A method and system for identifying risk business behavior of a life insurance company

Abstract

The application provides a life insurance company risk business behavior identification method and system, the method comprises the following steps: taking a policy, a customer and a sales agent as a node, taking a policy, a customer and a sales agent between a policy relationship, a sales relationship and a loan relationship as an edge, and establishing a risk business behavior identification business graph; according to the policy node, the sales agent node and the customer node included in the subgraph of the risk business behavior identification business graph and the edges between the policy node, the sales agent node and the customer node, a first risk business behavior is identified; according to the policy node, the customer node included in the subgraph of the risk business behavior identification business graph and the edges between the policy node and the customer node, a second risk business behavior is identified, and the risk business behavior identification efficiency and reliability are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of abnormal insurance behavior identification, in particular to a life insurance company risk business behavior identification method and system. BACKGROUND

[0002] In the life insurance company insurance business, "splitting" refers to the condition that the policy sold by the sales agent meets the "three same and one different" condition (the same customer, the same month, and the same insurance policy under different sales agents), and the "circular insurance" behavior refers to the customer's short-term insurance after the loan (the customer can apply for a loan according to the cash value of the effective policy), and the loan is not repaid when the insurance is purchased. It can be simply understood as "repairing the west wall with the east wall", and the above two behaviors are abnormal sales behaviors caused by sales team due to assessment pressure and arbitrage, which can easily cause false premiums, increase company sales costs, and cause losses and operational risks to the company.

[0003] Therefore, the life insurance company needs to timely discover the "splitting" or "circular insurance" behavior. The traditional identification method is generally based on a relational database, and the splitting behavior or circular insurance behavior is identified by analyzing the association of multiple business libraries. The splitting and circular insurance behaviors are identified by a large number of table associations, the query cost is high, the identification efficiency is difficult to guarantee, the coding amount required to realize the function is high, the identification result is difficult to quickly visualize, and the "risk" behavior cannot be intuitively understood, so that the identification efficiency of the splitting behavior and / or the circular insurance behavior is not high.

[0004] In order to solve the problem of low identification efficiency of the splitting behavior and / or the circular insurance behavior in the prior art, the present application is proposed. SUMMARY

[0005] In order to solve the problems existing in the prior art, the present application innovatively provides a life insurance company risk business behavior identification method and system, which effectively solves the problem of low identification efficiency of the splitting behavior and / or the circular insurance behavior caused by the prior art, and effectively improves the identification efficiency of the splitting behavior and / or the circular insurance behavior.

[0006] The first aspect of the present application provides a life insurance company risk business behavior identification method, comprising:

[0007] A risk business behavior identification business graph is established by taking the policy, the customer, and the sales agent as nodes, and taking the insurance relationship, the sales relationship, and the loan relationship among the policy, the customer, and the sales agent as edges. The risk business behavior includes a first risk business behavior and / or a second risk business behavior, the first risk business behavior is a splitting behavior, and the second risk business behavior is a circular insurance behavior.

[0008] According to the risk business behavior, a subgraph included in the business graph is identified, and the subgraph includes a policy node, a sales agent node, a customer node, an edge between the policy node, the sales agent node, and the customer node, and a first risk business behavior is identified; and / or,

[0009] According to the risk business behavior, a subgraph included in the business graph is identified, and the subgraph includes a policy node, a sales agent node, a customer node, an edge between the policy node, the sales agent node, and the customer node, and a first risk business behavior is identified; and / or,

[0010] Optionally, each customer node in the risk business behavior identification business graph represents a customer; each policy node represents a policy; and each sales agent node represents a sales agent.

[0011] The customer node and the policy node have a first edge and a second edge, and the policy node and the sales agent node have a third edge; the first edge is from the customer node to the policy node and represents an insurance relationship; the second edge is from the customer node to the policy node and represents a policy loan relationship; the third edge is from the sales agent node to the policy node and represents a policy sales relationship; and the policy node includes policy type attribute information.

[0012] Further, the first edge is marked as an insurance type, the second edge is marked as a loan type, and the third edge is marked as a sales type.

[0013] The first edge includes first attribute information and second attribute information, the first attribute information is an insurance time, and the second attribute information is an insurance effective time.

[0014] The second edge includes fourth attribute information and fifth attribute information, the fourth attribute information is a policy loan time, and the fifth attribute information is a policy loan repayment time.

[0015] Further, according to the risk business behavior, a subgraph included in the business graph is identified, and the subgraph includes a policy node, a sales agent node, a customer node, an edge between the policy node, the sales agent node, and the customer node, and a first risk business behavior is identified.

[0016] If the policy node, the sales agent node, and the customer node included in a certain subgraph in the risk business behavior identification business graph, and the edges between the policy node, the sales agent node, and the customer node satisfy a preset first risk identification rule, then the insurance behavior involved in the subgraph constitutes a first risk business behavior.

[0017] Further, the first risk identification rule includes:

[0018] The same customer node is connected to at least two policy nodes; and

[0019] The first type of policy node corresponds to connection of at least two different sales agent nodes, wherein the first type of policy node is at least two policy nodes connected with the same customer node; and

[0020] The first attribute information or the second attribute information of the first edge between the same customer node and the first type of policy node is in the same year and month; and

[0021] The policy type attribute information of the first type of policy node is the same.

[0022] Optionally, the risk business behavior comprises the policy node, the customer node, and the edge between the policy node and the customer node in the subgraph of the business graph according to the risk business behavior, and the second risk business behavior specifically comprises:

[0023] If the policy node, the customer node, and the edge between the policy node and the customer node in the subgraph of the business graph satisfy a preset second risk identification rule, a policy application behavior involved in the subgraph constitutes the second risk business behavior.

