Risk analysis method and device, computer equipment and storage medium

Abstract

The invention relates to a risk analysis method and device, computer equipment, a storage medium and a computer program product, and relates to the technical field of big data. The method comprises the following steps: obtaining a knowledge graph representing a resource circulation relationship, the knowledge graph comprising respective resource node information of a plurality of resource nodes and resource circulation information among the resource nodes, and screening the resource nodes according to a node screening condition matched with the scale of the knowledge graph, screening a plurality of basic nodes of which the resource circulation information meets a node screening condition from the resource nodes, determining at least one basic link according to the position of each basic node in the knowledge graph, performing similarity matching on the resource node information of each basic node in the basic link, and determining the resource node information of each basic node in the basic link; by adopting the method, the basic links to which the risk nodes belong are determined as the risk links, so that the accurate risk links are obtained, the information analysis amount is greatly reduced, and the determination efficiency of the risk links is improved.

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to a risk analysis method, apparatus, computer equipment, storage medium and computer program product. Background Art

[0002] With the development of big data technology, traditional offline resource circulation business is constantly migrating online, and the amount of data generated by the resource circulation business involved in online business is increasing. Therefore, it is necessary to analyze the resource circulation situation.

[0003] In the traditional method, the resource record table of each resource node is analyzed. For example, a relational database is used to perform table association operations to obtain the analysis results of the resource circulation situation. This method involves a full table scan, which has a large amount of calculation and a long analysis time, resulting in low analysis efficiency. Summary of the invention

[0004] Based on this, it is necessary to provide a risk analysis method, apparatus, computer equipment, computer-readable storage medium and computer program product that can improve analysis efficiency in response to the above technical problems.

[0005] In a first aspect, the present application provides a risk analysis method. The method comprises:

[0006] A knowledge graph representing resource circulation relationships is obtained; the knowledge graph includes resource node information of each of a plurality of resource nodes, and resource circulation information between each of the resource nodes; according to a node screening condition matching the scale of the knowledge graph, a plurality of basic nodes whose resource circulation information satisfies the node screening condition are screened in each of the resource nodes; at least one basic link is determined according to the position of each of the basic nodes in the knowledge graph; the basic link is a resource circulation link between at least a part of the basic nodes; similarity matching is performed on the resource node information of each of the basic nodes in the basic link to obtain the node similarity between the basic nodes; in the case where there is a risk node whose node similarity satisfies the similarity condition between the resource nodes, the basic link to which the risk node belongs is determined as the risk link.

[0007] In one embodiment, resource circulation information includes resource circulation times and resource circulation amounts; the node screening conditions include circulation times screening conditions and circulation amount screening conditions; the screening of multiple basic nodes whose resource circulation information satisfies the node screening conditions in each resource node according to the node screening conditions that match the scale of the knowledge graph comprises: determining circulation times screening conditions and circulation amount screening conditions that match the scale of the knowledge graph; screening of multiple candidate nodes whose resource circulation times satisfy the circulation times screening conditions in each resource node according to the circulation times screening conditions; screening of multiple basic nodes whose resource circulation amounts satisfy the circulation amount screening conditions in each candidate node according to the circulation amount screening conditions.

[0008] In one of the embodiments, the amount of resource circulation between two resource nodes is the sum of the resource transfer amounts corresponding to each resource circulation between the two resource nodes; the multiple basic nodes whose resource circulation amounts meet the circulation amount screening conditions are screened in each of the candidate nodes according to the circulation amount screening conditions, including: determining a circulation amount threshold and a circulation amount interval representing the circulation amount screening conditions; for each candidate node, if there is a resource transfer amount greater than or equal to the circulation amount threshold, or if there is a resource transfer amount that belongs to the circulation amount interval, the candidate node is determined as a basic node.

[0009] In one of the embodiments, the candidate node corresponds to multiple resource circulation times; and according to the circulation amount screening condition, multiple basic nodes whose resource circulation amounts meet the circulation amount screening condition are screened in each candidate node, including: for the candidate node, determining the resource flow direction of each resource circulation; for each resource flow direction, according to the resource circulation amounts of each resource circulation corresponding to the resource flow direction, counting the total resource transfer amount corresponding to the resource flow direction; in each resource flow direction, if there is a target flow whose total resource transfer amount meets the circulation amount screening condition, the candidate node is determined as the basic node.

[0010] In one of the embodiments, similarity matching is performed on the resource node information of each basic node in the basic link to obtain the node similarity between each basic node, including: dividing each basic node in the basic link into node pairs to obtain at least one node pair; the node pair includes two adjacent nodes that are in adjacent node positions in the knowledge graph; for each of the node pairs, similarity matching is performed on the resource node information of each adjacent node in the node pair to obtain the node similarity corresponding to the node pair.

[0011] In one of the embodiments, the method further includes: obtaining resource record information of each of the plurality of resource nodes and a resource extraction template configured for the resource record information; for each of the resource nodes, extracting information from the resource record information according to the resource extraction template to obtain resource node information of the resource node and resource flow information between the resource node and other resource nodes; and constructing a knowledge graph representing the resource flow relationship between each of the resource nodes based on the resource node information of each of the resource nodes and the resource flow information between each of the resource nodes.

[0012] In a second aspect, the present application also provides a risk analysis device. The device comprises:

[0013] A graph acquisition module is used to acquire a knowledge graph representing resource circulation relationships; the knowledge graph includes resource node information of each of a plurality of resource nodes, and resource circulation information between each of the resource nodes; a node screening module is used to screen, in accordance with a node screening condition matching the scale of the knowledge graph, a plurality of basic nodes whose resource circulation information satisfies the node screening condition in each of the resource nodes; a link determination module is used to determine at least one basic link according to the position of each of the basic nodes in the knowledge graph; the basic link is a resource circulation link between at least a part of the basic nodes; an information matching module is used to perform similarity matching on the resource node information of each of the basic nodes in the basic link to obtain the node similarity between each of the basic nodes; a risk determination module is used to determine the basic link to which the risk node belongs as a risk link when there is a risk node whose node similarity satisfies the similarity condition between the resource nodes.

