Risk identification method and device based on complaint individual and gang, medium and product
By building a call network map of communication logs, analyzing call characteristics and risk levels, identifying complaint gangs and their associated individuals, the problem of high error rate of malicious complaint identification in the existing technology is solved, and more accurate and efficient risk identification and management is achieved .
Patent Information
- Application Number
- CN202410902889.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-05
- Publication Date
- 2024-11-01
AI Technical Summary
The existing malicious complaint identification methods based on expert rules and supervision models have the problem of high identification error rates, and it is difficult to effectively identify complaint gangs and their associated individuals.
By building a call network map of the communication log, combining the complaint number resource library, analyzing the call characteristics and risk levels, identifying the risk levels of the number to be identified, and determining the target risk numbers and complaint gang numbers associated with it, improving identification accuracy and efficiency.
It effectively improves the ability to identify and manage potential risk complaints, reduces the recognition error rate, monitors and evaluates the risk level of complaint behavior in real time, and provides a more accurate basis for analyzing risk prevention and control strategy.
Smart Images

Figure CN118886705A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of communication technology, and in particular to a risk identification method, device, medium and product based on complaining individuals and groups. Background Art
[0002] In the complex and ever-changing financial ecological environment, illegal complaint agencies have become a major thorny problem. These organizations usually recruit clients through Internet platforms under the guise of "legal consultation" or "agency protection", induce debtors to take malicious complaint methods, and even assist in the production of false materials to deceive financial institutions, thus causing serious infringement on the legitimate rights and interests of financial consumers, seriously disrupting the normal order of the financial market, and undermining the construction of the social credit system.
[0003] Traditional financial institutions mainly rely on expert rules or supervised models based on historical behavioral characteristics to identify and filter potential malicious complaint users. For example, they collect various data related to complaining users, such as basic user information, account activity records, and complaint records; clean and pre-process the collected data to ensure data quality and consistency; formulate a series of expert rules based on domain knowledge and historical data to identify potential malicious complaint users; use supervised learning methods to train models to automatically identify malicious complaint users; deploy the trained models to the actual system to identify and filter potential malicious complaint users in real time.
[0004] However, existing recognition methods rely on highly subjective expert rules and supervised models, and suffer from high recognition error rates. Summary of the invention
[0005] The present application provides a risk identification method, equipment, medium and product based on complaining individuals and groups, which is used to solve the problem that the existing identification methods rely on expert rules and supervised models with strong subjectivity and have a high identification error rate, so as to improve the identification efficiency and accuracy of risk groups.
[0006] In a first aspect, the embodiment of the present application provides a risk identification method based on complaining individuals and gangs, including:
[0007] Obtaining a call network graph of the communication log and complaint numbers in a complaint number resource library, wherein nodes in the call network graph are determined according to numbers in the communication log, and edges in the call network graph are determined according to target call features between numbers;
[0008] According to the complaint number, determine the number to be identified and the risk level of the number to be identified, wherein the number to be identified is a number that has a call feature relationship with the complaint number, and the risk level is determined according to the risk complaint level and call features of the complaint number;
[0009] Determine, according to the call network graph, a first associated number having a first side connection with the number to be identified, and a second associated number having a second side connection, wherein the first side connection and the second side connection are different;
[0010] Determine, according to the risk level, the first associated number and the second associated number, a target risk number that meets the target call feature requirements among the numbers to be identified, and the risk call features of the target risk number in the communication log;
[0011] Based on the risky call characteristics of the target risk number in the communication log, determine the complaint group number that has a risky association with the target risk number.
[0012] In a possible implementation, obtaining a call network map of a communication log and a complaint number in a complaint number resource library includes:
[0013] Obtain communication logs and complaint numbers from the complaint number resource library;
[0014] Determine the characteristics of calls between the complaint number and other numbers in the communication log;
[0015] According to the influence of call features on the risk association between the complaint number and other numbers in the communication log, the call features are screened and processed to determine the target call features;
[0016] The numbers in the communication log are used as nodes and the target call features between the numbers are used as edges to construct a call network graph.