[0024] Further, the second risk identification rule comprises:

[0025] The same customer node corresponds to at least two policy nodes, and at least one policy node in the first type of policy node and the customer node have a second edge therebetween, wherein the first type of policy node is at least two policy nodes connected with the same customer node; and

[0026] The first attribute information of the first type of first edge is less than the fourth attribute information of the first type of second edge, wherein the first type of first edge is the first edge between the same customer node and the policy node having the second edge with the customer node; the first type of second edge is the second edge between the same customer node and the policy node having the second edge with the customer node; and

[0027] The fourth attribute information of the first type of second edge is less than the first attribute information of the second type of first edge, wherein the second type of first edge is the first edge between the same customer node and the policy node not having the second edge with the customer node; and

[0028] The fifth attribute information of the first type of second edge is greater than the first attribute information of the second type of first edge; and

[0029] The policy type attribute information of the first type of policy node is the same.

[0030] Optionally, the more the number of sales agents included in each subgraph of the first risk business behavior, or the more the number of policy nodes corresponding to each sales agent in the sales agents included in each subgraph of the first risk business, the higher the risk level of the first risk business behavior corresponding to the subgraph.

[0031] The more the number of the policy nodes corresponding to the same customer node included in each subgraph involved in the second risk business behavior, or the more the number of the policy nodes having the second edge with the same customer node included in each subgraph involved in the second risk business, the higher the risk level of the second risk business behavior.

[0032] Optionally, the system further comprises:

[0033] labeling the customer nodes, the policy nodes, the sales agent nodes included in each subgraph involved in the first risk business behavior, and the edges between the customer nodes, the policy nodes, the sales agent nodes involved in the first risk business; and / or,

[0034] labeling the customer nodes, the policy nodes included in each subgraph involved in the second risk business behavior, and the edges between the customer nodes, the policy nodes involved in the second risk business.

[0035] The second aspect of the present application provides a system for identifying risk business behaviors of a life insurance company, comprising:

[0036] a building module for building a risk business behavior identification business graph with the policy, the customer, and the sales agent as nodes, and the insurance relationship, the sales relationship, and the loan relationship between the policy, the customer, and the sales agent as edges; wherein the risk business comprises a first risk business and a second risk business, the first risk business is a policy splitting behavior, and the second risk business is a circular insurance behavior;

[0037] a first identification module for identifying the first risk business according to the policy nodes, the sales agent nodes, the customer nodes included in a subgraph in the risk business behavior identification business graph, and the edges between the policy nodes, the sales agent nodes, and the customer nodes; and / or,

[0038] a second identification module for identifying the second risk business according to the policy nodes, the customer nodes included in a subgraph in the risk business behavior identification business graph, and the edges between the policy nodes and the customer nodes.

[0039] The technical solutions adopted by the present application have the following technical effects:

[0040] 1. The technical solution of this invention uses policies, customers, and sales agents as nodes, and the insurance, sales, and loan relationships among policies, customers, and sales agents as edges to establish a risk business behavior identification business graph. Based on the policy nodes, sales agent nodes, and customer nodes included in the subgraphs of the risk business behavior identification business graph, and the edges between policy nodes, sales agent nodes, and customer nodes, the first risk business behavior is identified. Based on the policy nodes, customer nodes, and the edges between policy nodes and customer nodes included in the subgraphs of the risk business behavior identification business graph, the second risk business behavior is identified. This effectively solves the problem of low identification efficiency of splitting and / or revolving insurance behaviors caused by existing technologies, and effectively improves the identification efficiency of splitting and / or revolving insurance behaviors.

[0041] 2. The technical solution of this invention abandons the traditional data modeling method based on relational databases. Based on the "graph model", it constructs a business graph in the graph database to identify "splitting" and "circular insurance" behaviors. Through ingenious data modeling, combined with the advantages of graph databases in graph data analysis and mining, it efficiently identifies splitting and circular insurance behaviors through "subgraph recognition", which reduces the difficulty of recognition, improves the recognition efficiency, and enhances the accuracy and comprehensiveness of the recognition results.

[0042] 3. In the technical solution of this invention, the customer nodes, policy nodes, and sales agent nodes included in each subgraph involved in the first risk business, as well as the edges between the customer nodes, policy nodes, and sales agent nodes involved in the first risk business, are marked; the customer nodes, policy nodes included in each subgraph involved in the second risk business, as well as the edges between the customer nodes and policy nodes involved in the second risk business, are marked. The identification return result is still graph-type data (subgraph), which is easy to visualize and can fully deconstruct the relevant parties of "splitting" and cyclic insurance behavior.

[0043] 4、The more the number of sales agents included in each subgraph involved in the first risk business behavior, or the more the number of policy nodes corresponding to each sales agent included in the first risk business involved in each subgraph, the higher the risk level of the first risk business behavior corresponding to the subgraph; the more the number of policy nodes corresponding to the same customer node included in each subgraph involved in the second risk business, or the more the number of policy nodes having a second edge with the same customer node included in each subgraph involved in the second risk business, the higher the risk level of the second risk business behavior, which can quickly identify subgraphs with high risk, reduce company losses and fraud risks, improve business quality, help life insurance companies efficiently identify "single splitting" and "circular insurance" behaviors, fully "understand" various representative behavior patterns involved in the "single splitting behavior" of the sales agent and the "circular insurance behavior" of the customer, and focus on the sales agents and customers who frequently produce the above risk business behaviors, risk prompts and warnings, guide business behavior to be more standardized, and help the company develop high-quality.

