Electronic Card Risk Early Warning Method and Device Based on Knowledge Graph

Abstract

An embodiment of the present application provides a method and device for electronic card risk early warning based on a knowledge graph, which can be used in the field of artificial intelligence. The method includes: obtaining electronic card application information, and constructing an initial state graph according to the electronic card application information and the association relationship between the electronic card application information; performing community cutting on the initial state graph according to a preset modularity algorithm model, and determining the financial risk level of each community according to the matching degree between each community obtained after the cutting and a preset gang fraud rule model, and performing a risk early warning operation on the community whose financial risk level exceeds the risk threshold; the present application can accurately prevent and control the risk of electronic card gang fraud behavior.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence and can also be used for financial risk control, and specifically to an electronic card risk warning method and device based on a knowledge graph. Background Art

[0002] As fraud methods are upgraded and high-tech trends become more evident, the gang nature of fraud methods is gradually revealed, which can easily form a black industry chain with a complete upstream, midstream and downstream structure. Especially with the increase in online and electronic channels, various methods of committing crimes are obvious, such as using packaged application materials, forging false mailing addresses and cross-regional migration.

[0003] Taking credit card applications as an example, the current number of online transactions is huge. When using relational databases, existing methods make it difficult to discover the internal connections between various types of card applications. The amount of calculation is large and the work efficiency is low. It is not easy to discover and explore the risk of gang fraud, and thus it is impossible to prevent and control the risk of fraud behavior of users in advance. Summary of the invention

[0004] In response to the problems in the prior art, the present application provides an electronic card risk warning method and device based on a knowledge graph, which can accurately prevent and control the risks of electronic card gang fraud.

[0005] In order to solve at least one of the above problems, the present application provides the following technical solutions:

[0006] In a first aspect, the present application provides an electronic card risk warning method based on a knowledge graph, comprising:

[0007] Acquire electronic card application information, and construct an initial state map according to the electronic card application information and the association relationship between each piece of electronic card application information;

[0008] The initial state map is divided into communities according to a preset modularity algorithm model, and the financial risk level of each community is determined according to the degree of matching between each community obtained after the division and a preset gang fraud rule model, and risk warning operations are performed on communities whose financial risk levels exceed a risk threshold.

[0009] Furthermore, the constructing of an initial state graph according to the electronic card application information and the association relationship between each electronic card application information includes:

[0010] According to the preset graph association relationship building component, the electronic card application information is subjected to a strong association relationship precise matching operation and a weak association relationship matching degree calculation operation, and the intermediate node of the initial state graph is determined according to the results of the precise matching operation and the matching degree calculation operation;

[0011] Determine the relationship edges of the initial state graph according to the intermediate node and the corresponding application node, where the electronic card application information is the application node of the initial state graph;

[0012] Construct the initial state graph according to the application node, the intermediate node, and the relationship edge.

[0013] Further, the operation of calculating the weak association relationship matching degree of the electronic card application information by the component constructed according to the preset graph association relationship includes:

[0014] Perform address fuzzy matching operation on the address type information in the electronic card application information according to the preset standard address database to obtain the weak association relationship matching degree of the address type information;

[0015] Perform pairwise matching operation on the non-address type information in the electronic card application information to obtain the weak association relationship matching degree of the non-address type information.

[0016] Further, after the initial state graph is cut by the preset modularity algorithm model, it includes:

[0017] Determine the initial modularity of the initial state graph according to the preset modularity algorithm model;

[0018] Allocate each node in the initial state graph to the communities of adjacent nodes in turn until the change value of modularity before and after allocation is the largest, to obtain each community of the initial state graph.

[0019] Further, determining the risk level of each community according to the matching degree between each community obtained after the cutting and the preset gang fraud rule model includes:

[0020] Obtain the preset gang fraud rule model, judge the number of rule hits and the content of rule hits of the preset gang fraud rule model in each community obtained after the cutting, and determine the risk level of each community according to the number of rule hits and the content of rule hits.

[0021] Further, determining the relationship edges of the initial state graph according to the intermediate node and the corresponding application node includes:

[0022] Establish a relationship edge between the application node with exactly the same electronic card application information and the intermediate node during the precise matching operation of the strong association relationship, and establish a relationship edge between the application node with partially the same electronic card application information and the intermediate node during the weak association relationship matching degree operation.

[0023] Second aspect, the present application provides an electronic card risk early warning device based on a knowledge graph, including:

[0024] An initial state graph construction module, configured to obtain electronic card application information, and construct an initial state graph according to the electronic card application information and the association relationships between the electronic card application information;

[0025] A community risk early warning module, configured to perform community cutting on the initial state graph according to a preset modularity algorithm model, determine the risk level of each community according to the matching degree between each community obtained after the cutting and a preset gang fraud rule model, and perform a risk early warning operation on the community whose risk level exceeds a risk threshold.

