Risk identification method and device, storage medium and electronic equipment
By constructing and updating the business map, identifying the correlation between the accounts to be identified and other accounts, the problem of difficulty in identifying risk accounts in complex trading networks in the prior art is solved, and more efficient risk identification and feature extraction is achieved.
Patent Information
- Application Number
- CN202410921101.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-09
- Publication Date
- 2024-11-08
AI Technical Summary
It is difficult for the prior art to quickly and accurately identify nodes that realize risk transactions through complex trading networks, especially in the decentralized third-generation Internet, where the recognition efficiency and accuracy of risk accounts are low.
By constructing a business map, the correlation between the account to be identified and other accounts is determined, the business map is updated until the preset conditions are met, and the characteristics of the account to be identified are determined, thereby performing risk identification. The specific steps include determining that the account to be identified and other accounts are nodes, building a business graph, updating the target account with a correlation degree greater than the preset value, until the update end condition is reached, and finally determining the characteristics of the account to be identified based on the business graph for risk identification.
It improves the accuracy and efficiency of account identification for risk services through complex transaction networks, reduces manual participation, saves computing resources, and improves the efficiency of feature extraction and risk identification.
Smart Images

Figure CN118917850A_ABST
Abstract
Description
Technical Field
[0001] This specification relates to the field of computers, and in particular to a risk identification method, device, storage medium, and electronic device. Background Art
[0002] With the development of computer technology, the Internet is moving towards the decentralized third generation Internet, which uses decentralized technologies such as distributed networks and blockchain to reduce the risks caused by data concentration. However, due to its decentralized and anonymized characteristics, risky accounts are more difficult to identify.
[0003] In the prior art, risk identification of an account is usually performed based on the features of an account by extracting features from an account. Specifically, taking the risk identification of each node in a blockchain as an example, the transaction records of the node to be identified within a period of time are usually randomly sampled, and the features of the sampled transaction records are extracted, and then the features of the node to be identified are determined, and the features of the node to be identified and the features of the transaction records are fused as the representation of the node to be identified. Then, through model construction, the risk identification of the node to be identified is performed based on the representation of the node to be identified, which prevents the leakage of personal information while achieving risk identification of the node to be identified.
[0004] However, when there are nodes that implement risky transactions through complex transaction networks, or nodes that implement risky transactions using new methods, the existing technology cannot quickly and accurately identify risky nodes. Therefore, this specification provides a risk identification method, device and computer system. Summary of the invention
[0005] This specification provides a risk identification method, device, medium and equipment to partially solve the above-mentioned problems existing in the prior art.
[0006] This manual adopts the following technical solutions:
[0007] This manual provides a risk identification method, including:
[0008] Determine the account to be identified, and construct a business graph using other accounts that have executed business with the account to be identified and the account to be identified as nodes, and business data of business executed between nodes as edges;
[0009] Other accounts that have performed business with the account to be identified are used as extended accounts, and for each extended account, other accounts that have performed business with the extended account are determined as target accounts, and a first degree of association between the target account and the extended account is determined;
[0010] Determine a target account with a first association degree greater than a first preset value as a node, and update the business graph with business data of executing business between the nodes as an edge;
[0011] Re-using the target account with the first correlation degree greater than the first preset value as the extended account, continuing to update the business graph until the update end condition is met, determining a final business graph, and determining the characteristics of the account to be identified according to the final business graph;
[0012] Risk identification is performed on the account to be identified based on the characteristics.
[0013] Optionally, a business graph is constructed with business data of business executed between nodes as edges, specifically including:
[0014] For each other account, determine each business performed between the other account and the account to be identified;
[0015] For each determined business, an edge between the node of the other account and the node of the account to be identified in the business graph is determined according to the business data of the business.
[0016] Optionally, updating the business graph with business data of business executed between nodes as edges specifically includes:
[0017] For each target account, determine each business executed between the target account and the extended account;
[0018] For each determined business, an edge between the node of the target account and the node of the extended account in the business graph is determined according to the business data of the business.
[0019] Optionally, other accounts that have performed business with the account to be identified are used as extended accounts, specifically including:
[0020] For each other account, determining a second degree of association between the account to be identified and the other account according to business data of business executed between the other account and the account to be identified;
[0021] Other accounts whose second correlation degree is greater than a second preset value are determined as extended accounts.
[0022] Optionally, the business data includes a timestamp when the business is executed; the account that initiates the business is the upstream account, and the account that receives the business is the downstream account;
[0023] Determining a target account with a first association degree greater than a first preset value as a node, and updating the business graph with business data of executing business between the nodes as an edge, specifically includes:
[0024] For each extended account, determine the timestamp of each business initiated by the extended account to the account to be identified, and determine the largest timestamp among the timestamps of the businesses as the first timestamp;
[0025] From the upstream accounts in the target account of the extended account, determine each upstream account with a first correlation degree of the extended account greater than a first preset value as an update account;
[0026] Determine the timestamp of each business initiated by each updating account to the extended account as the first timestamp of each updating account;
[0027] Determine that there is an updated account whose second timestamp is less than the first timestamp, as a newly added node;
[0028] The edge between the newly added node and the extended account is determined based on the business data whose second timestamp is less than the first timestamp, and the business graph is updated based on the newly added node and the edge between the newly added node and the extended account.
