[0128] The shortest path algorithm and heuristic search algorithm in graph theory are used to determine the hierarchical path step between the related person and the related party corresponding to each revised related party regulatory identification.

[0129] In one embodiment, in order to accurately determine the hierarchical path step length between the associated person and the associated party corresponding to each revised associated party supervision identifier, the shortest path algorithm and heuristic search algorithm in graph theory are used.

[0130] The shortest path algorithm is an important algorithm in graph theory, and its goal is to find the shortest path between two nodes in a given directed or undirected graph. In determining the hierarchical path of the related party relationship, the related parties are regarded as nodes in the graph, and the relationship between the related parties is regarded as the edge in the graph. The shortest path algorithm can be used to find the shortest hierarchical path between the related person and the specific related party. For example, for a relationship network containing multiple related parties, the related person node is the starting point, and the related party node corresponding to the revised related party regulatory identification is the end point. The shortest path algorithm determines the shortest path from the starting point to the end point by calculating the distance between the nodes and the weight of the edge.

[0131] Commonly used shortest path algorithms include Dijkstra algorithm, Bellman-Ford algorithm, etc. These algorithms calculate the shortest distance between nodes in the graph in different ways and can handle different types of graph structures and edge weights. In determining the hierarchical path of the associated party, the appropriate shortest path algorithm can be selected according to the actual situation to improve the calculation efficiency and accuracy.

[0132] First, a directed graph of the relationship between related parties is constructed. Each related party is represented as a node in the graph, and the relationship between related parties is represented as a directed edge. The weight of the edge can be determined based on factors such as the hierarchical relationship and relationship strength between related parties. For example, the edge weight between directly related parties can be set to a smaller value, and the edge weight between indirect related parties can be appropriately increased according to the hierarchical distance.

[0133] Then, select the appropriate shortest path algorithm for calculation. Starting from the associated person, traverse the nodes in the graph and update the distance and path information between nodes according to the rules of the algorithm. During the calculation process, the weight of the edge and the structure of the graph need to be considered to ensure that the shortest path found is accurate.

[0134] Finally, the shortest hierarchical path step between the related party and the related party corresponding to the revised related party supervision identifier is output. This step represents the hierarchical distance between the related party and the specific related party, which can be used to evaluate the closeness and regulatory risk between related parties.

[0135] Heuristic search algorithm is a search algorithm based on heuristic information. It uses specific knowledge and experience of the problem in the search process to guide the search in a more promising direction. In the determination of the hierarchical path of related parties, the heuristic search algorithm can use the known relationship between related parties, regulatory rules and other heuristic information to quickly find the hierarchical path step between the related person and the specific related party.

[0136] Heuristic search algorithms usually use an evaluation function to measure the quality of search states. The evaluation function combines the goal of the problem and the heuristic information to evaluate the search state and select the next most promising search state for expansion. In determining the hierarchical path of related parties, the evaluation function can consider factors such as the hierarchical distance between related parties, the strength of the relationship, and regulatory risks to determine the direction and priority of the search.

[0137] First, define the evaluation function of the heuristic search algorithm. The evaluation function can be designed based on factors such as the hierarchical relationship between related parties, relationship strength, and regulatory rules. For example, the hierarchical distance and relationship strength between related parties can be used as the main factors of the evaluation function, while considering the impact of regulatory rules on related parties.

[0138] Then, starting from the associated person, a heuristic search is performed. During the search process, the next most promising associated party node is selected for expansion according to the evaluation function. The expansion process can include steps such as checking the neighbor nodes of the node, calculating the distance between nodes, and evaluating the function value.

[0139] When the related party node corresponding to the revised related party supervision identifier is found, the search stops and the hierarchical path step between the related person and the related party is output. This step is the shortest hierarchical path step found by the heuristic search algorithm and can be used for subsequent related party analysis and supervision decisions.

[0140] By using the shortest path algorithm and heuristic search algorithm in graph theory, the hierarchical path step length between the related party and each related party corresponding to the revised related party regulatory identification can be accurately determined. This method can help enterprises better understand the relationship between related parties, assess regulatory risks, and take appropriate measures for risk management and compliance management.

[0141] In specific implementation, after performing step 105: determining the hierarchical path step between the associated person and each associated party corresponding to the revised associated party regulatory identifier, perform step 106: delete the associated relationships corresponding to the hierarchical path step exceeding the preset value in the associated relationship hierarchical map, and obtain the revised associated person's associated relationship hierarchical map.

[0142] In the embodiment, the determination of the preset value needs to comprehensively consider the regulatory requirements and the actual business situation. The regulatory requirements may have certain restrictions on the hierarchical relationship between related parties, such as stipulating that related parties exceeding a certain level may not be included in the regulatory scope. At the same time, the actual business situation will also affect the determination of the preset value, such as the organizational structure of the enterprise, business processes, and the actual relationship between related parties.

[0143] For example, in the financial industry, regulators may have clear regulations on the hierarchical relationship between financial institutions and their related parties to prevent excessive risk transmission. Enterprises can determine a reasonable preset value based on these regulatory requirements and their own business characteristics.

[0144] According to the determined preset value, the associations corresponding to the hierarchical path steps exceeding the value are identified in the association hierarchy map. This requires analyzing each association in the map one by one, calculating its hierarchical path step, and comparing it with the preset value.

[0145] For example, for a hierarchical graph containing multiple associated parties, we can start from the associated person node and traverse down along the hierarchical relationship step by step to calculate the hierarchical path step between each associated party and the associated person. If the hierarchical path step between a certain associated party and the associated person exceeds the preset value, the association between the associated party and the upper-level associated party is marked as an association to be deleted.

[0146] The revised relationship hierarchy map can be applied to the company's related party management, regulatory compliance, and risk control. For example, the company can determine which related parties need to be included in the regulatory scope based on the revised map and formulate corresponding regulatory strategies and risk control measures. At the same time, the revised map can also provide support for the company's business decision-making, such as providing reference for transaction approval involving related parties and partner selection.