[0026] Further, the initial state graph construction module includes:

[0027] An intermediate node determination unit, configured to perform a precise matching operation for strong association relationships and a matching degree calculation operation for weak association relationships on the electronic card application information according to a preset graph association relationship construction component, and determine the intermediate node of the initial state graph according to the results of the precise matching operation and the matching degree calculation operation;

[0028] A relationship edge determination unit, configured to determine the relationship edge of the initial state graph according to the intermediate node and the corresponding application node, where the electronic card application information is the application node of the initial state graph;

[0029] A graph construction unit, configured to construct the initial state graph according to the application node, the intermediate node, and the relationship edge.

[0030] Further, the intermediate node determination unit includes:

[0031] An address type weak association matching subunit, configured to perform an address fuzzy matching operation on the address type information in the electronic card application information according to a preset standard address database to obtain an address type information weak association relationship matching degree;

[0032] A non-address type weak association matching subunit, configured to perform a pairwise matching operation on the non-address type information in the electronic card application information to obtain a non-address type information weak association relationship matching degree.

[0033] Further, the community risk early warning module further includes:

[0034] An initial modularity determination unit, configured to determine the initial modularity of the initial state graph according to a preset modularity algorithm model;

[0035] A community segmentation unit is used to sequentially assign each node in the initial state graph to the communities of adjacent nodes until the modularity change value before and after the assignment is the largest, thereby obtaining each community of the initial state graph.

[0036] In a third aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the electronic card risk warning method based on the knowledge graph are implemented.

[0037] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the electronic card risk warning method based on the knowledge graph are implemented.

[0038] As can be seen from the above technical solutions, the present application provides an electronic card risk warning method and device based on a knowledge graph. By using the electronic card application information and the association relationships between the electronic card application information, an initial state graph is constructed. Then, the modularity algorithm model is used to perform community segmentation on the generated initial state graph until the best graphing method is obtained. Next, the gang fraud rule model is combined to evaluate the risk levels of the constructed communities as the basis for gang fraud risk warning, so as to accurately prevent and control the risk of electronic card gang fraud behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] 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 required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0040] Figure 1 It is a schematic flowchart of the electronic card risk warning method based on the knowledge graph in an embodiment of the present application;

[0041] Figure 2 It is a schematic flowchart of the electronic card risk warning method based on the knowledge graph in an embodiment of the present application;

[0042] Figure 3 It is a schematic flowchart of the electronic card risk warning method based on the knowledge graph in an embodiment of the present application;

[0043] Figure 4 It is a schematic flowchart of the electronic card risk warning method based on the knowledge graph in an embodiment of the present application;

[0044] Figure 5One of the structural diagrams of the electronic card risk warning device based on the knowledge graph in the embodiments of the present application;

[0045] Figure 6 Another structural diagram of the electronic card risk warning device based on the knowledge graph in the embodiments of the present application;

[0046] Figure 7 Another structural diagram of the electronic card risk warning device based on the knowledge graph in the embodiments of the present application;

[0047] Figure 8 Another structural diagram of the electronic card risk warning device based on the knowledge graph in the embodiments of the present application;

[0048] Figure 9 Schematic diagram of the initial state graph in an embodiment of the present application;

[0049] Figure 10 Schematic diagram of the node relationship in an embodiment of the present application;

[0050] Figure 11 Schematic diagram of the structure of the electronic device in the embodiments of the present application. Detailed implementation manners

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are some but not all of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0052] Considering that in the prior art, the current number of online transactions is huge. In the case of using a relational database, it is difficult to discover the internal connections between various card application documents by existing means, the calculation amount is large and the work efficiency is low, it is not easy to discover and dig out the risk of gang fraud, and thus it is impossible to prevent and control the risk of fraud in user behavior in advance. The present application provides an electronic card risk warning method and device based on a knowledge graph. An initial state graph is constructed through the electronic card application document information and the association relationships between the electronic card application document information. Then, the modularity algorithm model is used to perform community cutting on the generated initial state graph until the best graphing method is obtained. Then, in combination with the gang fraud rule model, risk level assessment is carried out for each constructed community as the basis for gang fraud risk warning, so as to accurately prevent and control the risk of electronic card gang fraud behavior.

[0053] It is understandable that the electronic card is a physical card / virtual card containing electronic information, such as a credit card or a debit card containing financial information or account information, or a virtual membership card or consumption card containing financial information or account information.