[0029] Optionally, determining the target accounts with a first association degree greater than a first preset value as nodes, and updating the business graph with business data of executing business between the nodes as edges, specifically includes:
[0030] For each extended account, determine the timestamps of each service initiated by the account to be identified to the extended account, and determine the smallest timestamp among the timestamps of each service as the third timestamp;
[0031] From the downstream accounts in the target account of the extended account, determine each downstream account with a first correlation degree of the extended account greater than a first preset value as an update account;
[0032] Determine the timestamp of each service initiated by the extended account to each update account as the fourth timestamp of each update account;
[0033] Determine that there is an updated account whose fourth timestamp is greater than the third timestamp, as a newly added node;
[0034] The edge between the newly added node and the extended account is determined based on the business data whose fourth timestamp of the newly added node is greater than the third timestamp, and the business graph is updated based on the newly added node and the edge between the newly added node and the extended account.
[0035] Optionally, the service data includes at least one of a timestamp of executing the service, a service volume, and a service execution count;
[0036] Determining a first degree of association between the target account and the extended account specifically includes:
[0037] The first association degree between the target account and the extended account is determined according to at least one of a timestamp, a business volume, and a business execution count of the business executed between the target account and the extended account.
[0038] Optionally, until the update end condition is met, a final business graph is determined, specifically including:
[0039] Determining a third degree of association between the node corresponding to the account to be identified and other nodes in the business graph;
[0040] When at least one of the following conditions is met: the business graph reaches a preset number of hops, the third degree of association is less than a third preset value, or there is no target account with a second degree of association greater than a second preset value, the most recently updated business graph is determined as the final business graph.
[0041] Optionally, determining the characteristics of the account to be identified according to the final business graph specifically includes:
[0042] Determine the characteristics of each node according to the attributes of each node in the final business graph;
[0043] Determining the characteristics of each edge according to the business data corresponding to each edge in the final business graph;
[0044] The features of the nodes and the features of the edges are fused according to preset weights to determine fused features as the features of the account to be identified.
[0045] This specification provides a risk identification device, including:
[0046] A construction module, used to determine an account to be identified, and construct a business graph with the account to be identified and other accounts that perform business with the account to be identified as nodes and the executed business data as edges;
[0047] An updating module is used to take other accounts that have executed business with the account to be identified as extended accounts, and for each extended account, determine other accounts that have executed business with the extended account as target accounts, and determine a first degree of association between the target account and the extended account; determine the target accounts with the first degree of association greater than a first preset value as nodes, and update the business graph with business data of business executed between the nodes as edges;
[0048] a determination module, configured to re-use the target account with the first correlation degree greater than the first preset value as an extended account, continue to update the business graph until an update end condition is met, determine a final business graph, and determine the characteristics of the account to be identified according to the final business graph;
[0049] An identification module is used to identify the risk of the account to be identified based on the characteristics.
[0050] This specification provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the risk identification method described above is implemented.
[0051] This specification provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the risk identification method is implemented when the processor executes the program.
[0052] At least one of the above technical solutions adopted in this specification can achieve the following beneficial effects:
[0053] In a risk verification method described in the present specification, a business graph is constructed with the account to be identified and other accounts that perform business with the account to be identified as nodes and business data as edges. Then, based on the degree of association between the account to be identified and the other accounts, the business graph is further updated and expanded. When the business graph reaches the update end condition, the final business graph is determined, and the characteristics of the account to be identified are confirmed based on the final business graph. Then, risk identification of the account to be identified is performed based on the characteristics of the account to be identified.
[0054] It can be seen from the above method that the business graph that reaches the preset number of hops is determined by the correlation between the account to be identified and other accounts, and the characteristics of the account to be identified are determined according to the characteristics of the edges and nodes in the business graph. When risk identification is performed on the account to be identified, the node to be identified can be characterized by the characteristics and business data of the account with a higher correlation with the account to be identified. This can better identify accounts that implement risky businesses through complex transaction networks, and at the same time, automatically identify the risks of the account to be identified, reduce manual participation, improve the efficiency of feature extraction and risk identification, and save computing resources. BRIEF DESCRIPTION OF THE DRAWINGS
[0055] The drawings described herein are used to provide a further understanding of this specification and constitute a part of this specification. The exemplary embodiments and descriptions of this specification are used to explain this specification and do not constitute an improper limitation on this specification.
[0056] In the figure:
[0057] Figure 1 A schematic diagram of the process of a risk identification method provided in this manual;
[0058] Figure 2 A schematic diagram of a business diagram provided for this specification;
[0059] Figure 3 A schematic diagram of an updated business diagram provided for this specification;
[0060] Figure 4 A schematic diagram of a business diagram provided for this specification;
[0061] Figure 5 A schematic diagram of a business diagram provided for this specification;
[0062] Figure 6 A schematic diagram of a business graph update provided in this specification;
[0063] Figure 7 A schematic diagram of a business graph update provided in this specification;
[0064] Figure 8 A schematic diagram of a risk identification model provided for this specification;
[0065] Fig. 9 A schematic diagram of edge feature extraction provided in this specification;
[0066] Fig.10 A schematic diagram of a risk identification method provided for this specification;
[0067] Fig.11 A schematic diagram of a business diagram provided for this specification;
[0068] Fig.12 A schematic diagram of a risk identification device provided in this manual;
[0069] Fig.13 The corresponding Figure 1 Schematic diagram of electronic equipment. DETAILED DESCRIPTION
[0070] In order to make the purpose, technical solutions and advantages of this specification more clear, the technical solutions of this specification will be clearly and completely described in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0071] At present, the Internet is constantly moving towards the decentralized third generation Internet (Web 3.0), which uses decentralized technologies such as distributed networks and blockchain to reduce the risks caused by data concentration. At the same time, digital signatures, public key encryption and other means are used to ensure the security and immutability of data, so that accounts can interact safely without mutual trust. However, due to its trustless and decentralized characteristics, if there is an account that conducts risky business, it is difficult for other accounts in Web 3.0 to identify the risky account of the risky business, which affects Internet security. Therefore, how to accurately and quickly identify the risk of the identified account has become an urgent problem to be solved.