[0014] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0015] A knowledge graph representing resource circulation relationships is obtained; the knowledge graph includes resource node information of each of a plurality of resource nodes, and resource circulation information between each of the resource nodes; according to a node screening condition matching the scale of the knowledge graph, a plurality of basic nodes whose resource circulation information satisfies the node screening condition are screened in each of the resource nodes; at least one basic link is determined according to the position of each of the basic nodes in the knowledge graph; the basic link is a resource circulation link between at least a part of the basic nodes; similarity matching is performed on the resource node information of each of the basic nodes in the basic link to obtain the node similarity between the basic nodes; in the case where there is a risk node whose node similarity satisfies the similarity condition between the resource nodes, the basic link to which the risk node belongs is determined as the risk link.

[0016] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0017] A knowledge graph representing resource circulation relationships is obtained; the knowledge graph includes resource node information of each of a plurality of resource nodes, and resource circulation information between each of the resource nodes; according to a node screening condition matching the scale of the knowledge graph, a plurality of basic nodes whose resource circulation information satisfies the node screening condition are screened in each of the resource nodes; at least one basic link is determined according to the position of each of the basic nodes in the knowledge graph; the basic link is a resource circulation link between at least a part of the basic nodes; similarity matching is performed on the resource node information of each of the basic nodes in the basic link to obtain the node similarity between the basic nodes; in the case where there is a risk node whose node similarity satisfies the similarity condition between the resource nodes, the basic link to which the risk node belongs is determined as the risk link.

[0018] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0019] A knowledge graph representing resource circulation relationships is obtained; the knowledge graph includes resource node information of each of a plurality of resource nodes, and resource circulation information between each of the resource nodes; according to a node screening condition matching the scale of the knowledge graph, a plurality of basic nodes whose resource circulation information satisfies the node screening condition are screened in each of the resource nodes; at least one basic link is determined according to the position of each of the basic nodes in the knowledge graph; the basic link is a resource circulation link between at least a part of the basic nodes; similarity matching is performed on the resource node information of each of the basic nodes in the basic link to obtain the node similarity between the basic nodes; in the case where there is a risk node whose node similarity satisfies the similarity condition between the resource nodes, the basic link to which the risk node belongs is determined as the risk link.

[0020] The above-mentioned risk analysis method, device, computer equipment, storage medium and computer program product obtain a knowledge graph representing resource circulation relations, wherein the knowledge graph contains resource node information of each of a plurality of resource nodes and resource circulation information between each of the resource nodes, and selects a plurality of basic nodes whose resource circulation information satisfies the node screening condition in each of the resource nodes according to the node screening condition matching the scale of the knowledge graph, thereby reducing the number of resource nodes and the amount of subsequent information analysis. According to the position of each of the basic nodes in the knowledge graph, at least one basic link is determined, wherein the basic link is a resource circulation link between at least a part of the basic nodes, and the basic nodes are further analyzed based on the basic link, and the correlation between the basic nodes is taken into consideration, thereby improving the accuracy of subsequent determination of the risk link. The resource node information of each of the basic nodes in the basic link is matched for similarity, and the node similarity between the basic nodes is obtained. By using the above-mentioned method, when there is a risk node whose node similarity satisfies the similarity condition between the resource nodes, the basic link to which the risk node belongs is determined as the risk link, thereby greatly reducing the amount of information analysis and improving the efficiency of determining the risk link under the premise of ensuring that an accurate risk link is obtained. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] FIG1 is a diagram of an application environment of a risk analysis method in one embodiment;

[0022] FIG2 is a schematic diagram of a risk analysis method in one embodiment;

[0023] FIG3 is a schematic diagram of a flow chart of a method for determining and screening resource nodes in one embodiment;

[0024] FIG4 is a schematic diagram of a flow chart of a method for monitoring resource backflow in one embodiment;

[0025] FIG5 is a structural block diagram of a risk analysis device in one embodiment;

[0026] FIG. 6 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. It should be noted that in the embodiments of the present application, certain software, groups, models and other existing solutions in the industry may be mentioned, which 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, but it does not mean that the applicant has or will necessarily use the solution.

[0028] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations. The acquisition, storage, use, and processing of data in the technical solution of this application are in compliance with relevant provisions of national laws and regulations.

[0029] The risk analysis method provided in the embodiment of the present application can be applied in the application environment shown in FIG1 . In which, the terminal 102 communicates with the server 104 through a network. The data storage system can store data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers.

[0030] The server 104 obtains a knowledge graph representing resource circulation relations, and the knowledge graph includes resource node information of each of a plurality of resource nodes, and resource circulation information between each resource node. The server 104 selects a plurality of basic nodes whose resource circulation information satisfies the node screening conditions in each resource node according to the node screening conditions that match the scale of the knowledge graph. The server 104 determines at least one basic link according to the position of each basic node in the knowledge graph, and the basic link is a resource circulation link between at least a part of each basic node. The server 104 performs similarity matching on the resource node information of each basic node in the basic link, and obtains the node similarity between each basic node. When there is a risk node whose node similarity satisfies the similarity condition between each resource node, the server 104 determines the basic link to which the risk node belongs as a risk link, and the server 104 sends the risk link to the terminal 102.

[0031] The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, IoT devices, and portable wearable devices. The IoT devices may be smart speakers, smart TVs, smart air conditioners, smart car-mounted devices, etc. The portable wearable devices may be smart watches, smart bracelets, head-mounted devices, etc. The server 104 may be implemented as an independent server or a server cluster consisting of multiple servers.

[0032] In one embodiment, as shown in FIG. 2 , a risk analysis method is provided, which is described by taking the method applied to the server 104 in FIG. 1 as an example, including:

[0033] S202, obtaining a knowledge graph representing resource circulation relations, wherein the knowledge graph includes resource node information of each of a plurality of resource nodes, and resource circulation information between the resource nodes.

[0034] The resource circulation relationship refers to the flow of resources between resource entities. Resource entities refer to institutions or departments with resource management capabilities, such as banks, regulatory agencies, individuals, etc.