[0054] In order to accurately prevent and control the risk of electronic card gang fraud, this application provides an embodiment of an electronic card risk warning method based on a knowledge graph. Refer to Figure 1 The electronic card risk warning method based on the knowledge graph specifically includes the following content:

[0055] Step S101: Obtain electronic card application information, and construct an initial state graph according to the electronic card application information and the association relationship between the electronic card application information.

[0056] Optionally, this application can obtain relevant data such as applications and transactions through a financial transaction system, including user personal application information and operation behavior information. Among them, the user personal application information includes, but is not limited to: ID number, mobile phone number, contact mobile phone number, unit telephone number, contact name, unit name, unit address, residential address. The operation behavior information includes, but is not limited to: login IP address, MAC address. This application can generate an application ID for each financial transaction obtained, mark each transaction node with this, and use it as a node in the initial state graph (i.e., the knowledge graph) (i.e., the method of generating an ID).

[0057] In another embodiment of this application, after obtaining the electronic card application information, this application can also use spark technology to preprocess the data, clean the data such as the application address and mobile phone number in the obtained transaction information, and remove the information data that is empty or does not conform to the specified format.

[0058] Optionally, this application can find the relationship between various types of information between each application according to the obtained electronic card application information, specifically including two categories: strong association relationships and weak association relationships, so as to construct an initial state graph. Among them, the strong association relationship means that the nodes are exactly the same, such as the ID number. The weak association relationship means that the node information is partially similar, for example, the mobile phone number segments are the same, and the address similarity is greater than the threshold, etc.

[0059] Step S102: Perform community cutting on the initial state graph according to a preset modularity algorithm model, determine the risk level of each community according to the matching degree between each community obtained after the cutting and a preset gang fraud rule model, and perform a risk warning operation on the community whose risk level exceeds the risk threshold.

[0060] Optionally, this application can perform community cutting on the initial state graph according to a preset modularity algorithm model. The preset modularity algorithm model can be, for example, the existing Louvain algorithm model (the Louvain algorithm takes modularity as the optimization goal, and the larger the value, the better the partitioning effect). Thus, a large community graph is cut into several small communities, thereby discovering small gang communities.

[0061] Optionally, based on the matching degree between the divided communities and the preset gang fraud rule model, determine the risk level of each community, and perform a risk warning operation on the community whose risk level exceeds the risk threshold.

[0062] Specifically, first obtain the gang fraud rules, locate each application and relationship graph in the community through the community number, and use the association relationship as the rule for forming the community relationship network, such as the applicant identity information in the network being the same. Store the rules obtained in each community, store them according to the community number, and record the number of rules obtained in each community.

[0063] At the same time, in addition to obtaining fraud rules through community information, this application can also provide a manual rule maintenance interface for users. Based on the gang frauds that have occurred, incorporate the accumulated fraud rules into the recognition engine process. When assigning weights to the rules, according to whether the rule has been involved in the occurred gang fraud, if it has been involved once, the proportion is increased by 10.

[0064] Finally, using the community number as the primary key, search for all rules within the community, including the obtained rules and the manually maintained rules, and record the community number, the number of application nodes, the number of rule hits, and the content of the hit rules. Set the community risk levels to high, medium, and low risk categories. When the community hit rules include manually maintained rules, that is, when the rules violated in the occurred gang fraud are involved, it is listed as a high risk level. When the community hit rules are the obtained community rules and there are multiple of them, it is listed as a medium risk level. When the community hit rule is only one of the obtained community rules, it is listed as a low risk level. Finally, based on the obtained community information of different risk categories, perform a gang fraud risk warning.

[0065] As can be seen from the above description, the electronic card risk warning method based on a knowledge graph provided by the embodiments of this application can construct an initial state graph through the electronic card application information and the association relationship between the electronic card application information, and then use the modularity algorithm model to perform community cutting on the generated initial state graph until the best graph construction method is obtained. Then, combined with the gang fraud rule model, the risk levels of the constructed communities are evaluated as the basis for gang fraud risk warning, so as to accurately prevent and control the risk of electronic card gang fraud.

[0066] In order to accurately construct the initial state graph of the electronic card application information, in an embodiment of the electronic card risk warning method based on the knowledge graph in the present application, refer to Figure 2 The above step S101 may specifically include the following content:

[0067] Step S201: Perform an accurate matching operation of strong association relationships and a matching degree calculation operation of weak association relationships on the electronic card application information by components according to the preset graph association relationships, and determine the intermediate nodes of the initial state graph according to the results of the accurate matching operation and the matching degree calculation operation.

[0068] Step S202: Determine the relationship edges of the initial state graph according to the intermediate nodes and the corresponding application nodes, where the electronic card application information is the application node of the initial state graph.

[0069] Step S203: Construct the initial state graph according to the application nodes, the intermediate nodes, and the relationship edges.