[0072] Commonly used risk identification methods generally identify risk accounts based on the characteristics of determined risk accounts. For example, risk accounts often have characteristics such as abnormal event execution of business, frequent address changes, etc. When an account is identified as meeting these characteristics, it can be considered that the account is a risk account. Therefore, in order to reduce the probability of being identified for executing risky business, risk accounts often execute business through interactions between multiple accounts. The more complex the transaction network, the more difficult it is to identify risk accounts. It can be seen that it is often difficult to accurately identify accounts that perform risky business through a complex business execution network by only identifying the risk of the account based on the characteristics of the account. When the account executes a new type of risky business, if the characteristics of the account in the new type of risky business execution are not determined, it will cause the existing risk identification method to be difficult to accurately identify the new type of risky business execution method within a period of time, which reduces the efficiency and accuracy of risk identification. On this basis, this specification provides a risk identification method to partially solve the above problems.
[0073] It should be noted that in one or more embodiments of this specification, the risk identification method can be an account that performs risk identification on an account that is about to conduct business exchanges to ensure the safety of the business, or it can be a management server in a centralized network that performs risk identification on each account to protect the security of each account in the centralized network. This specification does not limit the subject that executes the risk identification method, which can be a central server in a distributed system or a terminal where a node in a blockchain is located.
[0074] To facilitate subsequent descriptions, in one or more embodiments of this specification, taking the risk identification of a certain node in the blockchain for a transaction node as an example, the risk identification method is described with the device where the node is located as the execution subject. The device can be a personal computer, a mobile terminal, and a server, etc. However, since subsequent steps involve operations such as business record acquisition, model calling, feature extraction, and feature fusion, and such operations are generally performed by a server, this specification will also take the server executing the risk identification method as an example for description. The server can be a single device, or composed of multiple devices, for example, a distributed server, and this specification does not limit this.
[0075] The technical solutions provided by the embodiments of this specification are described in detail below in conjunction with the accompanying drawings.
[0076] Figure 1 A schematic diagram of the process of a risk identification method provided in this manual, specifically:
[0077] S100: Determine an account to be identified, and construct a business graph with the account to be identified and other accounts that perform business with the account to be identified as nodes and business execution data as edges.
[0078] Since it is difficult to accurately determine whether the account is a risk account based only on the attributes of the account to be identified, in order to better identify the account to be identified, the business performed by the account to be identified within a preset time should also be determined, and the account to be identified should be characterized based on the business performed by the account to be identified within a preset time and other accounts that perform business with the account to be identified. In one or more embodiments of the present specification, a method for constructing a business graph based on the business performed by the account to be identified within a preset time and other accounts that perform business with the account to be identified can be used to extract features for characterizing the node to be identified.
[0079] Specifically, the server first determines the business data of the account to be identified and other accounts that perform business for the account to be identified. In one or more embodiments of this specification, the server is not limited to the method by which it determines the business data of the account to be identified. For example, if the account to be identified and the account corresponding to the server are both nodes in the blockchain, the server can directly search for the transaction record of the account to be identified on the blockchain as the business data of the account to be identified. If the account to be identified and the account corresponding to the server are two servers in a distributed server cluster, the server can obtain the business data of the account to be identified through the management server. Of course, the business data of the account to be identified can also be obtained through other methods. Different acquisition methods can be adopted according to actual conditions.
[0080] Then, other accounts that have performed business with the account to be identified and the account to be identified are used as nodes, and the business data of the business performed between the nodes are used as edges to construct a business graph, such as Figure 2 As shown, Figure 2 A schematic diagram of a business graph provided for this specification, wherein a circle with a symbol represents a node, each node corresponds to an account, and each edge represents business data between two nodes.
[0081] It should be noted that in one or more embodiments of this specification, the other accounts that perform business with the node to be identified are accounts that have direct business connections with the account to be identified. Of course, the specific content of the business performed between accounts is not limited, and it can be resource conversion, resource allocation, etc., such as node a1 allocating its own computing resources to node a2. The above-mentioned business data is a collection of business data between the account to be identified and other accounts within a preset time. The business data includes at least one of the timestamp of executing the business, the business volume, and the number of business executions.
[0082] S102: other accounts that have executed business with the account to be identified are taken as extended accounts, and for each extended account, other accounts that have executed business with the extended account are determined as target accounts, and a first association degree between the target account and the extended account is determined.
[0083] In order to further improve the precision and accuracy of risk identification and better identify risky accounts that implement risky businesses through complex business networks, it is necessary to better characterize the node to be identified. Therefore, it is necessary to further update the business graph based on the businesses performed by other accounts that have performed businesses with the account to be identified, so that the account to be identified can be subsequently characterized based on the characteristics of other nodes in the updated business graph.