[0147] For example, when reviewing transactions involving related parties, enterprises can quickly determine the hierarchical relationship and regulatory requirements between the two parties based on the revised hierarchical map of related parties, thereby more accurately assessing the risks and compliance of transactions. When selecting partners, enterprises can refer to the revised map to avoid cooperating with related parties whose hierarchical relationships are too complex or risky.

[0148] In one embodiment, Figure 3 The specific example diagram of a method for determining the association relationship of an associated person shown in the figure deletes the association relationship corresponding to the hierarchical path step length exceeding a preset value in the association relationship hierarchical map to obtain a revised association relationship hierarchical map of the associated person, including:

[0149] Repeat the following operations until there are no more relationships in the graph whose hierarchical path step length exceeds the preset value, and obtain the revised hierarchical graph of the relationship between the associated persons:

[0150] Step 301: for each regulatory agency identifier, based on a greedy algorithm, determining the relationship attributes of different association relationships corresponding to the regulatory agency identifier; the relationship attributes include the connection frequency of the association relationship;

[0151] Step 302: According to the relationship attributes, the different association relationships corresponding to the regulatory agency identifier are selectively modified.

[0152] In one embodiment, the associations corresponding to the hierarchical path steps exceeding a preset value in the association hierarchical graph are deleted to obtain a revised association hierarchical graph of the associated persons. The specific steps are as follows:

[0153] A greedy algorithm is an algorithm that takes the best decision under the current state in each step of selection. In this embodiment, for each regulatory agency identifier, a greedy algorithm is used to determine the relationship attributes of different associations corresponding to the regulatory agency identifier. The connection frequency in the relationship attribute reflects the activity and importance of the association in the association hierarchy map. By counting and analyzing the connection frequency, we can better understand the status and role of different associations in the entire regulatory system.

[0154] For example, for a specific regulatory agency identifier, there may be multiple corresponding relationships. Through the analysis of historical data and real-time monitoring, the connection frequency of each relationship can be determined. If the connection frequency of a relationship is high, it means that it appears frequently in the relationship hierarchy map and may have a greater impact on the regulatory results. On the contrary, if the connection frequency of a relationship is low, its importance may be relatively low.

[0155] In order to determine the connection frequency of the associated relationships, it is necessary to collect and analyze relevant data. This can include information such as transaction records, communication records, equity relationships, etc. between the associated parties. By analyzing these data, the number of occurrences of different associated relationships can be determined, thereby calculating the connection frequency.

[0156] For example, we can extract transaction records between related parties from the enterprise's database and count the number of occurrences of each relationship within a certain period of time. Or we can monitor the communication traffic between related parties to determine the activity level of different relationships. These data can provide input for the greedy algorithm to help determine the relationship attributes of each relationship.

[0157] According to the relationship attributes, formulate strategies for selectively modifying different associations. For associations with higher connection frequencies, it may be necessary to modify them more carefully to avoid significant impact on regulatory results. For associations with lower connection frequencies, appropriate adjustments or deletions can be made according to their specific circumstances.

[0158] For example, if the connection frequency of an association is high and it plays an important role in supervision, it can be considered to be modified by adjusting its hierarchical path step or adding a regulatory mark to ensure that it meets regulatory requirements. If the connection frequency of an association is low and it is not within the scope of the regulatory related party identification rules, it can be considered to directly delete the association to simplify the hierarchical map of the association.

[0159] According to the formulated correction strategy, the different relationships corresponding to the regulatory agency logo are selectively corrected. During the correction process, it is necessary to consider the mutual impact between the relationships and the overall regulatory effect. If the correction of one relationship will affect other relationships, a comprehensive assessment is required to ensure that the revised relationship hierarchy map can still accurately reflect the relationship between the related parties and the regulatory requirements.

[0160] For example, when deleting a relationship with a low connection frequency, it is necessary to check whether the relationship has dependencies with other important relationships. If there are dependencies, it is necessary to consider whether the deleted relationship can be replaced by adjusting other relationships to maintain the integrity and accuracy of the relationship hierarchy map.

[0161] Repeat the above operation, i.e., determine the relationship attributes and make selective corrections for each regulatory agency identifier, until there are no more relationships in the graph whose hierarchical path step length exceeds the preset value. In this process, it is necessary to set appropriate iteration termination conditions to ensure that the operation can be completed within a reasonable time.

[0162] For example, a maximum number of iterations can be set. When this number is reached, the iteration stops even if there are still associations in the graph whose hierarchical path step length exceeds the preset value. Alternatively, a convergence condition can be set according to the change of the graph. When the change of the graph is less than a certain threshold, the iteration is considered to have converged and the operation stops.

[0163] When the operation stops, the revised hierarchical map of the associated persons is obtained. In order to ensure the accuracy and effectiveness of the map, the final result needs to be verified. This can include comparison with other data sources, simulation of regulatory scenarios, etc.

[0164] For example, the revised relationship hierarchy map can be compared with the company's internal database, regulatory requirements and actual business conditions to check whether there are any missing relationships or incorrect regulatory identification. At the same time, some regulatory scenarios can be simulated to verify the effectiveness and accuracy of the revised map in actual supervision.

[0165] Through the above steps, the association relationships corresponding to the hierarchical path steps exceeding the preset value in the association relationship hierarchy map can be deleted to obtain a revised association relationship hierarchy map of the associated persons, thereby improving the accuracy and efficiency of related party management and meeting regulatory requirements.

[0166] In one embodiment, Figure 4 The specific example diagram of a method for determining the association relationship of an associated person shown in the figure deletes the association relationship corresponding to the hierarchical path step length exceeding a preset value in the association relationship hierarchical map to obtain a revised association relationship hierarchical map of the associated person, including:

[0167] Step 401: Determine the range of the preset value according to the supervision intensity level carried by the related party supervision information;

[0168] Step 402: Delete the associations corresponding to the hierarchical path steps exceeding a preset value in the association hierarchical graph, and obtain a revised association hierarchical graph of the associated persons.