[0070] Specifically, first, set the association relationships for generating the graph, that is, set an independent association relationship component for generating the graph, which is used to help the user define the basis for graph construction. This component supports setting one or more association elements, and the association elements are application information in the application, including ID card number, name, mobile phone number, unit phone number, mailing address, IP address of the application device, MAC address of the application device, etc.

[0071] Then, perform the strong association relationship operation, that is, perform the strong association relationship operation after setting the association relationships for generating the graph, and perform an accurate matching of the set relationship elements. When the association elements are completely consistent, record the element information and the application ID, assign a matching degree of 1. After completing the strong association relationship operation, enter the weak association relationship operation, find the weak association relationship, and obtain the operation relationship matching degree.

[0072] Finally, construct the initial state graph based on the relationships, that is, according to the strong relationship and weak relationship results obtained from the above operations, convert the various types of information that match into intermediate nodes, and establish a relationship edge between the application node and the intermediate node, refer to Figure 9 Specifically:

[0073] (1) When performing the accurate matching operation of the strong association relationship, establish a relationship edge between the application node with completely consistent electronic card application information and the intermediate node. That is, in the case of strong association, convert the relevant element information into intermediate information, and establish a strong association relationship edge between the application ID with consistent information and the intermediate information, and assign a weight of 1 to the strong association relationship edge.

[0074] (2) When performing the weak association relationship matching degree operation, establish a relationship edge between the application node with partially consistent electronic card application information and the intermediary node. That is, when there is a weak association, convert relevant information such as mobile phone number segments, area codes, the first 4 digits of landline numbers, and the first 3 groups of IP addresses into corresponding intermediary information, and establish a weak association relationship edge between the application and the various types of converted intermediary information. Assign the weight of the weak association relationship edge as the relationship matching degree calculated in the third step. In this way, a knowledge graph in the initial state is established.

[0075] Among them, for fixed - line phone - related information, it is defined that when the area code and the first four digits of the phone number are the same, it is considered that a weak association relationship edge is established between the two matching application IDs through the fixed - line phone; when the mobile phone number segments are the same, it is considered that a weak association relationship edge is established between the two matching application IDs; for IP - address - related information, it is defined that when the first three segments of the address are the same, it is considered that a weak association relationship edge is established between the two matching application IDs through the IP address.

[0076] In order to be able to perform the matching degree operation of the weak association relationship on the electronic card application information when constructing the knowledge graph in the initial state, in an embodiment of the electronic card risk warning method based on the knowledge graph of the present application, refer to Figure 3 , the above - mentioned step S201 may specifically include the following content:

[0077] Step S301: Perform an address fuzzy matching operation on the address - related information in the electronic card application information according to a preset standard address database to obtain the weak association relationship matching degree of the address - related information.

[0078] Step S302: Perform a pairwise matching operation on the non - address - related information in the electronic card application information to obtain the weak association relationship matching degree of the non - address - related information.

[0079] In some embodiments of the present application, when performing the weak association relationship matching degree operation, it may specifically include:

[0080] (1) Perform an address fuzzy matching operation on the address - related information. Establish a standard address library, which contains various aliases of addresses. Use the standard address library to compare the application address with the standard address, and then return the result and the matching degree. Here, the cosine comparison algorithm is used to calculate the matching degree. When some parts are omitted in the application address, the standard address library will complete the application address to the provincial, municipal, and district / county levels. When performing the address fuzzy matching operation, a threshold (between 0 and 1) needs to be set. When the matching degree is greater than the threshold, convert the compared standard address into an address ID, and establish a weak association relationship edge between the application and the address ID. Here, the relationship matching degree between the two is the matching degree calculated by the cosine comparison algorithm.

[0081] (2) Perform pairwise matching operations on non-address information. Compare information such as mobile phone numbers, IP addresses, and company phone numbers pairwise. When the mobile phone number segments, area codes, the first 4 digits of company phone numbers, and the first 3 groups of IP address information are the same, a weak association relationship is established. The matching degree between the two here is the number of identical digits / the total number of digits.

[0082] In order to accurately determine each community in the initial state graph, in an embodiment of the electronic card risk warning method based on a knowledge graph in the present application, refer to Figure 4 , the above step S102 may further specifically include the following content:

[0083] Step S401: Determine the initial modularity of the initial state graph according to a preset modularity algorithm model.

[0084] Step S402: Allocate each node in the initial state graph to the communities of its adjacent nodes in turn until the change value of modularity before and after allocation is the largest, and obtain each community of the initial state graph.