[0084] Specifically, first, other accounts that have performed business with the account to be identified are used as extended accounts.
[0085] Then, for each extended account, other accounts that have performed business with the extended account are determined as target accounts. It should be noted that when determining other accounts that have performed business with the extended account, it is not restricted whether other accounts that have performed business with the extended account have had business exchanges with the account to be identified. For example, account a is the account to be identified, and account b and account c are other accounts of the account to be identified that have performed business with account a. Then, the business graph is updated based on account b and account c, and it is determined that account b and account c have also performed business. In this case, account c can also be used as the target account of account b, but account c is also another account of account a to be identified.
[0086] Finally, in order to improve the efficiency of risk identification and save computing resources, when updating the business graph, the server can also determine the business execution status between the target account and the corresponding extended account based on the network connection and interaction frequency, cooperation strategy, etc. between the two accounts, that is, determine the association between the target account and the corresponding extended account, and use the target account with a higher association as the node to be updated in the business graph. Specifically, based on the business data between the target account and the extended account, the first association between the target account and the extended account is determined. Of course, in one or more embodiments of this specification, there is no restriction on the specific method used to calculate the first association between the target account and the extended account.
[0087] The first degree of association between the target account and the extended account can be determined through the business data between the target account and the extended account, for example, by determining the number of business executions between the target account and the extended account, whether the active time periods overlap, or the business content, etc. The first degree of association between the target account and the extended account can also be determined by fusing the business data between the target account and the extended account according to a certain weight through a trained model, and scoring the fusion result through a trained scoring model, thereby determining the first degree of association between the target account and the extended account. Since there are many methods that can be used, they are not listed here one by one.
[0088] S104: Determine the target accounts with a first correlation degree greater than a first preset value as nodes, and update the business graph with the business data of the businesses executed between the nodes as edges.
[0089] In order to better characterize the account to be identified and determine a suitable business graph, the server determines a target account with a first correlation degree greater than a first preset value as a node, and updates the business graph with business data of the business executed between the nodes as an edge.
[0090] Specifically, Figure 3 As shown, Figure 3 A schematic diagram of an updated business graph provided in this specification, wherein each circle represents a node, the line between the nodes is an edge, which is used to characterize the business between two accounts, the black-filled nodes represent nodes whose first correlation is greater than the first preset value, the node b1 is a node to be identified, the white-filled nodes represent nodes whose first correlation is less than the first preset value, the dotted edges and dotted circles represent nodes whose first correlation is less than the first preset value, and their corresponding businesses. The dotted edges and dotted circles will be clipped in the actual updated business graph.
[0091] It should be noted that the above-mentioned first preset value can be determined according to actual conditions, and can be equal to the second preset value, or a value different from the second preset value. If the required accuracy of risk identification is high, the first preset value can be smaller. If the required accuracy of risk identification is low and high efficiency is required, the first preset value can be set larger. This manual does not limit the method for determining the first preset value.
[0092] S106: Re-select the target account with the first correlation degree greater than the first preset value as the extended account, continue to update the business graph until the update end condition is met, determine the final business graph, and determine the characteristics of the account to be identified based on the final business graph.
[0093] After completing a round of updating the business graph, in order to further determine a suitable business graph and better complete the graph feature propagation of the account to be identified according to the business graph, the server may re-select the target account with a first correlation degree greater than the first preset value as an extended account, determine other accounts that have a direct business relationship with the re-determined extended account as re-determined target accounts, and calculate the correlation degree between the re-determined extended account and the re-determined target account, thereby continuing to update the business graph.
[0094] In addition, accounts whose correlation degree with corresponding extended accounts is greater than a fourth preset value may be determined as extended accounts from among target accounts whose correlation degree is greater than a first preset value, and the business graph may be continuously updated.
[0095] When the update end condition is met, the final business graph is determined. It should be noted that after updating the business graph, the third degree of association between the node corresponding to the account to be identified and other nodes in the business graph should also be determined. When the third degree of association is less than the third preset value, the other nodes need to be cut off, and the most recently updated business graph is determined as the final business graph. When the business graph reaches the preset number of hops or there is no target account with a second degree of association greater than the second preset value, the most recently updated business graph can also be determined as the final business. Of course, the third preset value and the fourth preset value mentioned above can also be selected according to actual conditions, and are generally less than the first preset value and the second preset value.
[0096] After determining the final business graph, the server determines the characteristics of each node based on the attributes of each node in the final business graph, determines the characteristics of each edge based on the business data corresponding to each edge in the final business graph, and then fuses the characteristics of each node and the characteristics of the edge according to preset weights to determine the fused characteristics as the characteristics of the account to be identified.
[0097] Of course, it is also possible to determine the features of the edges and nodes directly connected to the node corresponding to the account to be identified, fuse the features of the edges and nodes directly connected to the node corresponding to the account to be identified with the features of the node corresponding to the account to be identified, determine the fused features, and then use the trained graph neural network to perform graph feature propagation based on the final business graph and the fused features to determine the features used to characterize the account to be identified.
[0098] S108: Perform risk identification on the account to be identified based on the characteristics.
[0099] After determining the characteristics of the account to be identified, the server may analyze the characteristics of the account to be identified to determine whether the account to be identified is a risk account, or may analyze the characteristics of the account to be identified by a preset risk identification model to determine whether the account to be identified is a risk account. This specification does not limit this.