[0017] In a possible implementation, determining the number to be identified and the risk level of the number to be identified according to the complaint number includes:
[0018] According to the complaint number, the numbers in the communication log are selected to obtain the numbers to be identified that have a call feature relationship with the complaint number;
[0019] According to the preset call time period, determine the call characteristics between the complaint number and the number to be identified in the communication log, where the call characteristics include call frequency, duration, number of days of call, call direction, and connection status;
[0020] Based on the call characteristics and the preset risk weights of the call characteristics, the risk level index of the number to be identified and the risk level of the number to be identified corresponding to the risk level index are determined, wherein the risk weight of call frequency is greater than the risk weight of duration, and the risk weight of duration is greater than the risk weight of call days.
[0021] In a possible implementation, determining the risk level index of the number to be identified and the risk level of the number to be identified corresponding to the risk level index according to the call characteristics and the preset risk weights of the call characteristics includes:
[0022] Determine the channels for obtaining complaint numbers, including industry supervision channels, consumer rights protection channels, and public complaint channels;
[0023] Determine the risk complaint level of the complaint number based on the way the complaint number is obtained;
[0024] According to the preset risk weights of the call features, the call features are weighted and summed to obtain the risk level index of the number to be identified;
[0025] The risk level of the number to be identified is determined from a preset number risk level range according to the risk complaint level of the complaint number and the risk level index of the number to be identified.
[0026] In a possible implementation manner, after determining the number to be identified in the number and the risk level of the number to be identified according to the complaint number, the method further includes:
[0027] Determine the node attributes in the call network graph according to the risk level of the number to be identified;
[0028] The call network map is displayed on a visual display interface based on node attributes, wherein the visual display interface provides a visual tool for further analysis so that the user can identify nodes of different risk levels in the visual display interface.
[0029] In a possible implementation, determining a target risk number that meets the target call feature requirement among the numbers to be identified and the risk call feature of the target risk number in the communication log according to the risk level, the first associated number, and the second associated number includes:
[0030] Analyze the target call feature of the number to be identified according to the first associated number and the second associated number to obtain target call feature distribution information;
[0031] Determine the target call feature requirements based on the target call feature distribution information and the risk level of the number to be identified in the call network map;
[0032] According to the target call feature requirements, determine the target risk number from the numbers to be identified;
[0033] According to the risk level of the target risk number, determine the risk call characteristics of the target risk number in the communication log.
[0034] In a possible implementation, determining the complaint group number having a risk association relationship with the target risk number according to the risk call characteristics of the target risk number in the communication log includes:
[0035] Determine the call closeness index between the target risk number and other numbers based on the risk call characteristics of the target risk number in the communication log;
[0036] According to the call density index, the clustering relationship map between the target risk number and other numbers is obtained;
[0037] According to the preset risk association requirements, the cluster relationship map is analyzed to determine the complaint group numbers that have a risk association relationship with the target risk number.
[0038] In a possible implementation, after determining the complaint group number having a risk association relationship with the target risk number according to the risk call characteristics of the target risk number in the communication log, the method further includes:
[0039] Get updated communication logs;
[0040] According to the target risk number, the number in the updated communication log is selected to obtain the initial risk number that meets the preset call characteristic frequency requirement with the target risk number;
[0041] Screen the initial risk numbers to identify potential risk numbers that are different from the complaint group numbers;
[0042] Determine the potential risk level of the potential risk number based on the call characteristics between the target risk number and the potential risk number;
[0043] According to the updated communication log and the potential risk level of the potential risk number, the call network map is updated to obtain an updated call network map.
[0044] In a second aspect, the embodiment of the present application provides a risk identification device based on complaining individuals and gangs, including:
[0045] An acquisition module, used to acquire a call network graph of a communication log and complaint numbers in a complaint number resource library, wherein nodes in the call network graph are determined according to numbers in the communication log, and edges in the call network graph are determined according to target call features between numbers;
[0046] A first determination module is used to determine, according to the complaint number, the number to be identified among the numbers and the risk level of the number to be identified, wherein the number to be identified is a number among the numbers having a call feature relationship with the complaint number, and the risk level is determined according to the risk complaint level and the call feature of the complaint number;
[0047] A second determination module is used to determine, according to the call network map, a first associated number having a first side connection with the number to be identified and a second associated number having a second side connection, wherein the first side connection and the second side connection are different;
[0048] A third determination module is used to determine a target risk number that meets the target call feature requirements among the numbers to be identified, and the risk call features of the target risk number in the communication log according to the risk level, the first associated number and the second associated number;
[0049] The fourth determination module is used to determine the complaint group number that has a risk association relationship with the target risk number based on the risk call characteristics of the target risk number in the communication log.