[0035] A knowledge graph is a structured semantic knowledge base that describes concepts and physical relationships in the physical world. It stores and expresses information in the form of triples such as entities, relationships, and attributes, enabling computers to understand and reason about complex situations in the real world.

[0036] Exemplarily, the knowledge graph representing resource circulation relationships can store and express information through triples such as resource nodes as entities, resource node attributes as attributes, and resource circulation information as relationships.

[0037] Resource nodes can represent the nodes in the knowledge graph corresponding to institutions or departments with resource management capabilities.

[0038] Resource node information refers to the attribute information of the resource node itself, for example, the identification information of the resource subject. The identification information includes the name of the resource subject, the number of the resource subject, etc.

[0039] Resource circulation information may be information used to characterize the resource circulation relationship between any two resource nodes. For example, if there is a resource transfer between a first resource node and a second resource node, the resource circulation information may be the amount of resource transfer, the number of resource transfers, etc. More specifically, the resource circulation information may be a transfer of X yuan from company A to company B.

[0040] S204, according to the node screening conditions that match the scale of the knowledge graph, multiple basic nodes whose resource circulation information meets the node screening conditions are screened in each resource node.

[0041] The scale of the knowledge graph includes the number of entities, the number of relationships, the number of attributes, and the depth of knowledge in the knowledge graph. The larger the number of entities, the number of relationships, and the number of attributes, the larger the scale of the knowledge graph. The deeper the knowledge depth, the larger the scale of the knowledge graph.

[0042] The node screening condition indicates the condition for screening each resource node. Specifically, the node screening condition may be screening each resource node according to the number of node circulations, or screening each resource node according to the amount of node circulations, or screening each resource node according to the number of node circulations and the amount of node circulations.

[0043] The basic nodes are resource nodes that meet the node screening conditions.

[0044] When the scale of the knowledge graph is large, the number of node screening conditions that match the scale of the knowledge graph is small, and the time spent on screening each resource node according to the small number of node screening conditions is short. Conversely, when the scale of the knowledge graph is small, the number of node screening conditions that match the scale of the knowledge graph is large, and the basic nodes obtained by screening each resource node according to the large number of node screening conditions are more accurate.

[0045] Specifically, the server pre-builds a knowledge graph that represents resource circulation relationships and determines the scale of the knowledge graph.

[0046] In one of the embodiments, the server determines the scale of the knowledge graph based on the number of entities, the number of relationships, the number of attributes, and the knowledge depth in the knowledge graph.

[0047] S206, determining at least one basic link according to the location of each basic node in the knowledge graph. The basic link is a resource circulation link between at least a part of the basic nodes.

[0048] Among them, the position of the basic node in the knowledge graph represents the position of the entity corresponding to the basic node in the knowledge graph.

[0049] The basic link is a resource circulation link between at least a part of the basic nodes. The basic link can be a closed link or a non-closed link.

[0050] The resource circulation link refers to a link composed of various basic nodes and having a certain resource circulation direction. Specifically, the basic nodes in the resource circulation link are connected according to the resource circulation direction.

[0051] Specifically, the server constructs and processes each basic node based on the resource circulation relationship between the basic nodes to obtain a resource circulation link constructed by multiple basic nodes. The server then divides the resource circulation link according to the position of each basic node in the knowledge graph to obtain at least one basic link.

[0052] S208 , performing similarity matching on the resource node information of each basic node in the basic link to obtain the node similarity between the basic nodes.

[0053] The similarity matching refers to performing similarity matching on information of multiple resource nodes.

[0054] The node similarity represents the similarity between the resource node information of each basic node. The higher the similarity, the more likely the basic nodes are the same node. Furthermore, the basic link described by the basic node is more likely to be a risk link.

[0055] For example, if the basic node is a company, if there are two companies with high node similarity in the basic link, the two companies may be the same company, or branches of each other. That is, the basic link may be a risky link with resource reflux risk.

[0056] Specifically, the server performs similarity matching on the resource node information of each basic node in the basic link to obtain the node similarity between each basic node and other basic nodes.

[0057] In one of the embodiments, each basic node in the basic link is divided into node pairs to obtain at least one node pair; the node pair includes two adjacent nodes that are in adjacent node positions in the knowledge graph; for each node pair, similarity matching is performed on the resource node information of each adjacent node in the node pair to obtain the node similarity corresponding to the node pair.

[0058] S210: When there is a risk node whose node similarity satisfies a similarity condition among resource nodes, a basic link to which the risk node belongs is determined as a risk link.

[0059] The risk node may be a resource node whose node similarity satisfies a similarity condition. The similarity condition indicates a similarity threshold configured for the node similarity. The risk link may be a basic link including the risk node.

[0060] Specifically, when there is a risk node whose node similarity satisfies a similarity condition among resource nodes, the server determines the basic link to which the risk node belongs as a risk link.

[0061] In one embodiment, the server obtains a similarity threshold, and when there is a risk node between each resource node, determines the basic link to which the risk node belongs as the risk link. The risk node is a basic node (resource node) whose node similarity is greater than the similarity threshold.

[0062] In the above risk analysis method, a knowledge graph representing resource circulation relations is obtained. The knowledge graph contains resource node information of each of multiple resource nodes and resource circulation information between each resource node. According to the node screening conditions that match the scale of the knowledge graph, multiple basic nodes whose resource circulation information meets the node screening conditions are screened in each resource node, thereby reducing the number of resource nodes and the amount of subsequent information analysis. According to the position of each basic node in the knowledge graph, at least one basic link is determined. The basic link is a resource circulation link between at least a part of each basic node. The basic nodes are further analyzed based on the basic link. Considering the correlation between the basic nodes, the accuracy of subsequent risk link determination is improved. The resource node information of each basic node in the basic link is matched with similarity to obtain the node similarity between each basic node. Using the above method, when there is a risk node whose node similarity meets the similarity conditions between each resource node, the basic link to which the risk node belongs is determined as a risk link. Under the premise of ensuring that an accurate risk link is obtained, the amount of information analysis is greatly reduced and the efficiency of determining the risk link is improved.