[0100] based on Figure 1 In the risk identification method described, a business graph is constructed with the account to be identified and other accounts that perform business with the account to be identified as nodes and business data as edges. Then, based on the correlation between the account to be identified and the other accounts, the business graph is further updated and expanded. When the business graph reaches a preset number of hops, the final business graph is determined. Based on the final business graph, the characteristics of the account to be identified are confirmed, and risk identification of the account to be identified is performed based on the characteristics of the account to be identified.
[0101] It can be seen from the above system that the business graph that reaches the preset number of hops is determined by the correlation between the account to be identified and other accounts, and the characteristics of the account to be identified are determined according to the characteristics of the edges and nodes in the business graph. When risk identification is performed on the account to be identified, the node to be identified can be characterized by the characteristics and business data of accounts with a higher correlation with the account to be identified. This can better identify accounts that implement risky businesses through complex transaction networks, and at the same time, automatically identify the risks of the account to be identified, reduce manual participation, improve the efficiency of feature extraction and risk identification, and save computing resources.
[0102] In addition, when constructing the business graph in step S100, in order to integrate the sequence attribute of the business into the characteristics of the node to be identified, the server can also determine, for each other account, the various businesses executed between the other account and the account to be identified, and then, for each determined business, determine an edge between the node of the other account in the business graph and the node of the account to be identified based on the business data of the business, such as Figure 4 As shown, Figure 4 A schematic diagram of a business graph provided in this specification. a1 to a5 represent nodes of other accounts respectively, and the lines between the nodes represent the business data of each transaction between the nodes.
[0103] Of course, in step S104, when updating the business graph, in order to integrate the sequence attribute of the business into the characteristics of the node to be identified, the server can also determine, for each target account, the various businesses executed between the target account and the extended account, and for each determined business, determine an edge between the node of the target account and the node of the extended account in the business graph according to the business data of the business, such as Figure 5 As shown, Figure 5 A schematic diagram of a business graph provided in this specification. Each circle filled in black represents a node, and the line between the nodes is an edge, which is used to characterize the business between two accounts. The nodes filled in black represent nodes with a first correlation greater than a first preset value, and the node b1 is a node to be identified. The nodes filled in white represent nodes with a first correlation less than a first preset value. The dotted edges and dotted circles represent nodes with a first correlation less than a first preset value. Each edge represents the business data of the business between two nodes.
[0104] Furthermore, in order to better characterize the node to be identified, the server may also integrate the time characteristics of the service execution into the node to be identified. Specifically, Figure 6 As shown, Figure 6 A schematic diagram of a business graph update provided in this specification, wherein circles represent nodes in the business graph, edges with directions represent node relationships, that is, node a is the upstream node of node b, node b is the downstream node of node a, t0-t6 represent the timestamps of the direct business of the nodes, wherein t0 <t1<t2,t3> t0, t5 <t0,t4<t0,t6> t0.
[0105] For each extended account, determine the timestamps of each business initiated by the extended account to the account to be identified, and determine the largest timestamp among the timestamps of the businesses as the first timestamp, that is, Figure 6 At t0 in the figure, from the upstream accounts in the target account of the extended account, each upstream account whose first correlation with the extended account is greater than the first preset value is determined as the update account, namely, node c and node d. Then, the timestamp of each business initiated by each update account to the extended account is determined as the second timestamp of each update account, namely, t3 and t4. Then, it is determined that there is an update account whose second timestamp is less than the first timestamp, as a newly added node, and then the edge between the newly added node and the extended account is determined based on the business data whose second timestamp of the newly added node is less than the first timestamp. According to the newly added node and the edge between the newly added node and the extended account, the business graph is updated. Therefore, the updated business graph should be Figure 6For the graph represented by the solid-line circle and solid-line edges, both node e and node f are upstream nodes of node b, not meeting the above update conditions. Although the timing meets the requirements, node d, although being a downstream node of node b, has a timestamp earlier than time t0, so it does not meet the update conditions. That is, the dashed-line edges and dashed-line circles are the edges and nodes to be trimmed. Of course, if the extended node has no upstream nodes, there is no need to perform the above steps, and this extended node is the updated node.
[0106] Then, for each extended account, the server determines the timestamps of the various services initiated by the account to be identified to this extended account, and determines the minimum timestamp among the timestamps of the various services as the third timestamp. From the downstream accounts of the target account of this extended account, it determines the downstream accounts with the first association degree greater than the first preset value with this extended account as the updated accounts, determines the timestamps of the various services initiated by this extended account to the updated accounts as the fourth timestamps of the updated accounts. Then it determines the updated accounts with the fourth timestamp greater than the third timestamp as the new nodes, uses the service data with the fourth timestamp of the new nodes greater than the third timestamp to determine the edges between the new nodes and the extended account, and updates the service graph according to the new nodes and the edges between the new nodes and the extended account. Specifically, as Figure 7 shown, Figure 7 is a schematic diagram of service graph update provided in this specification. Among them, the circles represent the nodes in the service graph, and the edges with directions represent the node relationships. That is, node a is a downstream node of node m, and node m is an upstream node of node a. t0 - t6 represent the timestamps of the direct services of the nodes. Among them, t0 < t1 < t2, t3 > t2, t5 < t2, t4 < t2, t6 > t2. Since t0 < t1 < t2, it can be seen that the second target timestamp is t2. Among them, both node p and node q are downstream nodes of node m, not meeting the update conditions. The timestamp t4 of node o < t2, meeting the update conditions. The timestamp t3 of node n > t2, not meeting the update conditions. Therefore, the dashed-line edges and dashed-line circles are the edges and nodes to be trimmed, and the graph represented by the solid-line circle and solid-line edges is the updated service graph.