[0050] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor;
[0051] Memory stores computer-executable instructions;
[0052] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.
[0053] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementations of the first aspect.
[0054] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.
[0055] The risk identification method, device, medium and product based on complaint individuals and groups provided in the embodiments of the present application collect and analyze communication logs and complaint number resource libraries to construct a call network map, and then identify and evaluate the risk level and correlation of numbers in the communication log through complaint numbers, thereby effectively improving the ability to identify and manage potential risk complaints; then, by analyzing the first associated number and the second associated number of each number in the call network map, the target risk number that meets the target call feature requirements and the risk call characteristics of the target risk number in the communication log can be determined among the numbers to be identified, that is, by analyzing the call characteristics of the associated numbers, the high-risk numbers that meet the requirements are determined, thereby improving the identification accuracy of the risk numbers; finally, by analyzing the risk call characteristics of the target risk number in the communication log, the means of determining the complaint group number that has a risk-related relationship with the target risk number is determined, and the complaint group that is closely connected with the complaint individual corresponding to the high-risk number and has frequent calls can be obtained, thereby improving the identification efficiency of the complaint individuals and groups, and providing a more accurate analysis basis for formulating targeted risk prevention and control strategies. BRIEF DESCRIPTION OF THE DRAWINGS
[0056] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0057] Figure 1 A flowchart of a risk identification method based on complaining individuals and gangs provided for this application;
[0058] Figure 2 A flowchart of another risk identification method based on complaining individuals and gangs provided for this application;
[0059] Figure 3 A specific flow chart of another risk identification method based on complaining individuals and gangs provided for this application;
[0060] Figure 4 A schematic diagram of the structure of the risk identification device based on complaining individuals and gangs provided for this application;
[0061] Figure 5 A schematic diagram of the structure of the electronic device provided in this application.
[0062] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION
[0063] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.
[0064] It should be noted that the user information (including but not limited to communication logs and user-related personal information, telephone numbers, user device identification, etc.) and data (including but not limited to call records for analysis, stored call records, displayed call records, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards, and provide corresponding operation entrances for users to choose to authorize or refuse.
[0065] In the existing technology, the identification and filtering of potential malicious complaint users mainly rely on expert rules or supervised models based on historical behavior characteristics. For example, various data related to the complaining users are collected, such as basic user information, account activity records, and complaint records; the collected data is cleaned and preprocessed to ensure the quality and consistency of the data; a series of expert rules are formulated based on domain knowledge and historical data to identify potential malicious complaint users; supervised learning methods are used to train the model to automatically identify malicious complaint users; the trained model is deployed to the actual system to identify and filter potential malicious complaint users in real time. Among them, expert rules are usually formulated based on historical experience, which are highly subjective and difficult to adapt to the rapidly changing malicious complaint methods and behavior patterns, resulting in an increase in the recognition error rate. At the same time, the supervised model requires a large amount of historical labeled data for training, and there is data lag. Moreover, the behavior patterns of malicious complaint users may change over time, and the model needs to be regularly updated and retrained, increasing the maintenance cost. Therefore, this method fails to deeply explore the associated individuals around the gang and their potential risks, and the recognition efficiency is not high.
[0066] In order to solve the above problems, the embodiments of the present application provide a risk identification method, device, medium and product based on complaining individuals and groups. By collecting and analyzing communication logs and complaint number resource libraries, a call network map can be constructed. Then, the risk of the number can be quantified and prioritized in combination with the call characteristics and the risk level of the complaint number. When identifying and evaluating risk numbers, the complex relationship between numbers can be revealed, the identification and management capabilities of target risk numbers are improved, and clues are provided for further risk assessment and investigation. By analyzing the call characteristics and associated numbers of the target risk numbers, risk groups that are closely connected to high-risk numbers and frequently call can be identified, which improves the efficiency of identifying complaining individuals and groups. At the same time, due to the dynamic characteristics of the communication logs, the call characteristics in the call network map are real-time, thereby improving the accuracy of risk identification based on complaining individuals and groups.