[0063] In one embodiment, as shown in FIG. 3, a flow chart of a method for determining and screening resource nodes, resource circulation information includes resource circulation times and resource circulation amounts, and node screening conditions include circulation times screening conditions and circulation amount screening conditions. According to the node screening conditions that match the scale of the knowledge graph, multiple basic nodes whose resource circulation information meets the node screening conditions are screened in each resource node, including:

[0064] S302, determine the circulation frequency screening conditions and circulation amount screening conditions that match the scale of the knowledge graph.

[0065] The resource flow times represent the number of resource flows into and out of a resource node. Specifically, the resource flow times are the sum of the number of resource flows into the resource node and the number of resource flows out of the resource node.

[0066] The resource circulation amount indicates the amount of resources flowing into and out of a resource node. Specifically, the resource circulation times is the sum of the amount of resources flowing into the resource node and the amount of resources flowing out of the resource node.

[0067] S304, according to the circulation number screening condition, screening multiple candidate nodes whose resource circulation times meet the circulation number screening condition from among the resource nodes.

[0068] The candidate node represents a resource node whose resource circulation times meet the circulation times screening condition.

[0069] The circulation number screening condition represents the condition configured for the number of resource inflows and outflows. Exemplarily, the circulation number screening condition may be that the total number of resource inflows and outflows is less than a preset value, or the number of resource inflows is less than a preset value, or the number of resource outflows is less than a preset value.

[0070] Specifically, the server determines the circulation times screening condition and the circulation amount screening condition that match the scale according to the scale of the knowledge graph. For example, the larger the scale, the smaller the preset value corresponding to the circulation times screening condition, and the smaller the scale, the larger the preset value corresponding to the circulation times screening condition. Similarly, the larger the scale, the larger the preset value corresponding to the circulation amount screening condition, and the smaller the scale, the smaller the preset value corresponding to the circulation amount screening condition.

[0071] S306, according to the circulation amount screening condition, screening multiple basic nodes whose resource circulation amounts meet the circulation amount screening condition from among the candidate nodes.

[0072] Among them, the basic node represents a candidate node whose resource circulation amount meets the circulation amount screening condition.

[0073] The circulation amount screening condition represents the condition configured for the amount of resource inflow and outflow. For example, the circulation amount screening condition may be that the total amount of resource inflow and outflow is greater than a preset value, or the total amount of resource inflow is greater than a preset value, or the total amount of resource outflow is greater than a preset value.

[0074] Specifically, the server may determine a circulation frequency screening condition and a circulation amount screening condition that match the scale of the knowledge graph. The server may screen multiple candidate nodes whose resource circulation frequency satisfies the circulation frequency screening condition in each resource node according to the circulation frequency screening condition. The server may screen multiple basic nodes whose resource circulation amount satisfies the circulation amount screening condition in each candidate node according to the circulation amount screening condition.

[0075] In one of the embodiments, the server selects, according to the circulation number screening condition, a plurality of basic nodes whose resource circulation times satisfy the circulation number screening condition from among the resource nodes.

[0076] In one of the embodiments, the server selects multiple basic nodes whose resource circulation amounts satisfy the circulation amount screening conditions from among the resource nodes according to the circulation amount screening conditions.

[0077] In this embodiment, the circulation number screening condition and the circulation amount screening condition that match the scale of the knowledge graph are determined. According to the circulation number screening condition, multiple candidate nodes whose resource circulation number meets the circulation number screening condition are screened in each resource node. According to the circulation amount screening condition, multiple basic nodes whose resource circulation amount meets the circulation amount screening condition are screened in each candidate node, thereby reducing the number of resource nodes, reducing the amount of subsequent information analysis, and improving the efficiency of determining risk links.

[0078] In one embodiment, the resource circulation amount between two resource nodes is the sum of the resource transfer amounts corresponding to each resource circulation between the two resource nodes. According to the circulation amount screening condition, multiple basic nodes whose resource circulation amounts meet the circulation amount screening condition are screened in each candidate node, including: determining a circulation amount threshold and a circulation amount interval representing the circulation amount screening condition. For each candidate node, if there is a resource transfer amount greater than or equal to the circulation amount threshold, or if there is a resource transfer amount that falls within the circulation amount interval, the candidate node is determined as a basic node.

[0079] The amount of resource circulation between two resource nodes is the sum of the corresponding resource transfer amounts for each resource circulation between the two resource nodes.

[0080] The resource transfer amount indicates the amount generated each time a resource is circulated between two resource nodes. Specifically, if resource node A transfers X yuan to resource node B first and then transfers Y yuan, the resource transfer amount is X yuan or Y yuan, and the resource circulation amount is X+Y yuan.

[0081] The circulation amount threshold may be a threshold configured for the resource transfer amount. The circulation amount interval may be an interval configured for the resource transfer amount.

[0082] For example, the circulation amount threshold can represent a transfer amount of 50,000 yuan, and the circulation amount interval can represent an interval corresponding to 0.5 to 1.5 times the previous transfer amount. Candidate node A transfers 50,000 yuan to candidate node B, and candidate node B transfers 60,000 yuan to candidate node C. For candidate node B, B transfers 60,000 yuan to C (resource transfer amount) greater than or equal to 50,000 yuan (circulation amount threshold). B's transfer of 60,000 yuan (resource transfer amount) belongs to the circulation amount interval (candidate node A transfers 50,000 yuan to candidate node B × 0.5 ~ candidate node A transfers 50,000 yuan to candidate node B × 1.5).

[0083] Specifically, the server determines the circulation amount threshold and circulation amount range that represent the circulation amount screening conditions. For each candidate node, if the resource transfer amount is greater than or equal to the circulation amount threshold, or if the resource transfer amount falls within the circulation amount range, the candidate node is determined as the basic node.

[0084] In one of the embodiments, after the server determines the circulation amount threshold and the circulation amount interval representing the circulation amount screening condition, if there is a resource transfer amount greater than or equal to the circulation amount threshold, the candidate node is determined as the basic node.