[0044] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0046] Figure 1 A flowchart of the method of embodiment one in the present application scheme;

[0047] Figure 2 An explanatory diagram of "single splitting" and "circular insurance" behaviors in the method of embodiment one in the present application scheme;

[0048] Figure 3 An example diagram of risk business behavior identification business graph in the method of embodiment one in the present application scheme;

[0049] Figure 4 An example diagram of subgraph identification in the risk business behavior identification business graph in the method of embodiment one in the present application scheme;

[0050] Figure 5 An example diagram of single splitting behavior mode 1 in the method of embodiment one in the present application scheme;

[0051] Figure 6 An example diagram of single splitting behavior mode 2 in the method of embodiment one in the present application scheme;

[0052] Figure 7 Figure 3 is an example schematic diagram of the mode 3 of the single disassembling behavior (the first risk business behavior) in the method of the embodiment one in the present application solution;

[0053] Figure 8 Figure 4 is an example schematic diagram of the mode 4 of the single disassembling behavior in the method of the embodiment one in the present application solution;

[0054] Figure 9 Figure 5 is an example schematic diagram of the mode 1 of the circular insurance behavior (the second risk business behavior) in the method of the embodiment one in the present application solution;

[0055] Figure 10 Figure 6 is an example schematic diagram of the mode 2 of the circular insurance behavior in the method of the embodiment one in the present application solution;

[0056] Figure 11 Figure 7 is another flow schematic diagram of the method of the embodiment one in the present application solution;

[0057] Figure 12 Figure 8 is a structure schematic diagram of the system of the embodiment two in the present application solution. DETAILED DESCRIPTION

[0058] To clearly illustrate the technical features of the present solution, the present application is described in detail below with specific embodiments and in conjunction with the accompanying drawings. The following disclosure provides many different embodiments or examples for implementing the different structures of the present application. In order to simplify the disclosure of the present application, the components and settings of specific examples are described below. In addition, the present application can repeatedly refer to the numbers and / or letters in different examples. Such repetition is for the purpose of simplification and clarity, and it does not indicate the relationship between the various embodiments and / or settings discussed. It should be noted that the components illustrated in the drawings are not necessarily drawn to scale. The present application omits the description of well-known components and processing techniques and processes to avoid unnecessary limitations on the present application.

[0059] Embodiment one

[0060] As shown in Figure 1 , the present application provides a method for identifying risk business behaviors of a life insurance company, comprising:

[0061] S1, establishing a risk business behavior identification business graph taking the insurance policy, the customer, and the sales agent as nodes and taking the insurance relationship, the sales relationship, and the loan relationship among the insurance policy, the customer, and the sales agent as edges; wherein the risk business behavior comprises a first risk business behavior and / or a second risk business behavior, the first risk business behavior is a single disassembling behavior, and the second risk business behavior is a circular insurance behavior;

[0062] S2, according to the risk business behavior, identifying the policy node, the sales agent node, the customer node, the edge between the policy node, the sales agent node and the customer node included in the subgraph of the business graph, identifying the first risk business behavior; and / or,

[0063] S3, according to the risk business behavior, identifying the policy node, the customer node, and the edge between the policy node and the customer node included in the subgraph of the business graph, identifying the second risk business behavior.

[0064] In step S1, as shown in the following table, both the "single splitting" behavior identification and the "circular insurance" behavior identification need to analyze the business connection among the policy, the sales agent and the customer, as shown in the following table, therefore, this project takes maximizing the identification ability of the "single splitting" behavior and the "circular insurance" behavior as the principle, and constructs a special graph model, as shown in the following table, which contains all the element information required for the identification of the two risk behaviors. Figures 2-3 Figure 2 Figure 3

[0065] Each customer node in the risk business behavior identification business graph represents a customer; each policy node represents a policy; and each sales agent node represents a sales agent.

[0066] There is a first edge and a second edge between the customer node and the policy node, and the policy node and the sales agent node have a third edge; the first edge is from the customer node to the policy node, representing an insurance relationship; the second edge is from the customer node to the policy node, representing a policy loan relationship; the third edge is from the sales agent node to the policy node, representing a policy sales relationship; the policy node includes policy type attribute information.

[0067] The first edge is marked as an insurance type, the second edge is marked as a loan type, and the third edge is marked as a sales type.

[0068] The first edge includes first attribute information, second attribute information and third attribute information, the first attribute information is the insurance time, the second attribute information is the insurance effective time (or effective time), and the third attribute information is the insurance effective premium (or effective premium);

[0069] The second edge includes fourth attribute information, fifth attribute information and sixth attribute information, the fourth attribute information is the policy loan time (or loan time), the fifth attribute information is the policy loan repayment time (or repayment time), and the sixth attribute information is the policy loan amount (or loan amount).

[0070] ​​​The third edge includes seventh attribute information, eighth attribute information and ninth attribute information, the seventh attribute information is a time of insurance, the eighth attribute information is a time of insurance taking effect (or effective time), and the ninth attribute information is a premium of insurance taking effect (or effective premium), so that the management and maintenance personnel can know the attribute information of the nodes and edges involved in the first risk business behavior or the second risk business behavior in time.

[0071] That is, there are three point types in the risk business behavior identification business graph: a policy, a customer and a sales agent, which respectively represent a policy, a customer and a sales agent.

[0072] A first edge of a type of "insurance" is used to connect a node of the type of "customer" and a node of the type of "policy", the first edge is pointed from the node of the type of "customer" to the node of the type of "policy", and represents an insurance behavior of the customer for the policy; a third edge of a type of "sales" is used to connect a node of the type of "sales agent" and a node of the type of "policy", the third edge is pointed from the node of the type of "sales agent" to the node of the type of "policy", and represents a sales behavior of the sales agent for the policy; and a second edge of a type of "borrowing" is used to connect a node of the type of "customer" and a node of the type of "policy", the second edge is pointed from the node of the type of "customer" to the node of the type of "policy", and represents a borrowing behavior of the customer for the policy.

[0073] The first edge of the type of "insurance" includes three attributes of a time of insurance of the policy, a time of insurance taking effect of the policy and a premium of insurance taking effect of the policy, i.e., first attribute information, second attribute information and third attribute information.

[0074] The second edge of the type of "borrowing" includes three attributes of a time of policy borrowing of the customer, a time of repayment of policy borrowing of the customer and an amount of policy borrowing of the customer, i.e., fourth attribute information, fifth attribute information and sixth attribute information.

[0075] The third edge of the type of "sales" includes three attributes of a time of insurance of the policy, a time of taking effect of the policy and a premium of insurance taking effect of the policy, i.e., seventh attribute information, eighth attribute information and ninth attribute information.