[0067] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.
[0068] The execution subject of the risk identification method based on complaining individuals and gangs provided in the embodiment of the present application can be a server, wherein the server can be a mobile phone, a computer, a tablet and other devices. The embodiments of the present application do not impose any particular restrictions on the implementation method of the execution entity, as long as the execution entity can obtain the call network map of the communication log and the complaint numbers in the complaint number resource library, wherein the nodes in the call network map are determined according to the numbers in the communication log, and the edges in the call network map are determined according to the target call characteristics between the numbers; according to the complaint number, the number to be identified in the number and the risk level of the number to be identified are determined, wherein the number to be identified is a number in the number that has a call characteristic relationship with the complaint number, and the risk level is determined according to the risk complaint level and call characteristics of the complaint number; according to the call network map, a first associated number having a first edge connection with the number to be identified and a second associated number having a second edge connection are determined, and the first edge connection and the second edge connection are different; according to the risk level, the first associated number and the second associated number, the target risk number that meets the target call characteristic requirements in the number to be identified and the risk call characteristics of the target risk number in the communication log are determined; according to the risk call characteristics of the target risk number in the communication log, the complaint group number having a risk association relationship with the target risk number is determined.
[0069] Among them, individual complainers can refer to high-risk complaint users corresponding to target risk numbers that meet the target call feature requirements; and groups can refer to complaint groups with common characteristics or behavior patterns. In some instances, risk correlation relationship analysis can be performed to identify them based on the call features between the target risk number and other numbers.
[0070] Communication logs may refer to massive communication logs obtained from telecommunications operators, network service providers or related systems. These logs usually contain detailed records of communication activities. It should be noted that the communication logs involved in this application are all information and data authorized by users or fully authorized by all parties. The information related to the numbers obtained from the communication logs are all encrypted or de-identified information, that is, the information related to the number obtained through this application cannot be used to determine the user's personal information (such as personal identity information). For example, call records between numbers are obtained through communication logs, such as call frequency, duration, number of days, call direction, connection status, etc. within a specified time.
[0071] A call network graph can be a visualization and analysis tool for representing and studying the connections and interaction patterns between individuals (such as phone numbers, users) in telephone communications. For example, a call network graph is constructed with phone numbers as nodes and call records as edges. By analyzing the graph, the communication patterns between numbers can be identified, such as groups that communicate frequently or abnormal communication behaviors.
[0072] Figure 1 The flowchart of a method for risk identification based on complaining individuals and groups provided by this application is shown below. The execution subject of this method can be a server or other server storing the method for risk identification based on complaining individuals and groups. This embodiment is not particularly limited here. Figure 1 As shown, the method may include:
[0073] S101. Obtain a call network graph of the communication log and complaint numbers in a complaint number resource library, wherein nodes in the call network graph are determined according to numbers in the communication log, and edges in the call network graph are determined according to target call features between numbers.
[0074] Among them, the call network graph can be a network graph with mobile phone numbers as nodes and call records as edges. Furthermore, the mobile phone number is the number in the communication log, and the call record can refer to the target call characteristics such as call frequency, duration, and number of days between each number. For example, by characterizing the edge weight with call frequency, a graph based on call frequency can be constructed; by characterizing the edge weight with duration, a graph based on duration can be constructed. By constructing a multi-dimensional call network graph through the target call characteristics corresponding to different edge weights, the call relationship between numbers can be analyzed more comprehensively from various dimensions, thereby improving the accuracy of risk number assessment and identification.
[0075] In this step, the communication log can be the log information within a specified time period, which is used to determine the communication association between the complaint number and other numbers within the time period, so as to further determine the risk level of each number in the communication log, thereby analyzing the current call network map and identifying the target risk number within the time period. Furthermore, when the communication log is dynamically updated over time, the corresponding call network map will also be dynamically updated to update the target risk number in real time, thereby achieving the purpose of dynamically identifying risk groups.
[0076] The complaint number resource library can be a database constructed after classifying the collected complaint numbers. For example, known and publicly disclosed complaint numbers are collected through channels such as industry supervision, consumer rights protection, and public complaints, and the numbers are classified and stored in the database according to the way the complaint numbers were obtained, so that there are different categories of complaint numbers in the complaint number resource library, and different categories of complaint numbers correspond to different risk levels, so as to provide a basis for risk level judgment when identifying risk numbers in communication logs.