[0085] In one of the embodiments, after the server determines the circulation amount threshold and the circulation amount interval representing the circulation amount screening condition, if there is a resource transfer amount that falls within the circulation amount interval, the candidate node is determined as the basic node.

[0086] In this embodiment, the circulation amount threshold and circulation amount interval representing the circulation amount screening condition are determined. For each candidate node, if there is a resource transfer amount greater than or equal to the circulation amount threshold, or if there is a resource transfer amount that falls within the circulation amount interval, the candidate node is determined as the basic node. Reducing the number of resource nodes and the amount of subsequent information analysis reduces the efficiency of determining risk links.

[0087] In one embodiment, the candidate node corresponds to multiple resource circulation times, and according to the circulation amount screening condition, multiple basic nodes whose resource circulation amounts meet the circulation amount screening condition are screened in each candidate node, including: for the candidate node, determining the resource flow direction of each resource circulation, for each resource flow direction, according to the resource circulation amounts of each resource circulation corresponding to the resource flow direction, counting the total resource transfer amount corresponding to the resource flow direction, and in each resource flow direction, when there is a target flow direction whose total resource transfer amount meets the circulation amount screening condition, determining the candidate node as the basic node.

[0088] The resource flow direction indicates the flow direction of resources in the candidate node, which includes the resource inflow direction and the resource outflow direction.

[0089] The resource circulation amount indicates the total amount of resource circulation between the candidate node and a resource node. For example, multiple transfers are initiated between node A and node B, and the total transfer amount is X yuan. The resource circulation amount is X yuan.

[0090] The total amount of resource transfer indicates the total amount of resource flows between the candidate and multiple other resource nodes. For example, node A transfers X yuan to node B and Y yuan to node C. The total amount of resource transfer is X+Y yuan.

[0091] The circulation amount screening condition may also be a screening condition configured for the total amount of resource transfer. Specifically, the circulation amount screening condition may indicate that the total amount of resource transfer is greater than or equal to Z yuan. If the total amount of resource transfer is greater than or equal to Z yuan, the candidate node corresponding to the total amount of resource transfer is determined as the basic node.

[0092] Specifically, the server determines the resource flow direction of each resource flow for the candidate node, and for each resource flow direction, calculates the total resource transfer amount corresponding to the resource flow according to the resource flow amount of each resource flow corresponding to the resource flow direction. If there is a target flow in each resource flow direction whose total resource transfer amount meets the flow amount screening condition, the candidate node is determined as the basic node.

[0093] In one of the embodiments, the server obtains resource record information corresponding to the candidate node, and the server determines the resource flow direction recorded in the resource record information.

[0094] In this embodiment, for the candidate nodes, the resource flow directions of each resource circulation are determined. For each resource flow direction, the total resource transfer amount corresponding to the resource flow direction is calculated according to the resource circulation amount of each resource circulation corresponding to the resource flow direction. Among each resource flow direction, if there is a target flow whose total resource transfer amount meets the circulation amount screening condition, the candidate node is determined as the basic node. Taking the resource flow direction into consideration is conducive to maintaining the orderliness of subsequent basic links, while reducing the number of resource nodes, reducing the amount of analysis of subsequent information, and improving the efficiency of determining risk links.

[0095] In one embodiment, similarity matching is performed on the resource node information of each basic node in the basic link to obtain the node similarity between each basic node, including: dividing each basic node in the basic link into node pairs to obtain at least one node pair; the node pair includes two adjacent nodes that are in adjacent node positions in the knowledge graph, and for each node pair, similarity matching is performed on the resource node information of each adjacent node in the node pair to obtain the node similarity corresponding to the node pair.

[0096] The node pair represents a set constructed by two basic nodes. Specifically, the node pair includes two adjacent nodes that are in adjacent node positions in the knowledge graph.

[0097] Specifically, the server can divide each basic node in the basic link into node pairs in sequence to obtain multiple node pairs, and each node pair is complementary and overlapping. For each node pair, the resource node information of each adjacent node in the node pair is similarly matched to obtain the node similarity corresponding to the node pair.

[0098] In one of the embodiments, the server may randomly divide each basic node in the basic link into node pairs to obtain multiple node pairs, and for each node pair, perform similarity matching on the resource node information of each adjacent node in the node pair to obtain the node similarity corresponding to the node pair.

[0099] In this embodiment, each basic node in the basic link is divided into node pairs to obtain at least one node pair; the node pair includes two adjacent nodes that are in adjacent node positions in the knowledge graph, and for each node pair, similarity matching is performed on the resource node information of each adjacent node in the node pair to obtain the node similarity corresponding to the node pair. Under the premise of ensuring that an accurate risk link is obtained, the amount of information analysis is greatly reduced and the efficiency of determining the risk link is improved.

[0100] In one embodiment, the risk analysis method also includes: obtaining resource record information of each of the multiple resource nodes and a resource extraction template configured for the resource record information; for each resource node, extracting information from the resource record information according to the resource extraction template to obtain the resource node information of the resource node and the resource flow information between the resource node and other resource nodes; and constructing a knowledge graph representing the resource flow relationship between the resource nodes based on the resource node information of each resource node and the resource flow information between the resource nodes.

[0101] The resource record information refers to the information recorded by the resource node. The resource record information includes the record information of the resource node in the resource flow dimension. Exemplarily, the resource record information includes transfer information, collection information, loan information, transaction time information, transaction amount information, etc.

[0102] Resource extraction template is used to extract resource record information. Resource extraction template includes resource attribute template and resource circulation template. Resource attribute template is used to extract information of attribute dimension, and resource circulation template is used to extract information of circulation dimension.

[0103] Each resource node can be used as an entity in the knowledge graph, the resource node information of each resource node can be used as the attribute of the entity in the knowledge graph, and the resource flow information between resource nodes can be used as the relationship between entities in the knowledge graph.