[0076] The node of the type of "policy" includes an attribute of a type of insurance of the policy.

[0077] In step S2, according to the policy nodes, the sales agent nodes and the customer nodes included in the subgraph of the risk business behavior identification business graph and the edges between the policy nodes, the sales agent nodes and the customer nodes, the first risk business behavior specifically includes:

[0078] If the policy node, sales agent node, and customer node, and the edges between the policy node, sales agent node, and customer node in a certain subgraph of the risk business behavior identification business graph satisfy the preset first risk identification rule, then the insurance purchase behavior involved in that subgraph constitutes the first risk business behavior.

[0079] Specifically, the first risk identification rule includes:

[0080] Each customer node corresponds to at least two policy nodes; and,

[0081] The first type of policy node corresponds to at least two different sales agent nodes, wherein a first-type policy node is at least two policy nodes connected to the same customer node (i.e., different policy nodes connected to the same customer node); and,

[0082] The year and month of the first attribute information or second attribute information on the first side are the same for the same customer node and the first type of policy node; and...

[0083] The policy type attribute information of the first type of policy nodes is the same.

[0084] Specifically, compared to relational database modeling, graph models have a stronger ability to express business rules and are suitable for analyzing complex relationships between actors. Reflected in the graph, the "splitting of orders" behavior is a specific subgraph pattern (graph pattern: the basic unit describing a graph structure, including the definition of vertices and edges and the relationship types between them, edge types, edge directions, and the attributes of vertices and edges). Through "subgraph identification," the relationships between "sales agents" and "policies" that "conform" to the splitting of orders behavior rules can be discovered, such as... Figure 4 As shown, the nodes and edges with thickened lines constitute a subgraph that is identified as either the first or second business risk behavior.

[0085] This method is independent of specific graph database types. Taking Nebula graph database as an example, based on the business graph constructed above, the following nGQL (NebulaGraph declarative query language, compatible with most opencypher syntax) can be used to quickly identify subgraphs that conform to the "order splitting" behavior pattern. The expression is natural, and the specific implementation is as follows:

[0086] MATCH p = (a: Agent) - [: Sales] -> (b: Policy) <- [f: Insurance Application] - (e: Customer) - [g: Insurance Application] -> (c: Policy) <- [: Sales] - (d: Agent)

[0087] WHERE f. Insurance application date (month) == g. Insurance application date (month)

[0088] AND f. Insurance period. year == g. Insurance period. year

[0089] AND b! = c

[0090] AND b. Insurance policy. Type of insurance == c. Insurance policy. Type of insurance

[0091] AND a! = d

[0092] RETURN p limit 100 (to ensure smooth and fast execution);

[0093] Taking the identification of "order splitting" behavior as an example:

[0094] like Figure 5 As shown, in policy splitting behavior mode 1, the policy splitting behavior involves two agents, and each agent has only one policy involved in the splitting; the customer node (Wang xx) is connected to two policy nodes (policy 1 and policy 2); and,

[0095] The first type of policy node (Policy 1 and Policy 2) corresponds to two different sales agent nodes (Ai xx and Li xx); and,

[0096] The first attribute information (insurance application time) or the second attribute information (insurance effective time) on the first side of the same customer node and the first type of policy node are the same in both year and month; and the policy type attribute information of the first type of policy nodes are all the same.

[0097] like Figure 6 As shown, in policy splitting behavior mode 2, the policy splitting behavior involves two agents, and one agent has two or more policies involving policy splitting; the customer node (Guo xx) is connected to four policy nodes (Policy 1, Policy 2, Policy 3, Policy 4); and,

[0098] The first type of policy node (Policy 1, Policy 2, Policy 3, Policy 4) corresponds to two different sales agent nodes (Teng xx, Cao xx); and,

[0099] The first attribute information (insurance application time) or the second attribute information (insurance effective time) on the first side of the same customer node and the first type of policy node are the same in both year and month; and the policy type attribute information of the first type of policy nodes are all the same.

[0100] like Figure 7 As shown, in policy splitting behavior mode 3, the policy splitting behavior involves three or more agents, and one agent has two or more policies involving policy splitting; the customer node (Wang xx) is connected to four policy nodes (Policy 1, Policy 2, Policy 3, Policy 4); and,

[0101] The first type of policy node (policy 1, policy 2, policy 3, policy 4) is connected to three different sales agent nodes (Li XX, Huang XX, Li X); and,

[0102] The first attribute information (insurance time) or the second attribute information (insurance effective time) of the first edge between the same customer node and the first type of policy node is in the same year and month; and the policy type attribute information of the first type of policy node is the same.

[0103] As shown in Figure 8 The single behavior in the single behavior mode 4 involves three or more agents, and two or more policies of an agent are involved in the single; the customer node (Wang XX) is connected to six policy nodes (policy 1, policy 2, policy 3, policy 4, policy 5, policy 6); and,

[0104] The first type of policy node (policy 1, policy 2, policy 3, policy 4, policy 5, policy 6) is connected to three different sales agent nodes (Li XX, Jiang XX, Han XX); and,

[0105] The first attribute information (insurance time) or the second attribute information (insurance effective time) of the first edge between the same customer node and the first type of policy node is in the same year and month; and the policy type attribute information of the first type of policy node is the same.

[0106] As the complexity of the single behavior mode increases, the business risk also increases (the risk level of the single behavior mode 1, the single behavior mode 2, the single behavior mode 3, the single behavior mode 4 increases in turn), which fully proves the practicability and reliability of the recognition scheme in the field of "single" behavior recognition.

[0107] That is, the more the number of sales agents included in each subgraph involved in the first risk business behavior, or the more the number of policy nodes corresponding to each sales agent included in each subgraph involved in the first risk business, the higher the risk level of the first risk business behavior corresponding to the subgraph.