[0107] Of course, the above risk identification method can also be implemented through the method of fusing multiple graph neural networks, as Figure 8 shown, Figure 8A schematic diagram of a risk identification model provided in this specification. Among them, after determining the final business graph, for the node a1 to be identified, determine the nodes a2-a4 that have a direct business relationship with the node a1, and then extract the features of the edges directly connected to the node a1 in the business graph through the feature extraction layer, and extract the features of the node a1. Of course, in one or more embodiments of this specification, there is no restriction on the specific method to be adopted, which can be determined according to actual needs. For example, if the nodes in the final business graph are the nodes in the blockchain, when extracting the features of the edges, the feature extraction of each edge can be realized through the deep learning architecture of Transformer. In addition, the feature extraction of the edges can be realized through methods such as gated recurrent units (GRU), long short-term memory networks (LSTM), and then, a fusion network layer is used to realize the feature fusion of the node a1 to be identified and the edges directly connected, and then the graph feature propagation is realized according to the final business graph determined by the graph neural network, and the features of other nodes and edges in the final business graph are fused with the features of the node according to certain weights to determine the features used to characterize the node to be identified. Finally, through the risk identification layer, the risk of the node to be identified is identified according to the characteristics of the node to be identified and the identification result is output.
[0108] It should be noted that when extracting features from the edges of the final business graph, Fig. 9 As shown, Fig. 9 This is a schematic diagram of edge feature extraction provided in this specification, in which the business data of each edge is represented first. Taking the transaction in the blockchain as an example, the business data may include the resource flow, timestamp, account type and transaction type in the transaction, where E represents the node address, C represents the smart contract address, X represents resource transfer, and S represents resource exchange. Fig. 9 Taking the leftmost edge as an example, it can indicate that at time t0, 3 resource blocks flow from the node to be identified c1 to the node c2. After the feature extraction layer is thick, the feature h1 of the edge is extracted.
[0109] In one or more embodiments of this specification, the server can also implement risk identification through the following methods, such as Figure 1 As shown, Fig.10 A risk identification method provided in this manual, specifically:
[0110] S1000: Determine an account to be identified, and construct a business graph with the account to be identified and other accounts that perform business with the account to be identified as nodes and business execution data as edges.
[0111] S1002: For each other account, determine a second degree of association between the account to be identified and the other account based on business data of business executed between the other account and the account to be identified, and determine other accounts whose second degree of association is greater than a second preset value as extended accounts.
[0112] In order to further improve the precision and accuracy of risk identification and better identify risk accounts that implement risky businesses through complex business networks, it is necessary to better characterize the nodes to be identified. Therefore, it is necessary to further expand the business graph. However, in order to improve the efficiency of risk identification and save computing resources, when expanding the business graph, the server can update the business graph based on the correlation between accounts and other accounts with higher correlation.
[0113] Specifically, for each other account, the second degree of association between the other account and the account to be identified is determined. Of course, in one or more embodiments of this specification, there is no limitation on the specific method used to calculate the second degree of association between the other account and the account to be identified. It can be determined by the business data between the other account and the account to be identified, for example, by determining the number of business executions between the other account and the account to be identified, whether the active time period overlaps, or the business content, etc. to determine the second degree of association between the other account and the account to be identified. It is also possible to fuse the business data between the other account and the account to be identified according to a certain weight through a trained model, and score the fusion result through a trained scoring model to determine the second degree of association between the other account and the account to be identified. Since there are many methods that can be used, they are not listed here one by one.
[0114] Then, after determining the second degree of association between the other account and the account to be identified, determine whether the second degree of association is greater than a second preset value. If so, the other account whose second degree of association is greater than the second preset value can be used as an extended account for subsequent expansion of the business graph.
[0115] It should be noted that the setting of the second preset value can be determined according to actual needs. If the risk identification accuracy required is high, the first preset value can be smaller. If the risk identification accuracy required is low and high efficiency is required, the second preset value can be set larger. The most appropriate second preset value can be determined according to actual needs, and this manual does not impose any restrictions on this.
[0116] like Fig.11 As shown, Fig.11A schematic diagram of a business graph provided in this specification, wherein each circle represents a node, and the lines between the nodes are edges, which are used to characterize the business between two accounts, the black-filled nodes represent nodes whose first correlation is greater than a first preset value, node a is a node to be identified, and the white nodes other than node a represent nodes corresponding to extended accounts whose second correlation is less than a second preset value.
[0117] S1004: For each extended account, determine other accounts that have executed business with the extended account as target accounts, and determine a first degree of association between the target account and the extended account.
[0118] S1006: Determine the target accounts with a first correlation degree greater than a first preset value as nodes, and update the business graph with the business data of the executed businesses between the nodes as edges.
[0119] S1008: Re-select the target account with the first correlation degree greater than the first preset value as the extended account, continue to update the business graph until the update end condition is met, determine the final business graph, and determine the characteristics of the account to be identified based on the final business graph.
[0120] S1010: Perform risk identification on the account to be identified based on the characteristics.
[0121] The above is a risk identification method provided in the embodiments of this specification. Based on the same idea, this specification also provides a corresponding risk identification device, such as Fig.12 shown.