[0169] In one embodiment, the process of deleting the association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical map to obtain a revised association relationship hierarchical map of associated persons is described in detail as follows.

[0170] The supervision intensity level carried in the related party supervision information reflects the strictness of supervision on different related parties. A higher supervision intensity level usually means stricter supervision requirements on related parties, requiring more detailed control of the hierarchical path steps of the relationship. A lower supervision intensity level may allow a relatively longer hierarchical path step.

[0171] For example, for related parties with higher financial risks, the regulatory intensity level may be higher, which requires a relatively small preset value to ensure that the relationships in the relationship hierarchy map are relatively close, which is convenient for stricter supervision. On the contrary, for some related parties with relatively low risks, the regulatory intensity level is lower, and the preset value can be appropriately increased to allow a certain degree of longer hierarchical path step length.

[0172] By analyzing different regulatory intensity levels, corresponding rules can be formulated to determine the range of preset values. This can be based on a comprehensive consideration of historical data, industry standards, and regulatory requirements. For example, data on related parties at different regulatory intensity levels can be collected to analyze the distribution of their hierarchical path steps, thereby determining the preset value range corresponding to different regulatory intensity levels. Alternatively, refer to the standards and best practices of relevant industries and determine the appropriate preset value range based on the actual situation of the enterprise.

[0173] After determining the preset value range, the hierarchical graph of the association relationship is traversed to calculate the hierarchical path step of each association relationship. The association relationship whose hierarchical path step exceeds the preset value is marked as an association relationship to be deleted. For example, for a hierarchical graph containing multiple associated parties, it can start from the associated person node and traverse down step by step along the hierarchical relationship to calculate the hierarchical path step between each associated party and the associated person. If the hierarchical path step between an associated party and an associated person exceeds the preset value, the association relationship between the associated party and the upper-level associated party is marked as an association relationship to be deleted.

[0174] Delete the identified relationships to be deleted. The deletion operation can be implemented by modifying the data structure of the relationship hierarchy graph, such as deleting the corresponding edges or nodes. When performing the deletion operation, it is necessary to maintain the integrity and consistency of the graph to avoid isolated nodes or incomplete hierarchical structures.

[0175] For example, in a hierarchical graph of association relationships represented by a graph structure, the associations corresponding to the hierarchical path steps exceeding the preset value can be deleted by deleting the edges in the graph. At the same time, it is necessary to check whether the deletion operation will lead to the isolation of other nodes or the incompleteness of the hierarchical structure. If this happens, corresponding adjustments and repairs need to be made.

[0176] After the deletion operation is completed, the revised hierarchical graph of the associated persons is obtained. At this point, the revised graph needs to be updated and verified to ensure its accuracy and completeness. The update operation may include recalculating node attributes, hierarchical relationships, and regulatory identification information in the graph. The verification operation can be performed by comparing with other data sources, simulating business scenarios, etc. to ensure that the revised graph meets regulatory requirements and business realities.

[0177] For example, an enterprise can compare the revised relationship hierarchy map with the internal database, the requirements of the regulatory authorities and the actual business situation to check whether there are any missing relationships or incorrect regulatory identification. At the same time, some business scenarios can be simulated to verify the effectiveness and accuracy of the revised map in actual business.

[0178] The revised relationship hierarchy map can be applied to the company's related party management, regulatory compliance, and risk control. For example, the company can determine which related parties need to be included in the regulatory scope based on the revised map and formulate corresponding regulatory strategies and risk control measures. At the same time, the revised map can also provide support for the company's business decision-making, such as providing reference for transaction approval involving related parties and partner selection.

[0179] Two specific embodiments are given below to illustrate the specific application of the method of the present invention.

[0180] First embodiment:

[0181] 1. Application scenario background

[0182] Assume that a large financial group company has many subsidiaries, holding companies and various related natural persons. The financial group needs to strictly comply with the regulations of financial regulators, accurately identify and manage related parties, and ensure the accuracy of related party regulatory identification to avoid misreporting of transaction data involving related parties and triggering regulatory penalties.

[0183] 2. Implementation steps

[0184] 1. Related party declaration and map construction:

[0185] During the declaration of related parties, the staff enters the information of each related party. For example, the direct relationship between a holding subsidiary and the related party collection agency is entered as a holding relationship, and its superior related party is selected as the group headquarters, and the relationship between the two is clarified. In this way, a relationship map of the upper and lower hierarchical relationships is gradually formed. With the group headquarters as the core, and each subsidiary, holding company and related natural persons as nodes, a directed graph is constructed based on the actual relationship to ensure that there is no reverse cycle relationship.

[0186] 2. Obtain the corresponding regulatory agency identification of each related party. For example, a subsidiary calculates its own regulatory identification combination information and virtual wire information from multiple related parties to the subsidiary based on its relationship with the related party and the regulatory identification of the related party. This information is sorted and analyzed to determine the combination of the relationship between the related parties, in preparation for the subsequent revision of the regulatory identification.

[0187] 3. Determine regulatory rules and key elements:

[0188] A dedicated interpretation team carefully studies financial regulatory rules to obtain key elements that constitute a specific combination of related parties that are not within the scope of regulatory related party identification rules. For example, it is found that a certain type of related party and a specific related party relationship are not recognized under current regulatory rules. Taking a natural person shareholder as an example, his hierarchical relationship with the group company and his relationship with the middle-level related parties do not meet regulatory requirements and require regulatory identification corrections. Determine the relationship information between the natural person shareholder (related party A) and the upper-level related party, the hierarchical relationship with the middle-level related party (hierarchical jump) and the corresponding regulatory identification that needs to be corrected.

[0189] 4. Establish a related party supervision identification correction model:

[0190] Set up the key element item selection function: determine the set of relationship information items X, such as the relationship information set between the group headquarters and the first-level subsidiaries; the set of relationship information items Y, that is, the relationship information set between the first-level subsidiaries and the second-level subordinate related parties, and delete the reverse related person relationship. If the spouse of the group leader holds a position in a subsidiary, the subsidiary will no longer establish a reverse conduction relationship with the group leader to avoid building a directed cyclic graph. Determine the set of relationship information items Z, such as the relationship information set between the second-level related parties and the subordinate related parties, and clarify the hierarchical relationship between Z and Y. Set the corresponding regulatory identification that needs to be revised according to the actual situation.