[0085] Specifically, first perform initialization. Regard each point in the initial state graph as an independent node. The initial state graph regards each node as an independent community and calculates the modularity of the initial state. Here, the modularity is:

[0086]

[0087] Among them, A i,j represents the weight between node i and node j. Here, the matching degree obtained in the process of constructing the graph is adopted. m identifies the sum of the weights of all edges in the graph. k i represents the sum of the weights of the edges connected to point i. Here, it becomes the degree. k i = ∑ j A i,j . c i identifies the community to which the point is assigned. δ(c i , c j ) is used to determine whether point i and point j are divided into the same community. If so, it is 1; otherwise, it returns 0.

[0088] Then, divide the communities of the nodes, that is, try to assign each node to the communities where its each neighbor node is located in turn, and calculate the change in modularity ΔQ before and after the assignment.

[0089] Then, determine the node assignment, that is, select the partitioning method with ΔQ>0 and taking the maximum value as the final partitioning method. If maxΔQ<0, then keep it unchanged.

[0090] For example: The relationship between nodes X, Y, and Z is as Figure 10As shown, the calculation of each variable is as follows:

[0091] A x,y = A y,x = 1 A x,z = A z,x = 1 A z,y = A y,z = 0

[0092] A x,x = 0 A y,y = 0 A z,z = 0

[0093]

[0094] K x = 2 K y = 1 K z = 1

[0095] It can be seen from this that in the initial state, the three points are each a community, δ(c x , c x ) = δ(c y , c y ) = δc z , c z ) = 1. When i and j take other values, δ(c i , c j ) = 0, and the modularity Q = -3 / 8.

[0096] Among them, for point Y, there are two partitioning methods: X and Y in one community, and Y and Z in one community. The modularity Q values calculated for the two partitioning methods are -1 / 8 and -1 / 2 respectively, and the corresponding modularity changes ΔQ are 1 / 4 and -1 / 8 respectively. Therefore, X and Y are partitioned into one community.

[0097] Finally, set the maximum number N of points included in the community, and continuously run the above algorithm to ensure that all nodes in the community do not exceed N. Each community generates a globally unique number as the community number. Record the community number, the application information in the community, and the association relationships in the application information.

[0098] In order to accurately prevent and control the risks of electronic card gang fraud, this application provides an embodiment of an electronic card risk warning device based on a knowledge graph for implementing all or part of the content of the above-mentioned electronic card risk warning method based on a knowledge graph. See Figure 5 , the electronic card risk warning device based on a knowledge graph specifically includes the following content:

[0099] The initial state graph construction module 10 is used to obtain electronic card application information, and construct an initial state graph according to the electronic card application information and the association relationships among the electronic card application information.

[0100] The community risk warning module 20 is used to perform community cutting on the initial state graph according to a preset modularity algorithm model, determine the risk levels of the communities according to the matching degrees of the communities obtained after the cutting with a preset gang fraud rule model, and perform risk warning operations on the communities whose risk levels exceed the risk threshold.

[0101] As can be seen from the above description, the electronic card risk warning device based on a knowledge graph provided by the embodiments of the present application can construct an initial state graph through the electronic card application information and the association relationships among the electronic card application information, and then use the modularity algorithm model to perform community cutting on the generated initial state graph until the best graph composition method is obtained, and then combine the gang fraud rule model to evaluate the risk levels of the constructed communities as the basis for gang fraud risk warning, so as to accurately prevent and control the risk of electronic card gang fraud behavior.

[0102] In order to accurately construct the initial state graph of the electronic card application information, in an embodiment of the electronic card risk warning device based on a knowledge graph of the present application, see Figure 6 , the initial state graph construction module 10 includes:

[0103] The intermediate node determination unit 11 is used to perform precise matching operations for strong association relationships and matching degree calculation operations for weak association relationships on the electronic card application information according to a preset graph association relationship construction component, and determine the intermediate nodes of the initial state graph according to the results of the precise matching operations and the matching degree calculation operations.

[0104] The relationship edge determination unit 12 is used to determine the relationship edges of the initial state graph according to the intermediate nodes and the corresponding application nodes, where the electronic card application information is the application nodes of the initial state graph.

[0105] The graph construction unit 13 is used to construct the initial state graph according to the application nodes, the intermediate nodes, and the relationship edges.

[0106] In order to be able to perform matching degree calculation for weak association relationships on the electronic card application information when constructing the initial state graph, in an embodiment of the electronic card risk warning device based on a knowledge graph of the present application, see Figure 7 , the intermediate node determination unit 11 includes:

[0107] The address - type weak - association matching subunit 111 is used to perform address fuzzy matching operations on the address - type information in the electronic card application information according to a preset standard address database, and obtain the weak - association relationship matching degree of the address - type information.

[0108] The non - address - type weak - association matching subunit 112 is used to perform pairwise matching operations on the non - address - type information in the electronic card application information, and obtain the weak - association relationship matching degree of the non - address - type information.