[0108] In step S3, the second risk business behavior is identified according to the policy nodes, customer nodes, edges between the policy nodes and the customer nodes included in the subgraph in the risk business behavior identification business graph, which specifically includes:

[0109] If the policy nodes, customer nodes, and edges between the policy nodes and the customer nodes in the risk business behavior identification business graph meet the preset second risk identification rule, the insurance behavior involved in the subgraph constitutes a second risk business behavior.

[0110] Specifically, the second risk identification rule includes:

[0111] At least two policy nodes correspond to the same customer node, and there is a second edge between at least one policy node in the first type of policy nodes and the customer node, where the first type of policy nodes are at least two policy nodes corresponding to and connected to the same customer node; and,

[0112] The first attribute information of the first type of first edges is all less than the fourth attribute information of the first type of second edges, where the first type of first edges are the first edges (the first edges between the customer node with the second edge and the policy node) between the same customer node and the policy node with the second edge between the customer node and the policy node; the first type of second edges are the second edges (the second edges between the customer node with the second edge and the policy node) between the same customer node and the policy node with the second edge between the customer node and the policy node; and,

[0113] The fourth attribute information of the first type of second edges is all less than the first attribute information of the second type of first edges, where the second type of first edges are the first edges (the first edges between the customer node without the second edge and the policy node) between the same customer node and the policy node without the second edge between the customer node and the policy node; and,

[0114] The fifth attribute information of the first type of second edges is all greater than the first attribute information of the second type of first edges; and,

[0115] The policy type attribute information of the first type of policy nodes is all the same.

[0116] Specifically, compared with the relational database modeling method, the graph model has strong business rule expression ability and is suitable for analyzing complex associations between actors. Reflected in the graph, the "repeated insurance" behavior is a specific subgraph pattern (graph pattern: the basic unit describing the graph structure, including the definitions of nodes and edges and the relationship types, edge types, edge directions, and attributes of nodes and edges) between "customers" and "policies" that conform to the "repeated insurance" behavior rules can be discovered through "subgraph recognition", as Figure 4 shown.

[0117] This method does not depend on a specific graph database type. Taking the Nebula graph database as an example, based on the above constructed business graph, the following nGQL (NebulaGraph graph database declarative query language, compatible with most opencypher syntax) can be used to quickly identify subgraphs that conform to the "repeated insurance" behavior pattern, and the expression is natural. The specific implementation method is:

[0118] MATCH p=(c:policy)<-[g:insured]-(b:customer)-[f:borrow]->(a:policy)<-[e:insured]-(b:customer)

[0119] WHERE e.insured_time < f.borrow_time

[0120] AND f. Borrowing time < g. Insurance application time

[0121] AND f. Repayment time > g. Insurance application time

[0122] AND a!= c

[0123] AND a. Policy. Insurance type == c. Policy. Insurance type

[0124] RETURN p LIMIT 100;

[0125] As Figure 9 shown, the same customer node (Xu xx) corresponds to two policy nodes (Policy 1 and Policy 2), and there is a second edge (loan) between the policy node (Policy 1) in the first type of policy nodes (Policy 1 and Policy 2) and this customer node; and, [[ID=2,0]]

[0126] The first attribute information (insurance application time) of the first edge in the first type is less than the fourth attribute information (borrowing time) of the second edge in the first type. Among them, the first edge in the first type is the first edge between the same customer node (Xu xx) and the policy node (Policy 1) that has a second edge with this customer node; the second edge in the first type is the second edge between the same customer node (Xu xx) and the policy node (Policy 1) that has a second edge with this customer node; and,

[0127] The fourth attribute information (policy loan time) of the second edge in the first type is less than the first attribute information (insurance application time) of the first edge in the second type. Among them, the first edge in the second type is the first edge between the same customer node (Xu xx) and the policy node (Policy 2) that has no second edge with this customer node; and,

[0128] The fifth attribute information (repayment time of policy loan) of the second edge in the first type is greater than the first attribute information (insurance application time) of the first edge in the second type; and, the policy insurance type attribute information of the first type of policy nodes is the same.

[0129] As Figure 10 shown, the same customer node (Xu xx) corresponds to four policy nodes (Policy 1, Policy 2, Policy 3, Policy 4), and there is a second edge (loan) between the policy nodes (Policy 1, Policy 3, Policy 4) in the first type of policy nodes (Policy 1, Policy 2, Policy 3, Policy 4) and this customer node; and,

[0130] The first attribute information (insurance time) of the first edge of the first type is less than the fourth attribute information (loan time) of the second edge of the first type, wherein the first edge of the first type is the first edge between the same customer node (Xu XX) and the policy node (policy 1, policy 3, policy 4) having the second edge with the customer node; the second edge of the first type is the second edge between the same customer node (Xu XX) and the policy node (policy 1, policy 3, policy 4) having the second edge with the customer node; and

[0131] The fourth attribute information (policy loan time) of the second edge of the first type is less than the first attribute information (insurance time) of the first edge of the second type, wherein the first edge of the second type is the first edge between the same customer node (Xu XX) and the policy node (policy 2, policy 5) not having the second edge with the customer node; and

[0132] The fifth attribute information (policy loan repayment time) of the second edge of the first type is greater than the first attribute information (insurance time) of the first edge of the second type; and the policy type attribute information of the first policy node is the same.

[0133] Compared with the traditional method, the technical scheme of the application has simple recognition mode and higher recognition efficiency. The traditional relational database recognition method needs at least dozens of lines of code. The technical scheme of the application benefits from the cleverness of data modeling, does not need complex queries, and can be realized by ten lines of code or less. Moreover, the recognition content is easy to visualize and analyze.

[0134] Using visualization technology, the circular insurance behavior recognition result is visualized, the risk behavior is deconstructed through graph display, the related parties involved in the risk behavior are intuitively displayed, and risk warning and prevention and control are efficiently realized.

[0135] The scheme identifies the circular insurance behavior in the special business graph through the "subgraph identification" mode, the risk information identified is rich, is naturally suitable for visualization deconstruction, and can identify various behavior modes.