[0191] Measuring the shortest path based on the Dijkstra shortest path algorithm with adjusted limited path step length: Using the Dijkstra shortest path algorithm with adjusted limited path step length, measure the shortest path from the group headquarters (direct related person) to each related party. For example, under the traditional method, the number of combinations automatically generated by the system may increase geometrically as the transmission path becomes longer and the regulatory mark requirements increase, resulting in data redundancy. After adopting the new algorithm, if the regulatory transmission relationship only considers transmission level 3, the algorithm automatically deletes excessively long transmission paths according to the step length strategy, effectively reducing the regulatory transmission path relationship. Quickly locate the relationship and hierarchical step length between the group headquarters and each related party.

[0192] Automatically combine related relationships to generate specific related relationship combinations: The system automatically combines various related relationships to generate specific related relationship combinations that are not within the scope of the regulatory related party identification rules. For example, by analyzing the relationship information of various related parties, it is found that some related party combinations do not meet regulatory requirements, and these combinations are marked.

[0193] 5. Modification and penetration identification of related party regulatory identification: Automatically check the information of all related parties. If it is found that the combination of related party information of a certain related party meets the specific combination of related parties that is not within the scope of the regulatory related party identification rules, the system will automatically modify the regulatory identification of the related party. For example, for the above-mentioned natural person shareholders, the system determines that they are not related parties within the current regulatory identification scope based on the specific combination of related parties, and automatically reduces their regulatory identification. If the result of the correction causes a regulatory identification to be added to a related party that was not calculated according to the original patented method, the penetration identification of related parties that meet a certain regulatory identification standard is achieved.

[0194] 6. Adjustment of preset values ​​and modification of graphs: Determine the range of preset values ​​according to the regulatory intensity level carried by the regulatory information of related parties. For example, in a period of stricter financial regulation, the regulatory intensity level increases and the preset values ​​decrease accordingly, making the hierarchical path step length in the hierarchical graph of the relationship shorter and the relationship closer.

[0195] The associations corresponding to the hierarchical path steps exceeding the preset value in the association hierarchical map are deleted to obtain the revised association hierarchical map of the associated persons. For example, for some related parties with distant hierarchical relationships and not very important under the current regulatory requirements, their associations with the upper-level related parties are deleted to simplify the map and improve the pertinence and accuracy of supervision.

[0196] 3. Implementation effect: Through the above application examples, the financial group company can realize the automatic correction of the regulatory identification of related parties, reduce the manual workload, and improve the accuracy of the identification of regulatory identification of related parties. At the same time, by continuously adjusting and optimizing the hierarchical map of related relationships, it ensures that the company always meets financial regulatory requirements, reduces management costs, improves the quality of regulatory data submitted, and effectively avoids the risk of misreporting of transaction data involving related parties and regulatory penalties.

[0197] Of course, it is understandable that the above detailed process may have other variations, and the relevant variations should all fall within the protection scope of the present invention.

[0198] Second embodiment:

[0199] 1. During the declaration of related parties, when entering information, you need to select the direct relationship between the related party and the related party collection agency, or select its superior related party, select the related relationship and other information to form a related relationship map of the upper and lower levels of the related relationship;

[0200] 2. The related party obtains one or more regulatory agency identifiers, calculates one or more association relationship combination information of the above regulatory identifiers, and virtual wire information from multiple upper-level related parties to related party A formed by the above relationship combinations;

[0201] 3. Interpret the regulatory rules and obtain the key elements that constitute a specific combination of related relationships that are not within the scope of the regulatory related party identification rules. The elements include: the relationship information between the N-level related party and the N-1-level related party, the relationship information between the related party A and the superior related party, the hierarchical relationship between the related party A and the N-1-level related party (close hierarchy or hierarchical jump), and the corresponding regulatory identification that needs to be revised.

[0202] For example, the child F of the direct related person E needs to be identified as a related party. According to the existing regulatory mark transmission model, the grandchild G of E can also be identified as a related party by choosing the relationship with F (child) to inherit F's regulatory mark. However, under a certain regulatory rule definition, the grandchild G of the direct related person E does not fall within the definition of a related party. According to the analysis of the regulatory related party identification rule, it can be obtained that the relationship information between the child F (N-level related party) and the direct related person E (N-1-level related party) is a child, the relationship information between the grandchild G (related party A) and the child F (the superior related party of related party A) is a child, and the hierarchical relationship between the grandchild G (related party A) and the direct related person E (N-1-level related party) is a hierarchical jump.

[0203] 4. Establish a related party supervision identification correction model:

[0204] 4.1. Set up the key element selection function

[0205] The optional elements include: association information item set X (i.e., the association information set between the 0th level associated party (direct associated party) and the 1st level associated party), association information item set Y (i.e., the association information set between the 1st level associated party and the 2nd level lower associated party, and the reverse associated person relationship is deleted at the same time, as follows Figure 1 As shown, if the next-level spouse and wife A of the direct associated person P, and the husband P corresponding to the associated person direction of A, then the directed relationship will no longer be established in reverse, avoiding the construction of a directed cycle graph, thereby constructing a directed graph), the hierarchical relationship between Y and X (close hierarchy or hierarchical jump); the association relationship information item set Z (i.e., the association relationship information set between the second-level associated party and the lower-level associated party), the hierarchical relationship between Z and Y (close hierarchy or hierarchical jump), corresponding to the regulatory identification that needs to be revised (reduced or increased). The above X, Y, and Z can be a single association relationship information, or multiple association relationship information, and the association relationship information item set M can be added under the association relationship information item set Z, as well as the hierarchical relationship between M and Z, to infinity.