[0109] In order to accurately determine each community in the initial state graph, in an embodiment of the electronic card risk early - warning device based on a knowledge graph in this application, refer to Figure 8 , the community risk early - warning module 20 further includes:

[0110] The initial modularity determination unit 21 is used to determine the initial modularity of the initial state graph according to a preset modularity algorithm model.

[0111] The community segmentation unit 22 is used to sequentially assign each node in the initial state graph to the communities of its adjacent nodes until the change value of the modularity before and after the assignment is the largest, and obtain each community of the initial state graph.

[0112] From a hardware level, in order to accurately prevent and control the risk of electronic card gang fraud, this application provides an embodiment of an electronic device for implementing all or part of the content in the above - mentioned electronic card risk early - warning method based on a knowledge graph. The electronic device specifically includes the following:

[0113] A processor, a memory, a communication interface, and a bus; wherein, the processor, the memory, and the communication interface complete communication with each other through the bus; the communication interface is used to realize information transmission between the electronic card risk early - warning device based on a knowledge graph and related devices such as a core business system, a user terminal, and a related database, etc. This logic controller can be a desktop computer, a tablet computer, a mobile terminal, etc., and this embodiment is not limited thereto. In this embodiment, this logic controller can be implemented with reference to the embodiments of the electronic card risk early - warning method based on a knowledge graph and the embodiments of the electronic card risk early - warning device based on a knowledge graph, and its content is incorporated herein, and the repeated parts will not be elaborated.

[0114] It can be understood that the user terminal may include a smart phone, a tablet electronic device, a network set - top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle - mounted device, a smart wearable device, etc. Among them, the smart wearable device may include smart glasses, a smart watch, a smart bracelet, etc.

[0115] In practical applications, part of the electronic card risk warning method based on the knowledge graph can be executed on the side of the electronic device as described above, or all operations can be completed in the client device. Specifically, it can be selected according to the processing capacity of the client device and the limitations of the user usage scenario, etc. This application does not make any limitations in this regard. If all operations are completed in the client device, the client device may further include a processor.

[0116] The above-mentioned client device may have a communication module (i.e., a communication unit), which can be communicatively connected to a remote server to achieve data transmission with the server. The server may include a server on the side of the task scheduling center, and in other implementation scenarios, it may also include a server of an intermediate platform, such as a server of a third-party server platform having a communication link with the task scheduling center server. The server may include a single computer device, or may include a server cluster composed of multiple servers, or a server structure of a distributed device.

[0117] Figure 11 It is a schematic block diagram of the system composition of the electronic device 9600 according to an embodiment of the present application. As Figure 11 shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It should be noted that this Figure 11 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0118] In one embodiment, the function of the electronic card risk warning method based on the knowledge graph can be integrated into the central processing unit 9100. Among them, the central processing unit 9100 may be configured to perform the following controls:

[0119] Step S101: Obtain electronic card application information, and construct an initial state graph according to the electronic card application information and the association relationship between the electronic card application information.

[0120] Step S102: Perform community cutting on the initial state graph according to a preset modularity algorithm model, and determine the risk level of each community according to the matching degree between each community obtained after the cutting and a preset gang fraud rule model, and perform a risk warning operation on the community whose risk level exceeds the risk threshold.

[0121] As can be seen from the above description, the electronic device provided by the embodiments of the present application constructs an initial state graph through the electronic card application information and the association relationships between the electronic card application information. Then, the modularity algorithm model is used to perform community cutting on the generated initial state graph until the best composition method is obtained. Then, in combination with the gang fraud rule model, risk levels are evaluated for each constructed community, serving as the basis for gang fraud risk early warning, so as to accurately prevent and control the risk of electronic card gang fraud behavior.

[0122] In another embodiment, the electronic card risk early warning device based on the knowledge graph can be separately configured from the central processing unit 9100. For example, the electronic card risk early warning device based on the knowledge graph can be configured as a chip connected to the central processing unit 9100, and the functions of the electronic card risk early warning method based on the knowledge graph are realized through the control of the central processing unit.

[0123] As Figure 11 shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It should be noted that the electronic device 9600 does not necessarily have to include Figure 11 all the components shown in Figure 11 ; in addition, the electronic device 9600 may further include

[0124] components not shown in Figure 11 ; reference may be made to the prior art.

[0125] Among them, the memory 9140 can be, for example, one or more of a buffer, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory, or other suitable devices. The above information related to failure can be stored, and in addition, programs for executing relevant information can also be stored. And the central processing unit 9100 can execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0126] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to supply power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display can be, for example, an LCD display, but is not limited thereto.