[0104] Specifically, the server constructs a knowledge graph that represents the resource circulation relationship between resource nodes, using each resource node as an entity in the knowledge graph, the resource node information of each resource node as the attribute of the entity in the knowledge graph, and the resource circulation information between resource nodes as the relationship between entities in the knowledge graph.

[0105] Specifically, the resource extraction template includes a resource attribute template and a resource circulation template. The server extracts information from the resource record information according to the resource attribute template to obtain the resource node information of each resource node. The server extracts information from the resource record information and the resource record information corresponding to other nodes according to the resource circulation template to obtain the resource circulation information between the resource node and other resource nodes.

[0106] In this embodiment, for each resource node, according to the resource extraction template, the resource record information is extracted to obtain the resource node information of the resource node and the resource circulation information between the resource node and other resource nodes. According to the resource node information of each resource node and the resource circulation information between each resource node, a knowledge graph representing the resource circulation relationship between each resource node is constructed. Under the premise of ensuring that an accurate risk link is obtained, the amount of information analysis is greatly reduced and the efficiency of determining the risk link is improved.

[0107] In one embodiment, the method for monitoring resource backflow as shown in FIG4 includes:

[0108] S402, obtaining resource record information of each of a plurality of resource nodes and a resource extraction template configured for the resource record information.

[0109] Exemplarily, the resource record information for a resource node includes: basic information on all corporate debts, basic information on corporate customers, debt information, expenditure information, payment information, loan contract customer relationship information, loan contracts, corporate current deposit contract account details, loan contract expenditure information, basic corporate current deposit contract account information, discount business details and other information.

[0110] S404, for each resource node, extract resource record information according to the resource extraction template to obtain resource node information of the resource node and resource flow information between the resource node and other resource nodes.

[0111] S406, constructing a knowledge graph representing the resource circulation relationship between the resource nodes based on the resource node information of each resource node and the resource circulation information between the resource nodes.

[0112] Exemplarily, the process of constructing a knowledge graph includes: Step 1, extracting and identifying the account entity from the basic information data of the corporate demand deposit contract account. Step 2, based on the identified entity, collect the attribute information of the entity, such as account number, customer number, customer name, etc., and extract it to form the descriptive attribute information of the entity. Step 3, based on the existing identified entities, extract the association relationship between the entities. The extraction of relationships is mainly aimed at direct relationships, including loan relationships and transfer relationships. This type of relationship can be obtained from an intuitive analysis of the existing data source. For example, if account A and account B have transfer records, then the relationship between accounts AB is a transfer relationship. This type of relationship extraction method is to directly obtain the relationship.

[0113] S408, obtaining a knowledge graph representing resource circulation relations.

[0114] Among them, the knowledge graph contains the resource node information of multiple resource nodes and the resource flow information between the resource nodes.

[0115] S410, determining the circulation frequency screening condition and the circulation amount screening condition that match the scale of the knowledge graph.

[0116] The resource circulation information includes the number of resource circulations and the amount of resource circulations, and the node screening conditions include the number of circulations screening conditions and the amount of circulations screening conditions.

[0117] S412, determining a circulation amount threshold and a circulation amount range representing a circulation amount screening condition.

[0118] The amount of resource circulation between two resource nodes is the sum of the corresponding resource transfer amounts for each resource circulation between the two resource nodes.

[0119] S414: For each candidate node, if the resource transfer amount is greater than or equal to the circulation amount threshold, or if the resource transfer amount is within the circulation amount interval, the candidate node is determined as a basic node.

[0120] S416, for the candidate nodes, determining the resource flow direction of each resource flow.

[0121] S418, for each resource flow direction, according to the resource circulation amounts of each resource circulation corresponding to the resource flow direction, the total resource transfer amount corresponding to the resource flow direction is counted.

[0122] S420: In each resource flow, if there is a target flow whose total resource transfer amount satisfies the circulation amount screening condition, a candidate node is determined as a basic node.

[0123] S422, according to the circulation amount screening condition, screening multiple basic nodes whose resource circulation amounts meet the circulation amount screening condition from among the candidate nodes.

[0124] S424, determining at least one basic link according to the position of each basic node in the knowledge graph.

[0125] Among them, the basic link is a resource circulation link between at least a part of the basic nodes.

[0126] S426, dividing each basic node in the basic link into node pairs to obtain at least one node pair.

[0127] The node pair includes two adjacent nodes that are located at adjacent node positions in the knowledge graph.

[0128] S428, for each node pair, performing similarity matching on the resource node information of each adjacent node in the node pair to obtain the node similarity corresponding to the node pair.

[0129] S430: When there is a risk node whose node similarity satisfies a similarity condition among resource nodes, a basic link to which the risk node belongs is determined as a risk link.

[0130] Exemplarily, the process of determining the risk link includes: Step 1, reading account entities, loan and transfer relationships to construct a capital relationship graph. Step 2, filtering out nodes and associated relationships with in-and-out degrees greater than 5000 in the graph. Step 3, penetrating downward from the bank acceptance bill discount loan edge. Step 4, transfer edge penetration judgment: transaction time within 5 days; transfer amount greater than or equal to 50,000; greater than or equal to 0.5 times the previous transfer amount; less than or equal to 1.5 times the previous transfer amount; the previous transaction is a loan edge or a transfer edge that has not refluxed. Step 5, risk link judgment: the counterparty name is equal to the client's product name or the holder's direct predecessor name. Further, the risk link can be output and stored in the result table. The result table includes data date, credit funds flow, problem number, customer number, contract number, debt number, expenditure number, customer account number of the outflowing party, customer number of the outflowing party, customer name of the outflowing party, customer account number of the inflowing party, customer name of the inflowing party, customer number of the inflowing party, account institution number of the inflowing party, account institution name of the inflowing party, transaction amount, balance of the outflowing party account, balance of the inflowing party account, transaction date, transaction time, transaction remarks, credit funds flow transaction level description, credit funds payment method code, business number, principal product name, name of the holder's immediate predecessor, counterparty mark, borrower mark, etc.