[0206] 4.2. The Dijkstra shortest path algorithm that adjusts the limited path step can quickly respond to regulatory requirements in constructing the topological map path of the direct related person and the M-level related person information. For example, when the regulatory transmission relationship only considers the transmission level 3, the algorithm can automatically delete the overly long transmission path (such as the P->A->B->C part in path 1) according to the step strategy, effectively reducing the regulatory transmission path relationship and regulatory relationship path. At the same time, the algorithm can quickly find the shortest path between the direct related person and the target related party (such as the shortest path between P and D is path 5: P->B->D), so as to quickly locate the transmission related party relationship and the hierarchical step. In addition, the algorithm can also quickly respond to changes in regulatory signs. For example, when the supervision does not include the grandson in the supervision, the transmission relationship can be quickly adjusted to remove the corresponding supervision sign, and then the final regulatory related party sign can be quickly located to achieve regulatory penetration identification.

[0207] The specific operation is as follows: Based on the Dijkstra shortest path algorithm with adjusted limited path step length, the shortest path from directly related persons (legal persons, natural persons, controlling shareholders, etc.) to X, Y, and Z is measured.

[0208] The existing technology generally selects the above-mentioned key elements and constructs the set combination relationship based on the existing regulatory sign transmission model using the set hierarchy. The system automatically generates the number of combinations = X number × Y number × Z number × M number × ... This method may cause the set information data to grow exponentially as the transmission path becomes longer and the regulatory sign requirements increase. At the same time, it will also complicate the association relationship and cause data redundancy.

[0209] Construct the topological path of the direct associated person and M-level associated person information, such as Figure 5 As shown in the figure, a specific example of a method for determining the association relationship of an associated person, the regulatory transmission relationship between the traditional association relationship set D and P can have multiple paths, path 1: P->A->B->C->D, path 2: P->A->B->D, path 3: P->B->C->D, path 4: P->B->C->D, path 5: P->B->D.

[0210] The dijkstra shortest path algorithm method of adjusting the limited path step length in the present invention can quickly respond to regulatory requirements. For example, if the regulatory transmission relationship only considers transmission level 3, the algorithm automatically follows the step length strategy path 1: P->A->B->C and no longer passes on, deletes this transmission path, effectively reduces the P regulatory transmission path relationship, and reduces the regulatory relationship path. At the same time, through the dijkstra shortest path algorithm, the shortest path between P and D is quickly found as path 5: P->B->D, and the P transmission related party relationship and hierarchical step length are quickly located. In addition, changes in regulatory signs can be quickly responded to. If the supervision does not include the grandson in the supervision, the son transmission relationship that appears twice is quickly detected through the algorithm, and the supervision association sign is removed to reduce the next transmission relationship. In this way, the final regulatory related party sign of P is quickly located, thereby ultimately achieving regulatory penetration identification.

[0211] 4.3. Automatic combination of related relationships generates specific related relationship combinations that are not within the scope of the regulatory related party identification rules.

[0212] The regulatory identification of related party A is modified (reduced or increased): the system automatically checks the entire related party information. If it is found that the relationship combination information of related party A meets the above-mentioned specific relationship combination that is not within the scope of the regulatory related party identification rules, the system automatically modifies (reduced or increased) the regulatory identification of related party A.

[0213] 5. Realize penetration identification.

[0214] If the result of the above system automatically correcting (reducing or increasing) the regulatory identification of related party A leads to the addition of a regulatory identification that is not calculated by the existing technical method for related party A, the related party that meets a certain regulatory recognition standard is identified through penetration. The related party regulatory identification correction model in the present invention, and the calculation method that uses the correction model application elements to exhaust the specific association relationship combinations that are not within the scope of the regulatory related party identification rules, enable the system to automatically correct (reduce or increase) the regulatory identification of related parties, ensure the accuracy, completeness and timeliness of the data reported by related parties, and ensure the accuracy of the regulatory attribution information of transactions involving related parties. It makes up for the defects of the existing related party regulatory mark calculation model, reduces management costs, and improves the quality of reported regulatory data.

[0215] As described above, the second embodiment has the following beneficial effects:

[0216] 1. By processing related party relationships through the Dijkstra shortest path algorithm, a model for transmitting specific related relationships that are not within the scope of regulatory related party identification rules is implemented through the system, reducing system data dimension disasters caused by exhaustive combinations.

[0217] 2. By using the Dijkstra shortest path algorithm with a limited path step length to handle related party relationships, the path step length, weight, etc. can be used to quickly discard related penetration conduction paths, reduce the Dijkstra algorithm's drill-down related relationships, effectively reduce the algorithm path overhead, and quickly adapt to regulatory changes. At the same time, it effectively improves the performance of the entire algorithm and realizes the rapid positioning of related party relationships.

[0218] 3. By limiting the path step length and using the Dijkstra shortest path algorithm to process related party relationships, the shortest path between the direct related party and the next N levels of related parties can be effectively established. Especially when the related party transmission path is complex, redundant paths can be reduced, and the most effective related party relationship set can be established. Transactions involving related parties can be reduced through multi-layer transmission to avoid the transaction identification risks involving related parties brought about by hidden related parties.

[0219] 4. Construct a directed graph of related parties to avoid reverse cycle relationships, and quickly and effectively realize the related party relationship transmission topology graph.

[0220] 5. By limiting the path step length, Dijkstra can realize the system's automatic correction (reduction or increase) of the regulatory identification of related parties, establish a reliable, flexible and scalable related party identification model, improve the accuracy of the system's identification of related parties, and ensure the quality of regulatory reporting.

[0221] The present invention also provides a device for determining the association relationship of associated persons, as described in the following embodiments. Since the principle of solving the problem by the device is similar to that of the method for determining the association relationship of associated persons, the implementation of the device can refer to the implementation of the method for determining the association relationship of associated persons, and the repeated parts will not be repeated.