[0127] The memory 9140 can be a solid-state memory, for example, a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It can also be a memory that stores information even when power is off, can be selectively erased and has more data stored. Examples of such a memory are sometimes referred to as EPROMs, etc. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 can include an application / function storage section 9142 that is used to store application programs and function programs or the processes for operating the electronic device 9600 by the central processing unit 9100.

[0128] The memory 9140 can also include a data storage section 9143 that is used to store data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage section 9144 of the memory 9140 can include various drivers of the electronic device for communication functions and / or for performing other functions of the electronic device (such as a messaging application, an address book application, etc.).

[0129] The communication module 9110 is a transmitter / receiver 9110 that transmits and receives signals via the antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which can be the same as in the case of a conventional mobile communication terminal.

[0130] Based on different communication technologies, multiple communication modules 9110 can be provided in the same electronic device, such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, etc. The communication module (transmitter / receiver) 9110 is also coupled to the speaker 9131 and the microphone 9132 via the audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby implementing normal telecommunication functions. The audio processor 9130 can include any suitable buffer, decoder, amplifier, etc. Additionally, the audio processor 9130 is also coupled to the central processing unit 9100, so that recording can be performed on the local device via the microphone 9132, and the sound stored on the local device can be played via the speaker 9131.

[0131] Embodiments of the present application also provide a computer-readable storage medium capable of implementing all steps of the method for early warning of electronic card risks based on a knowledge graph with the execution subject being a server or a client in the above embodiments. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, all steps of the method for early warning of electronic card risks based on a knowledge graph with the execution subject being a server or a client in the above embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0132] Step S101: Obtain electronic card application information, and construct an initial state graph according to the electronic card application information and the association relationships between the electronic card application information.

[0133] Step S102: Perform community cutting on the initial state graph according to a preset modularity algorithm model, and determine the risk level of each community according to the matching degree between each community obtained after the cutting and a preset gang fraud rule model, and perform a risk warning operation on the community whose risk level exceeds the risk threshold.

[0134] As can be seen from the above description, the computer-readable storage medium provided by the embodiments of the present application constructs an initial state graph through the electronic card application information and the association relationships between the electronic card application information, and then uses the modularity algorithm model to perform community cutting on the generated initial state graph until the best graph structure is obtained, and then combines the gang fraud rule model to evaluate the risk level of each constructed community as the basis for early warning of gang fraud risks, so as to accurately prevent and control the risk of electronic card gang fraud.

[0135] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a device, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0136] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (devices), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate for implementing in the processFigure 1 means for the functions specified in one or more processes and / or blocks Figure 1 or in a block or blocks.

[0137] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction means which implements the functions specified in one Figure 1 or more processes and / or blocks Figure 1 or in a block or blocks.

[0138] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one Figure 1 or more processes and / or blocks Figure 1 or in a block or blocks.

[0139] Specific embodiments are used in the present invention to illustrate the principles and implementation manners of the present invention. The description of the above embodiments is only for helping to understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. An electronic card risk warning method based on a knowledge graph, characterized in that, the method includes: Obtain electronic card application information, and determine the relationships between various types of information among each electronic card application according to the obtained electronic card application information; Construct an initial state graph according to the electronic card application information and the association relationships between various types of information among the electronic card applications; Perform community cutting on the initial state graph according to a preset modularity algorithm model, and determine the risk levels of each community according to the matching degree between each community obtained after the cutting and a preset gang fraud rule model, and perform risk warning operations on the communities whose risk levels exceed the risk threshold; where the performing community cutting on the initial state graph according to a preset modularity algorithm model specifically includes: determining the initial modularity of the initial state graph according to the preset modularity algorithm model; the initial modularity is: Among them, A i,j represents the weight between node i and node j in the initial state graph spectrum. This weight uses the matching degree obtained during the process of constructing the graph spectrum. m represents the sum of the weights of all edges. Among them, k i represents the sum of the weights of the edges connected to node i, and k j represents the sum of the weights of the edges connected to node j. Among them, k i = ∑ j A i,j ; c i identifies the community to which the node is assigned. d(c i , c j ) is used to determine whether node i and node j are divided into the same community. If node i and node j are divided into the same community, the value is 1; otherwise, the value is 0. wherein, the constructing an initial state graph according to the electronic card application information and the association relationships between the electronic card application information includes: Perform precise matching operations for strong association relationships and matching degree calculation operations for weak association relationships on the electronic card application information according to a preset graph association relationship construction component, and determine the intermediate nodes of the initial state graph according to the results of the precise matching operations and the matching degree calculation operations; Determine the relationship edges of the initial state graph according to the intermediate nodes and the corresponding application nodes, where the electronic card application information is the application nodes of the initial state graph; Construct the initial state graph according to the application nodes, the intermediate nodes, and the relationship edges; the performing matching degree calculation operations for weak association relationships on the electronic card application information according to a preset graph association relationship construction component includes: Perform address fuzzy matching operations on the address type information in the electronic card application information according to a preset standard address database to obtain the weak association relationship matching degree of the address type information; Perform pairwise matching operations on the non-address type information in the electronic card application information to obtain the weak association relationship matching degree of the non-address type information; the determining the relationship edges of the initial state graph according to the intermediate nodes and the corresponding application nodes includes: Establish relationship edges between the application nodes with completely identical electronic card application information and the intermediate nodes during the precise matching operation for strong association relationships, and establish relationship edges between the application nodes with partially identical electronic card application information and the intermediate nodes during the matching degree calculation operation for weak association relationships, where the completely identical application nodes include ID card numbers, and the partially identical application nodes include consistent mobile phone number segments or address similarity greater than a threshold.