[0077] Among them, in the embodiment of the present application, the method of obtaining the call network map of the communication log and the complaint number in the complaint number resource library may include:
[0078] Obtain communication logs and complaint numbers from the complaint number resource library;
[0079] Determine the characteristics of calls between the complaint number and other numbers in the communication log;
[0080] According to the influence of call features on the risk association between the complaint number and other numbers in the communication log, the call features are screened and processed to determine the target call features;
[0081] The numbers in the communication log are used as nodes and the target call features between the numbers are used as edges to construct a call network graph.
[0082] Among them, other numbers in the communication log may refer to numbers in the communication log that are different from the complaint number. For example, given the complaint number a and the numbers a, D, and E in the communication log, it is necessary to determine the call characteristics between a and other numbers D and E respectively. Furthermore, the call frequency, duration, number of days of calls, call direction, and connection status between a and D, and a and E can be determined.
[0083] The degree of influence of call features on the risk association between the complaint number and other numbers in the communication log can refer to the degree to which each call feature characterizes a close connection with the complaint number. The closer the connection with the complaint number, the easier it is to identify as a risk number. Among them, the degree of close connection can be determined by human experience or by analyzing historical data. For example, the call data of historical risk groups is analyzed, and the result that the call frequency can better reflect the close connection between the historical risk group and the complaint number than the number of call days. Therefore, choosing call frequency as the target call feature to construct a call network map can more accurately identify risk numbers.
[0084] S102. Determine the numbers to be identified and the risk level of the numbers to be identified based on the complaint number, wherein the numbers to be identified are numbers that have a call feature relationship with the complaint number, and the risk level is determined based on the risk complaint level and call features of the complaint number.
[0085] The numbers in the communication log may include complaint numbers, numbers to be identified, and irrelevant numbers that have no call feature relationship with the complaint numbers. In this step, since the irrelevant numbers have no call feature relationship with the complaint numbers, the risk level of the irrelevant numbers is set to zero risk level by default; the numbers to be identified are classified into one of the risk levels 1-5.
[0086] The risk complaint level of a complaint number can be determined by the way the complaint number is obtained. Furthermore, the level can be set according to the reliability of the way the number is obtained. For example, since industry supervision belongs to the official public channel and has the highest reliability, the risk complaint level corresponding to the complaint number of this category is set to a high complaint level. Similarly, consumer rights protection is a medium complaint level, and public complaints are low complaint levels. In one example, complaint number a belongs to the industry supervision category, complaint number b belongs to the consumer rights protection category, and complaint number c belongs to the public complaint category. The risk complaint level of a is a high complaint level, the risk complaint level of b is a medium complaint level, and the risk complaint level of c is a low complaint level.
[0087] In the embodiment of the present application, the method for determining the number to be identified and the risk level of the number to be identified according to the complaint number may include:
[0088] According to the complaint number, the numbers in the communication log are selected to obtain the numbers to be identified that have a call feature relationship with the complaint number;
[0089] According to the preset call time period, determine the call characteristics between the complaint number and the number to be identified in the communication log, where the call characteristics include call frequency, duration, number of days of call, call direction, and connection status;
[0090] Based on the call characteristics and the preset risk weights of the call characteristics, the risk level index of the number to be identified and the risk level of the number to be identified corresponding to the risk level index are determined, wherein the risk weight of call frequency is greater than the risk weight of duration, and the risk weight of duration is greater than the risk weight of call days.
[0091] The preset call time period may refer to a specified time period set in advance for risk group analysis. For example, the call variables of the call frequency, call duration, call days, call direction (caller and called), and connection status (whether connected) between the number to be identified and the complaint number a in the past seven days are determined.