[0122] A construction module 1200 is used to determine an account to be identified, and to construct a business graph using the account to be identified and other accounts that perform business with the account to be identified as nodes and the executed business data as edges;
[0123] The updating module 1201 is used to take other accounts that have executed business with the account to be identified as extended accounts, and for each extended account, determine other accounts that have executed business with the extended account as target accounts, and determine a first association degree between the target account and the extended account; determine the target accounts with the first association degree greater than a first preset value as nodes, and update the business graph with business data of business executed between the nodes as edges;
[0124] A determination module 1202 is used to re-use the target account with the first correlation degree greater than the first preset value as an extended account, continue to update the business graph until the update end condition is met, determine a final business graph, and determine the characteristics of the account to be identified according to the final business graph;
[0125] The identification module 1203 is used to identify the risk of the account to be identified according to the characteristics.
[0126] Optionally, construction module 1200 is specifically used to determine, for each other account, each business performed between the other account and the account to be identified; for each determined business, determine, based on the business data of the business, an edge between the node of the other account and the node of the account to be identified in the business graph.
[0127] Optionally, construction module 1200 is specifically used to determine, for each target account, each business executed between the target account and the extended account; for each determined business, determine an edge between the node of the target account and the node of the extended account in the business graph according to the business data of the business.
[0128] Optionally, the update module 1201 is specifically used to determine, for each other account, a second degree of association between the account to be identified and the other account based on business data of business executed between the other account and the account to be identified; and determine other accounts whose second degree of association is greater than a second preset value as extended accounts.
[0129] Optionally, the business data includes a timestamp when the business is executed; when the account initiating the business is an upstream account and the account receiving the business is a downstream account, the update module 1201 is specifically used to determine, for each extended account, the timestamp of each business initiated by the extended account to the account to be identified, and determine the largest timestamp among the timestamps of each business as the first timestamp; from the upstream accounts in the target account of the extended account, determine each upstream account whose first correlation degree with the extended account is greater than a first preset value as the update account; determine the timestamp of each business initiated by each update account to the extended account as the second timestamp of each update account; determine the existence of an update account whose second timestamp is less than the first timestamp as a newly added node; determine the edge between the newly added node and the extended account based on the business data whose second timestamp of the newly added node is less than the first timestamp, and update the business graph according to the newly added node and the edge between the newly added node and the extended account.
[0130] Optionally, the business data includes a timestamp when the business is executed; when the account initiating the business is an upstream account and the account receiving the business is a downstream account, the update module 1201 is specifically used to determine, for each extended account, the timestamp of each business initiated by the account to be identified to the extended account, and determine the smallest timestamp among the timestamps of each business as a third timestamp; from the downstream accounts in the target account of the extended account, determine each downstream account whose first correlation degree with the extended account is greater than a first preset value as an update account; determine the timestamp of each business initiated by the extended account to each update account as a fourth timestamp of each update account; determine the existence of an update account whose fourth timestamp is greater than the third timestamp as a newly added node; determine the edge between the newly added node and the extended account based on the business data whose fourth timestamp of the newly added node is greater than the third timestamp, and update the business graph according to the newly added node and the edge between the newly added node and the extended account.
[0131] Optionally, when the business data includes at least one of a timestamp of executing the business, a business volume, and a business execution count, the update module 1201 is specifically used to determine a first degree of association between the target account and the extended account based on at least one of a timestamp of executing the business, a business volume, and a business execution count.
[0132] Optionally, the determination module 1202 is specifically used to determine the third degree of association between the node corresponding to the account to be identified and other nodes in the business graph; when at least one of the following conditions is met: the business graph reaches a preset number of hops, the third degree of association is less than a third preset value, or there is no target account with a second degree of association greater than a second preset value, the most recently updated business graph is determined as the final business graph.
[0133] Optionally, the determination module 1202 is specifically used to determine the characteristics of each node in the final business graph according to the attributes of each node; determine the characteristics of each edge in the final business graph according to the business data corresponding to the edge; fuse the characteristics of each node and the characteristics of the edge according to preset weights, and determine the fused characteristics as the characteristics of the account to be identified.
[0134] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0135] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0136] This specification also provides a computer-readable storage medium, which stores a computer program, which can be used to execute the above Figure 1 Provides risk identification methods.
[0137] This manual also provides Fig.13 The schematic structure diagram of the electronic device shown in FIG. Fig.13 As shown, at the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include other hardware required for the business. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to achieve the above Figure 1 The risk identification method described. Of course, in addition to the software implementation, this specification does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.
[0138] In the 1990s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the method flow). However, with the development of technology, many improvements to the method flow today can be regarded as direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using a hardware entity module. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask a chip manufacturer to design and produce a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.
[0139] The controller can be implemented in any appropriate manner, for example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (such as software or firmware) that can be executed by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art also know that in addition to implementing the controller in a purely computer-readable program code manner, the controller can be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, this controller can be considered as a hardware component, and the devices included therein for implementing various functions can also be regarded as structures within the hardware component. Or even, the devices for implementing various functions can be regarded as both software modules for implementing the method and structures within the hardware component.
[0140] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
[0141] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.
[0142] 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.
[0143] 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.
[0144] 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.
[0145] 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.
[0146] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.
[0147] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.
[0148] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.
[0149] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.
[0150] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, this specification may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, this specification 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.
[0151] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.
[0152] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.
[0153] The above description is only an embodiment of this specification and is not intended to limit this specification. For those skilled in the art, this specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this specification should be included in the scope of the claims of this application.