2. The electronic card risk warning method based on a knowledge graph according to claim 1, characterized in that, after performing community cutting on the initial state graph according to a preset modularity algorithm model, it includes: Determine the initial modularity of the initial state graph according to a preset modularity algorithm model; All nodes in the initial state graph are sequentially assigned to the communities of adjacent nodes until the change value of modularity before and after assignment is maximized, and each community of the initial state graph is obtained.

3. The method for warning of electronic card risks based on a knowledge graph according to claim 1, characterized in that determining the risk level of each of the communities according to the matching degree between each of the communities obtained after the cutting and a preset gang fraud rule model includes: Obtaining a preset gang fraud rule model, and judging the number of rule hits and the content of rule hits of the preset gang fraud rule model in each of the communities obtained after the cutting, and determining the risk level of each of the communities according to the number of rule hits and the content of rule hits.

4. An electronic card risk warning device based on a knowledge graph, characterized in that it includes: An initial state graph construction module, configured to obtain electronic card application information, and determine the relationship between various types of information between each electronic card application according to the obtained electronic card application information; Construct an initial state graph according to the electronic card application information and the association relationship between various types of information between each electronic card application; A community risk warning module, configured to perform community cutting on the initial state graph according to a preset modularity algorithm model, and determine the risk level of each of the communities according to the matching degree between each of the communities obtained after the cutting and a preset gang fraud rule model, and perform a risk warning operation on the communities whose risk level exceeds a risk threshold; wherein, The community risk warning module is specifically configured to determine the initial modularity of the initial state graph according to a preset modularity algorithm model; the initial modularity is: Among them, A i,j represents the weight between node i and node j in the initial state graph spectrum. This weight uses the matching degree obtained during the construction of the graph spectrum. m represents the sum of the weights of all edges. Among them, k i represents the sum of the weights of the edges connected to node i, and k j represents the sum of the weights of the edges connected to node j. Among them, k i = ∑ j A i,j ; c i identifies the community to which the node is assigned. d(c i , c j ) is used to determine whether node i and node j are divided into the same community. If node i and node j are divided into the same community, the value is 1; otherwise, the value is 0. The initial state graph spectrum construction module includes: An intermediate node determination unit, configured to perform an accurate matching operation of strong association relationships and a matching degree calculation operation of weak association relationships on the electronic card application information according to a preset graph association relationship construction component, and determine the intermediate nodes of the initial state graph according to the results of the accurate matching operation and the matching degree calculation operation; A relationship edge determination unit, configured to determine the relationship edges of the initial state graph according to the intermediate nodes and the corresponding application nodes, wherein the electronic card application information is the application nodes of the initial state graph; A graph construction unit, configured to construct the initial state graph according to the application nodes, the intermediate nodes, and the relationship edges; The intermediate node determination unit is specifically configured to perform an address fuzzy matching operation on the address type information in the electronic card application information according to a preset standard address database to obtain a weak association relationship matching degree of the address type information; perform a pairwise matching operation of digital characters on the non-address type information in the electronic card application information to obtain a weak association relationship matching degree of the non-address type information; The relationship edge determination unit is specifically configured to establish a relationship edge between the application node with completely identical electronic card application information and the intermediary node during the precise matching operation of strong association relationships, and establish a relationship edge between the application node with partially identical electronic card application information and the intermediary node during the matching degree calculation operation of weak association relationships. Among them, the application nodes with completely identical information include the ID number, and the application nodes with partially identical information include consistent mobile phone number segments or an address similarity greater than a threshold.

5. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, when the processor executes the program, the steps of the knowledge graph-based electronic card risk warning method according to any one of claims 1 to 3 are implemented.

6. A computer-readable storage medium, on which a computer program is stored, wherein, when the computer program is executed by the processor, the steps of the knowledge graph-based electronic card risk warning method according to any one of claims 1 to 3 are implemented.

Images