[0092] Before determining the risk level index of the number to be identified, a risk weight is assigned to each call feature. This weight reflects the importance of the feature in risk assessment. For example, call frequency may have a higher risk weight than call days. For example, when the number F to be identified calls the complaint number a and the call is connected, the corresponding call frequency is the main weight, the call duration is the secondary weight, and the call days are the final weight. Furthermore, since the magnitudes of different features may be different, they need to be standardized to the same scale for easy comparison and weighting; then the standardized value of each call feature is multiplied by its risk weight to obtain the risk score of each feature; finally, the risk scores of all call features are added to obtain the risk level index of the number to be identified, and the risk level is determined by the risk level index. Since the call features in the communication log will be updated and changed over time, the risk level of the number to be identified will also change dynamically over time to solve the problem of sample data lag in the existing technology when identifying risky numbers.
[0093] Among them, in the embodiment of the present application, according to the call characteristics and the preset risk weights of the call characteristics, the method for determining the risk level index of the number to be identified and the risk level of the number to be identified corresponding to the risk level index may include:
[0094] Determine the channels for obtaining complaint numbers, including industry supervision channels, consumer rights protection channels, and public complaint channels;
[0095] Determine the risk complaint level of the complaint number based on the way the complaint number is obtained;
[0096] According to the preset risk weights of the call features, the call features are weighted and summed to obtain the risk level index of the number to be identified;
[0097] The risk level of the number to be identified is determined from a preset number risk level range according to the risk complaint level of the complaint number and the risk level index of the number to be identified.
[0098] In one example, a correlation table between the risk level, the risk complaint level, and the risk level index can be set up, and the risk level of the number to be identified can be determined through the correlation table.
[0099] In the embodiment of the present application, after determining the number to be identified and the risk level of the number to be identified in the number according to the complaint number, the method may further include:
[0100] Determine the node attributes in the call network graph according to the risk level of the number to be identified;
[0101] The call network map is displayed on a visual display interface based on node attributes, wherein the visual display interface provides a visual tool for further analysis so that the user can identify nodes of different risk levels in the visual display interface.
[0102] In one example, the visual display interface can encode nodes and edges by different colors, sizes, or shapes to intuitively display the level of risk, making high-risk areas or patterns clear at a glance. The risk level is dynamically updated through real-time changes in call characteristics in the communication log, and the visual display interface can reflect the latest communication data and risk changes. Users can explore the data in depth through interactive tools, such as clicking on nodes to view detailed information, or adjusting parameters to observe changes in risk distribution, so as to understand the risk distribution and formulate corresponding strategies.
[0103] S103. Determine, according to the call network graph, a first associated number having a first side connection with the number to be identified and a second associated number having a second side connection, wherein the first side connection and the second side connection are different.
[0104] In one example, the first edge connection may refer to a direct connection with the number to be identified, that is, the node directly connected to the number to be identified in the call network graph is the first associated number. Furthermore, direct connection usually means that there is a direct call feature between the two numbers. The strength and frequency of these direct connections are evaluated by checking the attributes of the edges connecting the numbers to be identified, such as the number of calls, call duration, etc.
[0105] The second side connection may refer to an indirect connection with the number to be identified, that is, a number in the call network graph that has no direct connection with the number to be identified but is connected through at least one intermediate node is a second associated number. Furthermore, the second associated number may refer to a number connected to the number to be identified through an intermediate node, and the intermediate node is the first associated number connecting the number to be identified and the second associated number.
[0106] S104. Determine, according to the risk level, the first associated number and the second associated number, a target risk number that meets the target call feature requirements among the numbers to be identified, and the risk call features of the target risk number in the communication log.
[0107] Among them, whether the number to be identified is a target risk number that meets the target call feature requirements can be determined by performing risk assessment on the first associated number and the second associated number, that is, considering the connection strength between their call features and the number to be identified.
[0108] Among them, in the embodiment of the present application, according to the risk level, the first associated number and the second associated number, the method for determining the target risk number that meets the target call feature requirements among the numbers to be identified, and the risk call features of the target risk number in the communication log may include:
[0109] Analyze the target call feature of the number to be identified according to the first associated number and the second associated number to obtain target call feature distribution information;
[0110] Determine the target call feature requirements based on the target call feature distribution information and the risk level of the number to be identified in the call network map;
[0111] According to the target call feature requirements, determine the target risk number from the numbers to be identified;
[0112] According to the risk level of the target risk number, determine the risk call characteristics of the target risk number in the communication log.