[0136] For the circular insurance behavior recognition scene, the scheme can also identify various circular insurance behavior modes, such as Figure 9 As shown in the figure, the customer insures again after insuring for a policy loan. As shown in the figure, Figure 10 As shown in the figure, the customer insures again after insuring for multiple policy loans, and the loans are completed in multiple times. Figure 10 The risk level of the circular insurance behavior shown in the figure is higher than Figure 9 The risk level of the circular insurance behavior shown in the figure.

[0137] That is, the more the number of the policy nodes corresponding to the same customer node included in each subgraph involved in the second risk business behavior, or the more the number of the policy nodes having the second edge with the same customer node included in each subgraph involved in the second risk business, the higher the risk level of the second risk business behavior.

[0138] It should be noted that, as shown in Figure 1 Steps S2 and S3 can be executed separately or together, if only the first risk business behavior is identified, only step S2 can be executed, that is, the complete execution sequence is S1-S2; if only the second risk business behavior is identified, only step S3 can be executed, that is, the complete execution sequence is S1-S3; if the first risk business behavior and the second risk business behavior need to be identified, steps S2 and S3 can be executed, and the execution sequence of steps S2 and S3 is not required, that is, the complete execution sequence is S1-S2-S3 or S1-S3-S2.

[0139] In addition, it should be noted that the risk business behavior identification business graph in the technical scheme of the present application can support periodic or non-periodic updating, and the updating mode can be automatic updating or manual updating, which is not limited in the present application.

[0140] As shown in Figure 11 The technical scheme of the present application further provides a risk business behavior identification method of a life insurance company, which further comprises:

[0141] S4, marking the customer nodes, policy nodes and sales agent nodes included in each subgraph involved in the first risk business behavior, and the edges between the customer nodes, policy nodes and sales agent nodes involved in the first risk business; and / or,

[0142] S5, marking the customer nodes and policy nodes included in each subgraph involved in the second risk business behavior, and the edges between the customer nodes and policy nodes involved in the second risk business.

[0143] It should be noted that, as shown in Figure 11As shown, step S4 and step S5 can be executed separately or together, if only the first risk business behavior is identified, only step S4 can be executed, that is, the complete execution sequence is S1-S2-S4; if only the second risk business behavior is identified, only step S5 can be executed, that is, the complete execution sequence is S1-S3-S5; if the first risk business behavior and the second risk business behavior need to be identified, steps S4 and S5 can be executed, and the execution sequence of steps S4 and S5 is not required, that is, the complete execution sequence is S1-S2-S3-S4-S5, or S1-S2-S3-S5-S4, or S1-S3-S2-S4-S5, or S1-S3-S2-S5-S4, or S1-S2-S4-S3-S5, or S1-S3-S5-S2-S4.

[0144] Compared with the conventional mode, the technical scheme of the present application has simple identification mode and higher identification efficiency, the conventional relational database identification mode needs at least dozens of lines of code, the technical scheme of the present application is benefited from the ingenious data modeling, does not need complex query, and can be realized by ten lines of code or less, and the identification content is easy to be visualized and analyzed.

[0145] The visualization technology is used to visualize the single splitting behavior identification result, the risk behavior is deconstructed through the graph presentation mode, the related parties involved in the risk behavior are intuitively presented, and risk warning and prevention and control are efficiently realized.

[0146] The technical scheme of the present application identifies the single splitting and the circular insurance behaviors in the special business graph through the subgraph identification mode, the risk information is rich, is naturally suitable for visualization deconstruction, and can identify various behavior modes.

[0147] In the technical scheme of the present application, the insurance policy, the customer and the sales agent are taken as nodes, the insurance policy, the customer and the sales agent are taken as edges, a risk business behavior identification business graph is established, the first risk business behavior is identified according to the insurance policy node, the sales agent node and the customer node included in the subgraph of the risk business behavior identification business graph and the edges between the insurance policy node, the sales agent node and the customer node, and the second risk business behavior is identified according to the insurance policy node, the customer node included in the subgraph of the risk business behavior identification business graph and the edges between the insurance policy node and the customer node, the problem of low identification efficiency of the single splitting behavior and / or the circular insurance behavior caused by the prior art is solved, and the identification efficiency of the single splitting behavior and / or the circular insurance behavior is effectively improved.

[0148] The technical scheme of the present application discards the traditional data modeling method based on a relational database, constructs a business graph special for identifying the behaviors of "single splitting" and "circular insurance" based on a "graph model" in a graph database, efficiently identifies the behaviors of single splitting and circular insurance through "subgraph identification" by means of ingenious data modeling and the advantages of the graph database in graph data analysis and mining, reduces the identification difficulty, improves the identification efficiency, and improves the accuracy and comprehensiveness of the identification result.

[0149] In the technical scheme of the present application, the customer nodes, the insurance policy nodes, the sales agent nodes included in each subgraph involved in the first risk business, and the edges between the customer nodes, the insurance policy nodes and the sales agent nodes involved in the first risk business are marked; the customer nodes, the insurance policy nodes included in each subgraph involved in the second risk business, and the edges between the customer nodes and the insurance policy nodes involved in the second risk business are marked, and the returned result of identification is still graph data (subgraph), which is easy to visualize and can fully decompose the relevant parties of the behaviors of "single splitting" and "circular insurance".

[0150] In the technical scheme of the present application, the more the number of sales agents included in each subgraph involved in the first risk business behavior, or the more the number of insurance policy nodes corresponding to each sales agent among the sales agents included in each subgraph involved in the first risk business, the higher the risk level of the first risk business behavior corresponding to the subgraph; the more the number of insurance policy nodes corresponding to the same customer node included in each subgraph involved in the second risk business behavior, or the more the number of insurance policy nodes existing the second edge with the same customer node included in each subgraph involved in the second risk business, the higher the risk level of the second risk business behavior, which can quickly identify subgraphs with higher risk, reduce company losses and fraud risks, improve business quality, help life insurance companies efficiently identify the behaviors of "single splitting" and "circular insurance", fully "understand" various representative behavior patterns involved in the behaviors of "single splitting" of sales agents and "circular insurance" of customers, focus on sales agents and customers who frequently produce the above risk business behaviors, and give risk prompts and warnings to guide the business behaviors to be more standardized and help the company to develop with high quality.