[0222] The embodiment of the present invention also provides a device for determining the association relationship of an associated person, which is used to improve the recognition accuracy and efficiency of the associated person's supervision identification, simplify the association relationship of the associated person, avoid data redundancy, and ensure the accuracy and timeliness of the supervision reporting data. Figure 6 A schematic diagram of a structural diagram of a device for determining an associated relationship of associated persons is shown, the device comprising:

[0223] A graph determination module 601 is used to determine a hierarchical graph of association relationships of an associated person; the hierarchical graph of association relationships is marked with one or more regulatory agency identifiers of the associated person; the regulatory agency identifiers are used to characterize the association elements of the associated person; the association elements include each associated party of the associated person, and the hierarchy and association relationship between the associated person and each of its associated parties;

[0224] The text recognition processing module 602 is used to perform text recognition processing on the received related party supervision information to obtain the specified association relationship elements; the related party supervision information is used to describe the information that meets the specified related party identification rules;

[0225] The existing association element determination module 603 is used to identify the existing association elements of the existing association relationship corresponding to the regulatory agency identifier of the associated person; match the existing association elements with the specified association relationship elements; and treat the existing association elements that are not successfully matched as association elements to be corrected that do not comply with the specified associated party identification rules;

[0226] A regulatory agency identification correction module 604 is used to correct one or more regulatory agency identifications of the associated person according to the associated elements to be corrected, so as to obtain a corrected regulatory identification of the associated party;

[0227] A hierarchical path step length determination module 605 is used to determine the hierarchical path step length between the associated person and each associated party corresponding to the revised associated party supervision identifier;

[0228] The graph correction module 606 is used to delete the association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical graph, so as to obtain a corrected association relationship hierarchical graph of the associated persons.

[0229] In one embodiment, determining the association relationship hierarchy map of the associated persons includes:

[0230] Constructing a hierarchical graph of the associated relationships of the associated person according to the associated party input information of the associated person; the associated party input information includes the associated person's superior associated party and the type of the associated relationship between the associated person and its superior associated party;

[0231] Based on the regulatory identification configuration information, one or more regulatory agency identifications of the associated persons in the association relationship hierarchy map are generated.

[0232] In one embodiment, the received related party supervision information is subjected to text recognition processing to obtain the specified association relationship elements, including:

[0233] The semantic analysis algorithm and semantic role labeling technology in natural language processing technology are used to perform text recognition processing on the received related party supervision information to obtain specified association relationship elements; the semantic analysis algorithm is used to parse the semantic structure in the supervision information; the semantic role labeling technology is used to analyze the semantic roles in the supervision information to determine the related person entity, related party entity and association relationship entity.

[0234] In one embodiment, determining the hierarchical path step between the associated person and each associated party corresponding to the modified associated party supervision identifier includes:

[0235] The shortest path algorithm and heuristic search algorithm in graph theory are used to determine the hierarchical path step between the related person and the related party corresponding to each revised related party regulatory identification.

[0236] In one embodiment, the association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical map are deleted to obtain a revised association relationship hierarchical map of the associated persons, including:

[0237] Repeat the following operations until there are no more relationships in the graph whose hierarchical path step length exceeds the preset value, and obtain the revised hierarchical graph of the relationship between the associated persons:

[0238] For each regulatory agency identifier, determining, based on a greedy algorithm, relationship attributes of different association relationships corresponding to the regulatory agency identifier; the relationship attributes include a connection frequency of the association relationship;

[0239] According to the relationship attributes, the different association relationships corresponding to the regulatory agency identification are selectively modified.

[0240] In one embodiment, the association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical map are deleted to obtain a revised association relationship hierarchical map of the associated persons, including:

[0241] Determining the range of the preset value according to the supervision intensity level carried by the related party supervision information;

[0242] The association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical map are deleted to obtain a revised association relationship hierarchical map of the associated persons.

[0243] In an embodiment of the present invention, a hierarchical map of association relationships of an associated person is determined; one or more regulatory agency identifiers of the associated person are marked in the hierarchical map of association relationships; the regulatory agency identifier is used to characterize the association elements of the associated person; the association elements include each associated party of the associated person, and the hierarchy and association relationship between the associated person and each of its associated parties; the received associated party regulatory information is subjected to text recognition processing to obtain a specified association relationship element; the associated party regulatory information is used to describe information that complies with the specified associated party identification rule; the existing association elements of the existing association relationship corresponding to the regulatory agency identifier of the associated person are identified; the existing association elements are matched with the specified association relationship elements; the existing association elements that are not successfully matched are used as the association elements to be corrected that do not comply with the specified associated party identification rule; according to the association elements to be corrected, one or more regulatory agency identifiers of the associated person are corrected to obtain the corrected associated party regulatory identifier; the hierarchical path step between the associated person and the associated party corresponding to each corrected associated party regulatory identifier is determined; the association corresponding to the hierarchical path step that exceeds a preset value in the hierarchical map of association relationships is deleted to obtain the corrected hierarchical map of association relationships of the associated person, thereby The system can identify the related elements to be corrected that do not meet the specified related party identification rules and correct them to ensure the accuracy of the regulatory identification, solving the problem that the specific combination of related relationships is not within the scope of the regulatory identification rules under the existing technology, resulting in incorrect calculation of the regulatory identification; the received related party regulatory information is processed by text recognition in an automated manner, the existing related elements are automatically matched, and the elements that have not been successfully matched are corrected, reducing manual intervention, improving efficiency and accuracy, and solving the problem that the existing technology requires manual search and modification of the related relationships in the related party map one by one and layer by layer, which is labor-intensive, time-consuming and easy to miss; by determining the hierarchical path step and deleting the path that exceeds the preset value, reducing redundant paths, simplifying the relationship map, improving the performance and maintainability of the relationship system, solving the problem that the data is redundant due to the geometric growth of data and the complexity of the relationship as the transmission path becomes longer and the requirements of the regulatory mark change under the existing technology; by dynamically adjusting and correcting the regulatory mark, it can quickly respond to changes in regulatory rules to ensure the accuracy and timeliness of regulatory reporting data, solving the problem that it is difficult to quickly adapt to changes in regulatory rules under the existing technology, resulting in the inability to update regulatory marks in a timely manner and affecting the quality of reported data.