Claims
1. A risk identification method, comprising: Determine the account to be identified, and construct a business graph using other accounts that have executed business with the account to be identified and the account to be identified as nodes, and business data of business executed between nodes as edges; Other accounts that have performed business with the account to be identified are used as extended accounts, and for each extended account, other accounts that have performed business with the extended account are determined as target accounts, and a first degree of association between the target account and the extended account is determined; Determine a target account with a first association degree greater than a first preset value as a node, and update the business graph with business data of executing business between the nodes as an edge; Re-using the target account with the first correlation degree greater than the first preset value as the extended account, continuing to update the business graph until the update end condition is met, determining a final business graph, and determining the characteristics of the account to be identified according to the final business graph; Risk identification is performed on the account to be identified based on the characteristics.
2. The method according to claim 1, wherein the business data of the business executed between the nodes is used as the edge to construct a business graph, specifically comprising: For each other account, determine each business performed between the other account and the account to be identified; For each determined business, an edge between the node of the other account and the node of the account to be identified in the business graph is determined according to the business data of the business.
3. The method according to claim 1 or 2, wherein the business data of the business executed between the nodes is used as the edge, and the business graph is updated, specifically comprising: For each target account, determine each business executed between the target account and the extended account; For each determined business, an edge between the node of the target account and the node of the extended account in the business graph is determined according to the business data of the business.
4. The method according to claim 1, wherein other accounts that have performed business with the account to be identified are used as extended accounts, specifically comprising: For each other account, determining a second degree of association between the account to be identified and the other account according to business data of business executed between the other account and the account to be identified; Other accounts whose second correlation degree is greater than a second preset value are determined as extended accounts.
5. As claimed in claim 1 or the method, the business data includes a timestamp when the business is executed; the account that initiates the business is the upstream account, and the account that receives the business is the downstream account; Determining a target account with a first association degree greater than a first preset value as a node, and updating the business graph with business data of executing business between the nodes as an edge, specifically includes: For each extended account, determine the timestamp of each business initiated by the extended account to the account to be identified, and determine the largest timestamp among the timestamps of the businesses as the first timestamp; From the upstream accounts in the target account of the extended account, determine each upstream account with a first correlation degree of the extended account greater than a first preset value as an update account; Determine the timestamp of each business initiated by each updating account to the extended account as the second timestamp of each updating account; Determine that there is an updated account whose second timestamp is less than the first timestamp, as a newly added node; The edge between the newly added node and the extended account is determined based on the business data whose second timestamp is less than the first timestamp, and the business graph is updated based on the newly added node and the edge between the newly added node and the extended account.
6. The method according to claim 5, determining the target accounts with a first association degree greater than a first preset value as nodes, and updating the business graph with the business data of the business executed between the nodes as edges, specifically comprising: For each extended account, determine the timestamps of each service initiated by the account to be identified to the extended account, and determine the smallest timestamp among the timestamps of each service as the third timestamp; From the downstream accounts in the target account of the extended account, determine each downstream account with a first correlation degree of the extended account greater than a first preset value as an update account; Determine the timestamp of each service initiated by the extended account to each update account as the fourth timestamp of each update account; Determine that there is an updated account whose fourth timestamp is greater than the third timestamp, as a newly added node; The edge between the newly added node and the extended account is determined based on the business data whose fourth timestamp of the newly added node is greater than the third timestamp, and the business graph is updated based on the newly added node and the edge between the newly added node and the extended account.
7. The method according to claim 1, wherein the service data comprises at least one of a timestamp of executing the service, a service volume, and a service execution frequency; Determining a first degree of association between the target account and the extended account specifically includes: The first association degree between the target account and the extended account is determined according to at least one of a timestamp, a business volume, and a business execution count of the business executed between the target account and the extended account.
8. The method according to claim 1, until the update end condition is met, determining the final business graph, specifically comprising: Determining a third degree of association between the node corresponding to the account to be identified and other nodes in the business graph; When at least one of the following conditions is met: the business graph reaches a preset number of hops, the third degree of association is less than a third preset value, or there is no target account with a second degree of association greater than a second preset value, the most recently updated business graph is determined as the final business graph.
9. The method of claim 1, wherein determining the characteristics of the account to be identified based on the final business graph comprises: Determine the characteristics of each node according to the attributes of each node in the final business graph; Determining the characteristics of each edge according to the business data corresponding to each edge in the final business graph; The features of the nodes and the features of the edges are fused according to preset weights to determine fused features as the features of the account to be identified.
10. A risk identification device, comprising: A construction module, used to determine an account to be identified, and construct a business graph with the account to be identified and other accounts that perform business with the account to be identified as nodes and the executed business data as edges; an updating module, configured to use other accounts that have executed business with the account to be identified as extended accounts, and for each extended account, determine other accounts that have executed business with the extended account as target accounts, and determine a first degree of association between the target account and the extended account; Determine a target account with a first association degree greater than a first preset value as a node, and update the business graph with business data of executing business between the nodes as an edge; a determination module, configured to re-use the target account with the first correlation degree greater than the first preset value as an extended account, continue to update the business graph until an update end condition is met, determine a final business graph, and determine the characteristics of the account to be identified according to the final business graph; An identification module is used to identify the risk of the account to be identified based on the characteristics.
11. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method according to any one of claims 1 to 9.
12. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 9 when executing the program.