[0151] Embodiment two

[0152] As shown in Figure 12 , the technical scheme of the present application further provides a life insurance company risk business behavior identification system, which comprises:

[0153] The establishing module 101 takes the insurance policy, the customer and the sales agent as nodes, and takes the insurance relationship, the sales relationship and the loan relationship among the insurance policy, the customer and the sales agent as edges to establish a risk business behavior identification business graph; wherein, the risk business includes a first risk business and / or a second risk business, the first risk business is a policy disassembly behavior, and the second risk business is a circular insurance behavior.

[0154] The first identification module 102 identifies the first risk business according to the insurance policy node, the sales agent node and the customer node included in the subgraph of the risk business behavior identification business graph and the edges among the insurance policy node, the sales agent node and the customer node.

[0155] The second identification module 103 identifies the second risk business according to the insurance policy node and the customer node included in the subgraph of the risk business behavior identification business graph and the edges between the insurance policy node and the customer node.

[0156] The implementation processes of the monitoring and establishing module 101, the first identification module 102 and the second identification module 103 in the second embodiment correspond to the method steps in the first embodiment, and the second embodiment will not be repeated here.

[0157] In the technical scheme of the application, the insurance policy, the customer and the sales agent are taken as nodes, and the insurance relationship, the sales relationship and the loan relationship among the insurance policy, the customer and the sales agent are taken as edges to establish a risk business behavior identification business graph; the first risk business behavior is identified according to the insurance policy node, the sales agent node and the customer node included in the subgraph of the risk business behavior identification business graph and the edges among the insurance policy node, the sales agent node and the customer node; and the second risk business behavior is identified according to the insurance policy node and the customer node included in the subgraph of the risk business behavior identification business graph and the edges between the insurance policy node and the customer node, thereby effectively solving the problem of low identification efficiency of the policy disassembly behavior and / or the circular insurance behavior caused by the prior art and effectively improving the identification efficiency of the policy disassembly behavior and / or the circular insurance behavior.

[0158] The technical scheme of the application discards the traditional data modeling method based on a relational database, constructs a special business graph for policy disassembly and circular insurance behavior identification in a graph database based on a "graph model", efficiently identifies the policy disassembly and circular insurance behavior through "subgraph identification" by means of ingenious data modeling and the advantages of the graph database in graph-based data analysis and mining, reduces the identification difficulty, improves the identification efficiency, and improves the accuracy and comprehensiveness of the identification result.

[0159] The customer nodes, the policy nodes, the sales agent nodes included in each subgraph involved in the first risk business and the edges between the customer nodes, the policy nodes and the sales agent nodes involved in the first risk business are marked; the customer nodes, the policy nodes included in each subgraph involved in the second risk business and the edges between the customer nodes and the policy nodes involved in the second risk business are marked, and the return result is still graph data (subgraph), which is easy to visualize and can fully decompose the related parties of the "single splitting" and the circular insurance behavior.

[0160] The more the number of sales agents included in each subgraph involved in the first risk business behavior is, or the more the number of policy nodes corresponding to each sales agent in the sales agents included in each subgraph involved in the first risk business is, the higher the risk level of the first risk business behavior corresponding to the subgraph is; the more the number of policy nodes corresponding to the same customer node included in each subgraph involved in the second risk business is, or the more the number of policy nodes having the second edge with the same customer node included in each subgraph involved in the second risk business is, the higher the risk level of the second risk business behavior is, the risk of the subgraph with higher risk can be quickly identified, the company loss and the fraud risk are reduced, the business quality is improved, the "single splitting" and the "circular insurance" behavior of the life insurance company can be efficiently identified, the representative behavior modes involved in the "single splitting behavior" of the sales agent and the "circular insurance behavior" of the customer are fully "understood", the sales agent and the customer frequently producing the above risk business behaviors are focused on, the risk prompt and the early warning are performed, the business behavior is guided to be more standardized, and the high-quality development of the company is assisted.

[0161] Although the specific embodiments of the present application are described above with reference to the accompanying drawings, the present application is not limited to the above embodiments, and various modifications or changes can be made by those skilled in the art without creative labor on the basis of the technical solutions of the present application.

Claims

1. A method of identifying risk business behavior of a life insurance company, characterized by, Comprise: A risk business behavior identification business graph is established, taking a policy, a customer, and a sales agent as a node, and taking a policy, a customer, a sales agent, a relationship between a sales agent and a customer, a relationship between a sales agent and a policy, and a relationship between a customer and a policy as an edge; wherein the risk business behavior comprises a first risk business behavior and / or a second risk business behavior, the first risk business behavior is a policy disassembly behavior, and the second risk business behavior is a circular insurance behavior; the customer node and the policy node are connected by a first edge and a second edge, the first edge is from the customer node to the policy node, representing a one-time insurance relationship, and the second edge is from the customer node to the policy node, representing a one-time policy loan relationship; the policy node comprises policy type attribute information; the first edge is marked as an insurance type, and the second edge is marked as a loan type; the first edge comprises first attribute information, and the first attribute information is an insurance time; the second edge comprises fourth attribute information and fifth attribute information, the fourth attribute information is a policy loan time, and the fifth attribute information is a policy loan repayment time; the circular insurance behavior is that a customer insures, then makes a policy loan for the policy in a short period of time, and then insures the same type of policy again, and the loan is not repaid at the time of the insurance; According to the policy node, the sales agent node, and the customer node included in the subgraph of the risk business behavior identification business graph, and the edges between the policy node, the sales agent node, and the customer node, the first risk business behavior is identified; and / or, According to the policy node and the customer node included in the subgraph of the risk business behavior identification business graph, and the edges between the policy node and the customer node, the second risk business behavior is identified; specifically comprising: If the policy node, the customer node, and the edges between the policy node and the customer node in the risk business behavior identification business graph satisfy a preset second risk identification rule, the insurance behavior involved in the subgraph constitutes a second risk business behavior; wherein the second risk identification rule comprises: The same customer node corresponds to at least two policy nodes, and at least one policy node in the first type of policy node is connected to the customer node by a second edge, wherein the first type of policy node is at least two policy nodes connected to the same customer node; and The first attribute information of the first type of first edge is less than the fourth attribute information of the first type of second edge, wherein the first type of first edge is the first edge between the same customer node and the policy node connected to the customer node by the second edge; the first type of second edge is the second edge between the same customer node and the policy node connected to the customer node by the second edge; and The fourth attribute information of the first type of second edge is less than the first attribute information of the second type of first edge, wherein the second type of first edge is the first edge between the same customer node and the policy node not connected to the customer node by the second edge; and The fifth attribute information of the first type of second edge is greater than the first attribute information of the second type of first edge; and The policy type attribute information of the first type of policy node is the same.