[0131] Among them, the credit funds flow problem number is used to identify the credit funds return problem involved in a customer's expenditure. The customer number represents the 18-digit customer number. The credit funds flow transaction level description: loan, first-hand flow, second-hand flow, third-hand flow, fourth-hand flow, fifth-hand flow. The credit funds payment method code includes entrusted payment and autonomous payment.

[0132] In this embodiment, a knowledge graph representing resource circulation relations is obtained. The knowledge graph contains resource node information of each of multiple resource nodes and resource circulation information between each resource node. According to the node screening conditions that match the scale of the knowledge graph, multiple basic nodes whose resource circulation information meets the node screening conditions are screened in each resource node, thereby reducing the number of resource nodes and the amount of subsequent information analysis. According to the position of each basic node in the knowledge graph, at least one basic link is determined. The basic link is a resource circulation link between at least a part of each basic node. The basic node is further analyzed based on the basic link. Considering the correlation between the basic nodes, the accuracy of subsequent determination of the risk link is improved. The resource node information of each basic node in the basic link is matched with similarity to obtain the node similarity between each basic node. By using the above method, when there is a risk node whose node similarity meets the similarity conditions between each resource node, the basic link to which the risk node belongs is determined as a risk link. Under the premise of ensuring that an accurate risk link is obtained, the amount of information analysis is greatly reduced and the efficiency of determining the risk link is improved. The method uses the knowledge graph to mine the reflux of bill discount funds, monitor the flow of funds, reduce the output risk link duration, and improve the efficiency of risk analysis.

[0133] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0134] Based on the same inventive concept, the embodiment of the present application also provides a risk analysis device for implementing the risk analysis method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more risk analysis device embodiments provided below can refer to the limitations on the risk analysis method above, and will not be repeated here.

[0135] In one embodiment, as shown in FIG5 , a risk analysis device is provided, including: a graph acquisition module 502, a node screening module 504, a link determination module 506, an information matching module 508 and a risk determination module 510, wherein:

[0136] A graph acquisition module 502 is used to acquire a knowledge graph representing resource circulation relationships; the knowledge graph includes resource node information of each of a plurality of resource nodes, and resource circulation information between the resource nodes;

[0137] The node screening module 504 is used to screen multiple basic nodes whose resource circulation information meets the node screening condition in each resource node according to the node screening condition matching the scale of the knowledge graph;

[0138] The link determination module 506 is used to determine at least one basic link according to the location of each basic node in the knowledge graph; the basic link is a resource circulation link between at least a part of the basic nodes;

[0139] The information matching module 508 is used to perform similarity matching on the resource node information of each basic node in the basic link to obtain the node similarity between the basic nodes;

[0140] The risk determination module 510 is configured to determine the basic link to which the risk node belongs as a risk link when there is a risk node whose node similarity satisfies a similarity condition among the resource nodes.

[0141] In one of the embodiments, resource circulation information includes resource circulation times and resource circulation amounts; node screening conditions include circulation times screening conditions and circulation amount screening conditions; the node screening module 504 is further used to determine circulation times screening conditions and circulation amount screening conditions that match the scale of the knowledge graph; according to the circulation times screening conditions, multiple candidate nodes whose resource circulation times meet the circulation times screening conditions are screened in each resource node; according to the circulation amount screening conditions, multiple basic nodes whose resource circulation amounts meet the circulation amount screening conditions are screened in each candidate node.

[0142] In one of the embodiments, the resource circulation amount between two resource nodes is the sum of the resource transfer amounts corresponding to each resource circulation between the two resource nodes; the node screening module 504 is also used to determine the circulation amount threshold and circulation amount range that represent the circulation amount screening conditions; for each candidate node, if there is a resource transfer amount greater than or equal to the circulation amount threshold, or if there is a resource transfer amount that falls within the circulation amount range, the candidate node is determined as a basic node.

[0143] In one of the embodiments, the candidate node corresponds to multiple resource flows; the node screening module 504 is further used to determine the resource flow direction of each resource flow for the candidate node; for each resource flow direction, the total resource transfer amount corresponding to the resource flow direction is counted according to the resource flow amount of each resource flow corresponding to the resource flow direction; in each resource flow direction, if there is a target flow direction whose total resource transfer amount meets the flow amount screening condition, the candidate node is determined as the basic node.

[0144] In one embodiment, the information matching module 508 is also used to divide each basic node in the basic link into node pairs to obtain at least one node pair; the node pair includes two adjacent nodes that are in adjacent node positions in the knowledge graph; for each node pair, the resource node information of each adjacent node in the node pair is similarly matched to obtain the node similarity corresponding to the node pair.

[0145] In one of the embodiments, the device also includes: a graph construction module, which is used to obtain resource record information of each of multiple resource nodes and a resource extraction template configured for the resource record information; for each resource node, according to the resource extraction template, the resource record information is extracted to obtain the resource node information of the resource node, and the resource flow information between the resource node and other resource nodes; based on the resource node information of each resource node and the resource flow information between each resource node, a knowledge graph representing the resource flow relationship between each resource node is constructed.

[0146] Each module in the above risk analysis device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0147] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be shown in FIG6. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store knowledge graph data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a risk analysis method is implemented.

[0148] Those skilled in the art will understand that the structure shown in FIG. 6 is merely a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0149] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the above method steps are implemented when the processor executes the computer program.

[0150] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the above method steps are implemented.

[0151] In one embodiment, a computer program product is provided, comprising a computer program, which implements the above method steps when executed by a processor.

[0152] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0153] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0154] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A risk analysis method, characterized in that: The method comprises: Acquire a knowledge graph representing resource circulation relations; the knowledge graph includes resource node information of each of a plurality of resource nodes, and resource circulation information between the resource nodes; According to the node screening condition matching the scale of the knowledge graph, multiple basic nodes whose resource circulation information meets the node screening condition are screened in each of the resource nodes; Determine at least one basic link according to the position of each basic node in the knowledge graph; the basic link is a resource circulation link between at least a part of each basic node; Performing similarity matching on the resource node information of each basic node in the basic link to obtain the node similarity between the basic nodes; In the case that there is a risk node whose node similarity satisfies a similarity condition among the resource nodes, the basic link to which the risk node belongs is determined as a risk link.