[0244] The embodiment of the present invention provides an embodiment of a computer device for implementing all or part of the contents of the above-mentioned method for determining the association relationship of associated persons. The computer device specifically includes the following contents:

[0245] A processor, a memory, a communications interface, and a bus; wherein the processor, the memory, and the communications interface communicate with each other through the bus; the communications interface is used to realize information transmission between related devices; the computer device can be a desktop computer, a tablet computer, a mobile terminal, etc., but the present embodiment is not limited thereto. In the present embodiment, the computer device can be implemented with reference to the embodiment of the method for realizing the association relationship determination of associated persons and the embodiment of the device for realizing the association relationship determination of associated persons, and the contents thereof are incorporated herein, and the repeated parts are not repeated.

[0246] Figure 7 1 is a schematic block diagram of the system structure of the computer device 1000 according to an embodiment of the present application. Figure 7 As shown, the computer device 1000 may include a central processor 1001 and a memory 1002; the memory 1002 is coupled to the central processor 1001. It is worth noting that Figure 7 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0247] In one embodiment, the association relationship determination function of the associated person may be integrated into the central processor 1001. The central processor 1001 may be configured to perform the following control:

[0248] Determine a hierarchical map of the associated relationships of the associated person; the hierarchical map of the associated relationships is marked with one or more regulatory agency identifiers of the associated person; the regulatory agency identifiers are used to characterize the associated elements of the associated person; the associated elements include each associated party of the associated person, and the hierarchy and associated relationship between the associated person and each of its associated parties;

[0249] Performing text recognition processing on the received related party supervision information to obtain designated related relationship elements; the related party supervision information is used to describe information that complies with designated related party identification rules;

[0250] Identify the existing association elements of the existing association relationship corresponding to the regulatory agency identification of the associated person; match the existing association elements with the designated association relationship elements; and treat the existing association elements that have not been successfully matched as association elements to be corrected that do not comply with the designated associated party identification rules;

[0251] According to the association elements to be amended, one or more regulatory agency identifiers of the associated person are amended to obtain amended regulatory identifiers of the associated party;

[0252] Determine the hierarchical path step between the related party and each related party corresponding to the revised related party regulatory identification;

[0253] The association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical map are deleted to obtain a revised association relationship hierarchical map of the associated persons.

[0254] In another embodiment, the associated person's associated relationship determination device can be configured separately from the central processor 1001. For example, the associated person's associated relationship determination device can be configured as a chip connected to the central processor 1001, and the associated person's associated relationship determination function is implemented through the control of the central processor.

[0255] like Figure 7 As shown, the computer device 1000 may also include: a communication module 601003, an input unit 1004, an audio processor 1005, a display 1006, and a power supply 1007. It is worth noting that the computer device 1000 does not necessarily have to include Figure 7 In addition, the computer device 1000 may also include Figure 7 For components not shown, reference may be made to the prior art.

[0256] like Figure 7 As shown, the central processor 1001 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor devices and / or logic devices. The central processor 1001 receives inputs and controls the operations of various components of the computer device 1000 .

[0257] The memory 1002 may be, for example, a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices, or one or more thereof. The memory may store the above-mentioned information related to the device, and may also store a program for executing the related information. The CPU 1001 may execute the program stored in the memory 1002 to implement information storage or processing, etc.

[0258] The input unit 1004 provides input to the CPU 1001. The input unit 1004 is, for example, a key or a touch input device. The power supply 1007 is used to provide power to the computer device 1000. The display 1006 is used to display display objects such as images and text. The display may be, for example, an LCD display, but is not limited thereto.

[0259] The memory 1002 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 1002 may also be some other type of device. The memory 1002 includes a buffer memory 1021 (sometimes referred to as a buffer). The memory 1002 may include an application / function storage unit 1022, which is used to store application programs and function programs or processes for executing the operation of the computer device 1000 through the central processor 1001.

[0260] The memory 1002 may also include a data storage unit 1023 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the computer device. The driver storage unit 1024 of the memory 1002 may include various drivers for the computer device for communication functions and / or for executing other functions of the computer device (such as messaging applications, address book applications, etc.).

[0261] The communication module 1003 is a transmitter / receiver that sends and receives signals via the antenna 1008. The communication module (transmitter / receiver) 1003 is coupled to the central processor 1001 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0262] Based on different communication technologies, multiple communication modules 1003 may be provided in the same computer device, such as a cellular network module, a Bluetooth module and / or a wireless local area network module, etc. The communication module (transmitter / receiver) 1003 is also coupled to a speaker 1009 and a microphone 1010 via an audio processor 1005 to provide an audio output via the speaker 1009 and receive an audio input from the microphone 1010, thereby realizing a common telecommunication function. The audio processor 1005 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 1005 is also coupled to the central processor 1001, so that recording can be performed on the local machine through the microphone 1010, and the sound stored on the local machine can be played through the speaker 1009.

[0263] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for determining the association relationship of associated persons is implemented.

[0264] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the method for determining the association relationship of associated persons is implemented.