2. The method for identifying risky business practices of a life insurance company according to claim 1, characterized in that, Each customer node in the risk business behavior identification business graph represents a customer; each policy node represents a policy; and each sales agent node represents a sales agent. The third edge exists between the policy node and the sales agent node; the third edge is directed from the sales agent node to the policy node, representing a policy sales relationship.

3. The method for identifying risky business practices of a life insurance company according to claim 2, characterized in that, The third edge is marked as a sales type. The first edge further includes second attribute information, and the second attribute information is a policy taking effect time.

4. The method for identifying risky business practices of a life insurance company according to claim 3, characterized in that, According to the risk business behavior, the policy node, the sales agent node, and the customer node included in the subgraph of the business graph, and the edge between the policy node, the sales agent node, and the customer node, the first risk business behavior specifically includes: If the policy node, the sales agent node, and the customer node included in a subgraph of the risk business graph, and the edge between the policy node, the sales agent node, and the customer node satisfy a preset first risk identification rule, the insurance behavior involved in the subgraph constitutes a first risk business behavior.

5. The method of claim 4, wherein the step of identifying the risk business behavior of the life insurance company is characterized by, The first risk identification rule includes: The same customer node is connected to at least two policy nodes; and The first type of policy node is connected to at least two different sales agent nodes, wherein the first type of policy node is at least two policy nodes connected to the same customer node; and The first attribute information or the second attribute information of the first edge between the same customer node and the first type of policy node is the same in the year and month; and The policy type attribute information of the first type of policy node is the same.

6. The method of claim 1, wherein the method further comprises: The more the number of sales agents included in each subgraph involved in the first risk business behavior, or the more the number of policy nodes corresponding to each sales agent in the sales agents included in the first risk business, the higher the risk level of the first risk business behavior corresponding to the subgraph. ​ The more the number of policy nodes corresponding to the same customer node included in each subgraph involved in the second risk business behavior, or the more the number of policy nodes connected to the same customer node by the second edge included in each subgraph involved in the second risk business, the higher the risk level of the second risk business behavior.

7. The method of claim 1-6, further comprising: It includes: Marking the customer node, the policy node, the sales agent node included in each subgraph involved in the first risk business behavior, and the edge between the customer node, the policy node, and the sales agent node involved in the first risk business; and / or Marking the customer node, the policy node included in each subgraph involved in the second risk business behavior, and the edge between the customer node and the policy node involved in the second risk business.

8. A risk business behavior identification system of a life insurance company, characterized by comprising: The establishment module establishes a risk business behavior identification business graph taking a policy, a customer, and a sales agent as nodes and taking a policy, a customer, a sales agent, an insurance relationship, a sales relationship, and a loan relationship therebetween as edges; wherein, the risk business includes a first risk business and / or a second risk business, the first risk business is a policy disassembly behavior, and the second risk business is a circular insurance behavior; the customer node and the policy node have a first edge and a second edge, the first edge is from the customer node to the policy node and represents an insurance relationship, and the second edge is from the customer node to the policy node and represents a policy loan relationship; the policy node includes policy type attribute information; the first edge is marked as an insurance type, and the second edge is marked as a loan type; the first edge includes first attribute information, and the first attribute information is an insurance time; the second edge includes fourth attribute information and fifth attribute information, the fourth attribute information is a policy loan time, and the fifth attribute information is a repayment time of the policy loan; the circular insurance behavior is that a customer insures, then makes a policy loan for the policy in a short period of time, and then insures the same policy type again, and the policy loan is not repaid at the time of the insurance; a first identification module identifies the first risk business according to the policy nodes, the sales agent nodes, and the customer nodes included in a subgraph of the risk business behavior identification business graph and the edges between the policy nodes, the sales agent nodes, and the customer nodes; and / or a second identification module identifies the second risk business according to the policy nodes and the customer nodes included in a subgraph of the risk business behavior identification business graph and the edges between the policy nodes and the customer nodes; specifically including: if the policy nodes, the customer nodes, and the edges between the policy nodes and the customer nodes in the risk business behavior identification business graph satisfy a preset second risk identification rule, then an insurance behavior involved in the subgraph constitutes a second risk business behavior; wherein, the second risk identification rule includes: a same customer node corresponds to at least two policy nodes, and at least one policy node in a first type of policy nodes has a second edge with the customer node, wherein the first type of policy nodes are at least two policy nodes connected to a same customer node; and first attribute information of a first type of first edges is less than fourth attribute information of a first type of second edges, wherein the first type of first edges are the first edges between a same customer node and policy nodes having the second edges with the customer node, the first type of second edges are the second edges between the same customer node and the policy nodes having the second edges with the customer node; and the fourth attribute information of the first type of second edges is less than first attribute information of a second type of first edges, wherein the second type of first edges are the first edges between a same customer node and policy nodes not having the second edges with the customer node; and the fifth attribute information of the first type of second edges is greater than the first attribute information of the second type of first edges; and policy type attribute information of the first type of policy nodes is the same.

Images