2. The method according to claim 1, characterized in that The resource circulation information includes the resource circulation times and the resource circulation amount; the node screening conditions include the circulation times screening conditions and the circulation amount screening conditions; The step of screening multiple basic nodes whose resource circulation information satisfies the node screening condition in accordance with the node screening condition matching the scale of the knowledge graph includes: Determine a circulation frequency screening condition and a circulation amount screening condition that match the scale of the knowledge graph; According to the circulation number screening condition, screening multiple candidate nodes whose resource circulation times meet the circulation number screening condition from among the resource nodes; According to the circulation amount screening condition, multiple basic nodes whose resource circulation amounts meet the circulation amount screening condition are screened from the candidate nodes.

3. The method according to claim 2, characterized in that The resource flow amount between two resource nodes is the sum of the resource transfer amounts corresponding to each resource flow between the two resource nodes; The step of screening multiple basic nodes whose resource circulation amounts satisfy the circulation amount screening conditions from among the candidate nodes according to the circulation amount screening conditions includes: Determining a circulation amount threshold and a circulation amount range representing the circulation amount screening condition; For each candidate node, if the resource transfer amount is greater than or equal to the circulation amount threshold, or if the resource transfer amount falls within the circulation amount interval, the candidate node is determined as a basic node.

4. The method according to claim 2, characterized in that: The resource circulation times corresponding to the candidate node is multiple times; The step of screening multiple basic nodes whose resource circulation amounts satisfy the circulation amount screening conditions from among the candidate nodes according to the circulation amount screening conditions includes: For the candidate nodes, determining the resource flow direction of each resource flow; For each resource flow direction, according to the resource circulation amounts of each resource circulation corresponding to the resource flow direction, the total resource transfer amount corresponding to the resource flow direction is counted; In each of the resource flows, if there is a target flow whose total resource transfer amount satisfies the circulation amount screening condition, the candidate node is determined as the basic node.

5. The method according to claim 1, characterized in that The performing similarity matching on the resource node information of each basic node in the basic link to obtain the node similarity between the basic nodes includes: Performing node pair division on each basic node in the basic link to obtain at least one node pair; the node pair includes two adjacent nodes at adjacent node positions in the knowledge graph; For each of the node pairs, similarity matching is performed on the resource node information of each adjacent node in the node pair to obtain the node similarity corresponding to the node pair.

6. The method according to claim 1, characterized in that The method further comprises: Acquire resource record information of each of a plurality of resource nodes and a resource extraction template configured for the resource record information; For each of the resource nodes, extract the resource record information according to the resource extraction template to obtain the resource node information of the resource node and the resource flow information between the resource node and other resource nodes; Based on the resource node information of each of the resource nodes and the resource circulation information between the resource nodes, a knowledge graph representing the resource circulation relationship between the resource nodes is constructed.

7. A risk analysis device, characterized in that: The device comprises: A graph acquisition module, used to acquire a knowledge graph representing resource circulation relations; the knowledge graph includes resource node information of each of a plurality of resource nodes, and resource circulation information between the resource nodes; A node screening module, used to screen multiple basic nodes whose resource circulation information meets the node screening conditions in each resource node according to the node screening conditions matching the scale of the knowledge graph; A link determination module, used to determine at least one basic link according to the location of each basic node in the knowledge graph; the basic link is a resource circulation link between at least a part of each basic node; An information matching module, used for performing similarity matching on the resource node information of each basic node in the basic link to obtain the node similarity between the basic nodes; The risk determination module is used to determine the basic link to which the risk node belongs as a risk link when there is a risk node whose node similarity satisfies a similarity condition among the resource nodes.

8. The device according to claim 7, characterized in that The resource circulation information includes the resource circulation times and the resource circulation amount; the node screening condition includes the circulation times screening condition and the circulation amount screening condition; the node screening module is further used to determine the circulation times screening condition and the circulation amount screening condition that match the scale of the knowledge graph; According to the circulation number screening condition, screening multiple candidate nodes whose resource circulation times meet the circulation number screening condition from among the resource nodes; According to the circulation amount screening condition, multiple basic nodes whose resource circulation amounts meet the circulation amount screening condition are screened from the candidate nodes.

9. The device according to claim 8, characterized in that The resource circulation amount between two resource nodes is the sum of the resource transfer amounts corresponding to each resource circulation between the two resource nodes; the node screening module is further used to determine the circulation amount threshold and circulation amount interval representing the circulation amount screening condition; For each candidate node, if the resource transfer amount is greater than or equal to the circulation amount threshold, or if the resource transfer amount falls within the circulation amount interval, the candidate node is determined as a basic node.

10. The device according to claim 9, characterized in that The number of resource flows corresponding to the candidate node is multiple; the node screening module is further used to determine the resource flow direction of each resource flow for the candidate node; for each resource flow direction, according to the resource flow amount of each resource flow corresponding to the resource flow direction, the total resource transfer amount corresponding to the resource flow direction is counted; In each of the resource flows, if there is a target flow whose total resource transfer amount satisfies the circulation amount screening condition, the candidate node is determined as the basic node.

11. The device according to claim 7, characterized in that The information matching module is further used to divide each basic node in the basic link into node pairs to obtain at least one node pair; the node pair includes two adjacent nodes that are in adjacent node positions in the knowledge graph; For each of the node pairs, similarity matching is performed on the resource node information of each adjacent node in the node pair to obtain the node similarity corresponding to the node pair.

12. The device according to claim 7, characterized in that: The device also includes: a graph construction module, which is used to obtain resource record information of each of the multiple resource nodes and a resource extraction template configured for the resource record information; for each of the resource nodes, according to the resource extraction template, the resource record information is extracted to obtain the resource node information of the resource node and the resource circulation information between the resource node and other resource nodes; based on the resource node information of each of the resource nodes and the resource circulation information between the resource nodes, a knowledge graph representing the resource circulation relationship between the resource nodes is constructed.

13. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images