[0265] In an embodiment of the present invention, a hierarchical map of association relationships of an associated person is determined; one or more regulatory agency identifiers of the associated person are marked in the hierarchical map of association relationships; the regulatory agency identifier is used to characterize the association elements of the associated person; the association elements include each associated party of the associated person, and the hierarchy and association relationship between the associated person and each of its associated parties; the received associated party regulatory information is subjected to text recognition processing to obtain a specified association relationship element; the associated party regulatory information is used to describe information that complies with the specified associated party identification rule; the existing association elements of the existing association relationship corresponding to the regulatory agency identifier of the associated person are identified; the existing association elements are matched with the specified association relationship elements; the existing association elements that are not successfully matched are used as the association elements to be corrected that do not comply with the specified associated party identification rule; according to the association elements to be corrected, one or more regulatory agency identifiers of the associated person are corrected to obtain the corrected associated party regulatory identifier; the hierarchical path step between the associated person and the associated party corresponding to each corrected associated party regulatory identifier is determined; the association corresponding to the hierarchical path step that exceeds a preset value in the hierarchical map of association relationships is deleted to obtain the corrected hierarchical map of association relationships of the associated person, thereby The system can identify the related elements to be corrected that do not meet the specified related party identification rules and correct them to ensure the accuracy of the regulatory identification, solving the problem that the specific combination of related relationships is not within the scope of the regulatory identification rules under the existing technology, resulting in incorrect calculation of the regulatory identification; the received related party regulatory information is processed by text recognition in an automated manner, the existing related elements are automatically matched, and the elements that have not been successfully matched are corrected, reducing manual intervention, improving efficiency and accuracy, and solving the problem that the existing technology requires manual search and modification of the related relationships in the related party map one by one and layer by layer, which is labor-intensive, time-consuming and easy to miss; by determining the hierarchical path step and deleting the path that exceeds the preset value, reducing redundant paths, simplifying the relationship map, improving the performance and maintainability of the relationship system, solving the problem that the data is redundant due to the geometric growth of data and the complexity of the relationship as the transmission path becomes longer and the requirements of the regulatory mark change under the existing technology; by dynamically adjusting and correcting the regulatory mark, it can quickly respond to changes in regulatory rules to ensure the accuracy and timeliness of regulatory reporting data, solving the problem that it is difficult to quickly adapt to changes in regulatory rules under the existing technology, resulting in the inability to update regulatory marks in a timely manner and affecting the quality of reported data.

[0266] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0267] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0268] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0269] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0270] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for determining the association relationship of an associated person, characterized in that: include: Determine the relationship hierarchy of the associated persons; One or more regulatory agency identifiers of associated persons are marked in the association relationship hierarchy map; The regulatory agency identifier is used to characterize the associated elements of the associated person; the associated elements include each associated party of the associated person, and the hierarchy and associated relationship between the associated person and each associated party; Performing text recognition processing on the received related party supervision information to obtain designated related relationship elements; the related party supervision information is used to describe information that complies with designated related party identification rules; Identify the existing association elements of the existing association relationship corresponding to the regulatory agency identification of the associated person; match the existing association elements with the designated association relationship elements; and treat the existing association elements that have not been successfully matched as association elements to be corrected that do not comply with the designated associated party identification rules; According to the association elements to be amended, one or more regulatory agency identifiers of the associated person are amended to obtain amended regulatory identifiers of the associated party; Determine the hierarchical path step between the related party and each related party corresponding to the revised related party regulatory identification; The association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical map are deleted to obtain a revised association relationship hierarchical map of the associated persons.

2. The method according to claim 1, characterized in that Determine the relationship hierarchy of the associated person, including: Constructing a hierarchical graph of the associated relationships of the associated person according to the associated party input information of the associated person; the associated party input information includes the associated person's superior associated party and the type of the associated relationship between the associated person and its superior associated party; Based on the regulatory identification configuration information, one or more regulatory agency identifications of the associated persons in the association relationship hierarchy map are generated.

3. The method according to claim 1, characterized in that Perform text recognition processing on the received related party supervision information to obtain the specified relationship elements, including: The semantic analysis algorithm and semantic role labeling technology in natural language processing technology are used to perform text recognition processing on the received related party supervision information to obtain specified association relationship elements; the semantic analysis algorithm is used to parse the semantic structure in the supervision information; the semantic role labeling technology is used to analyze the semantic roles in the supervision information to determine the related person entity, related party entity and association relationship entity.

4. The method according to claim 1, characterized in that Determine the hierarchical path step between the related party and each related party corresponding to the revised related party regulatory identification, including: The shortest path algorithm and heuristic search algorithm in graph theory are used to determine the hierarchical path step between the related person and the related party corresponding to each revised related party regulatory identification.

5. The method according to claim 1, characterized in that Deleting the associations corresponding to the hierarchical path steps exceeding a preset value in the association hierarchical graph to obtain a revised association hierarchical graph of the associated persons includes: repeatedly performing the following operations until there are no more associations whose hierarchical path steps exceed the preset value in the graph to obtain a revised association hierarchical graph of the associated persons: For each regulatory agency identifier, determining, based on a greedy algorithm, relationship attributes of different association relationships corresponding to the regulatory agency identifier; the relationship attributes include a connection frequency of the association relationship; Based on the relationship attributes, different association relationships corresponding to the regulatory agency identifier are selectively modified.

6. The method according to claim 1, characterized in that The associations corresponding to the hierarchical path steps exceeding the preset value in the association hierarchical map are deleted to obtain a revised association hierarchical map of the associated persons, including: Determining the range of the preset value according to the supervision intensity level carried by the related party supervision information; The association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical map are deleted to obtain a revised association relationship hierarchical map of the associated persons.

7. A device for determining an associated relationship of an associated person, characterized in that: include: A graph determination module is used to determine a hierarchical graph of association relationships of an associated person; the hierarchical graph of association relationships is marked with one or more regulatory agency identifiers of the associated person; the regulatory agency identifiers are used to characterize association elements of the associated person; the association elements include each associated party of the associated person, and the hierarchy and association relationship between the associated person and each of its associated parties; A text recognition processing module is used to perform text recognition processing on the received related party supervision information to obtain the specified association relationship elements; the related party supervision information is used to describe the information that meets the specified related party identification rules; An existing association element determination module is used to identify the existing association elements of the existing association relationship corresponding to the regulatory agency identifier of the associated person; match the existing association elements with the specified association relationship elements; and treat the existing association elements that have not been successfully matched as association elements to be corrected that do not comply with the specified associated party identification rules; A regulatory agency identification correction module, used to correct one or more regulatory agency identifications of the associated person according to the associated elements to be corrected, to obtain a corrected regulatory identification of the associated party; A hierarchical path step length determination module, used to determine the hierarchical path step length between the associated person and each associated party corresponding to the revised associated party supervision identifier; The graph correction module is used to delete the association relationships corresponding to the hierarchical path steps exceeding a preset value in the association relationship hierarchical graph, so as to obtain a corrected association relationship hierarchical graph of the associated persons.

8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images