[0113] Among them, the target call feature distribution information can characterize the risk association degree between the number to be identified and the associated number. Furthermore, the target call feature distribution information can be statistically analyzed by quantile analysis or cluster analysis, and the target call feature requirements can be set based on the risk level of the number to be identified. In one example, the preliminary target risk number threshold is determined by quantile analysis, and the node numbers that meet the threshold requirements are screened, and then the target risk numbers with higher risk levels are selected from the node numbers according to the risk levels of these node numbers.
[0114] In another example, a preliminary target risk number threshold is determined, and node numbers that meet the threshold requirements are screened. When the risk level of the node number meets the preset level requirements, and the node number has a call feature relationship with multiple other numbers and the multiple other numbers meet the preset number of numbers, the node number is determined to be a target risk number.
[0115] Furthermore, after determining the target risk number, it is also necessary to determine its corresponding risk call characteristics, which may include multiple target call characteristics and other call characteristics that affect the risk association relationship with the target risk number.
[0116] S105. Determine the complaint group number that has a risk association relationship with the target risk number based on the risk call characteristics of the target risk number in the communication log.
[0117] Among them, the call closeness between other numbers and the target risk number is calculated by using the risk call characteristics. The call closeness can be determined according to the preset calculation rules to characterize the close connection between other numbers and the target risk number to determine the complaint group number. For example, the numbers that are in contact with the target risk number H include T, Y, U, I, and O. The risk call characteristics include call frequency, call duration, and call days. It is necessary to calculate the closeness between HT, HY, HU, HI, and HO respectively.
[0118] Among them, in the embodiment of the present application, according to the risk call characteristics of the target risk number in the communication log, the method for determining the complaint group number having a risk association relationship with the target risk number may include:
[0119] Determine the call closeness index between the target risk number and other numbers based on the risk call characteristics of the target risk number in the communication log;
[0120] According to the call density index, the clustering relationship map between the target risk number and other numbers is obtained;
[0121] According to the preset risk association requirements, the cluster relationship map is analyzed to determine the complaint group numbers that have a risk association relationship with the target risk number.
[0122] In one example, the call frequency between HTs is 10 times, the call duration is 300 minutes, and the call days are 5 days. The closeness index can meet the following preset calculation rules; call frequency weight*(call frequency / maximum call frequency)+call duration weight*(call duration / maximum call duration)+call days weight*(call days / maximum call days), among which the maximum call frequency, maximum call duration and maximum call days are the maximum values of the number of calls, duration and days in the communication log, which are used to standardize the score.
[0123] Among them, the method for determining the cluster relationship map can be: select the target risk number as the central node of the cluster analysis, and other numbers as potential cluster nodes; use the call density index to distribute the nodes in the cluster map, and the nodes with higher density are closer to the cluster center.
[0124] The preset risk association requirement may refer to the need to determine the risk association degree of the complaint group number, that is, the higher the risk association requirement, the closer the determined complaint group number is to the target risk number in the aggregation relationship map, and the more accurate the complaint group number is. Therefore, according to the next use of risk assessment and group identification (such as risk management, finding the hidden mode of illegal agency complaint organizations, etc.), appropriate risk association requirements can be set to identify risk groups with different risk ranges.
[0125] Among them, in the embodiment of the present application, after determining the complaint group number having a risk association relationship with the target risk number according to the risk call characteristics of the target risk number in the communication log, the method may also include:
[0126] Get updated communication logs;
[0127] According to the target risk number, the number in the updated communication log is selected to obtain the initial risk number that meets the preset call characteristic frequency requirement with the target risk number;
[0128] Screen the initial risk numbers to identify potential risk numbers that are different from the complaint group numbers;
[0129] Determine the potential risk level of the potential risk number based on the call characteristics between the target risk number and the potential risk number;
[0130] According to the updated communication log and the potential risk level of the potential risk number, the call network map is updated to obtain an updated call network map.
[0131] Among them, the preset call feature frequency requirement may refer to a quantitative requirement for significant call features with the target risk number, for example, a call frequency of more than 20 times in one day.
[0132] Furthermore, since there is a situation where the initial risk number and the complaint group number are repeated, the potential risk number is a newly added risk number that is not in the complaint group number. For example, before the communication log is updated, there are 30 complaint group numbers corresponding to the target risk number. After the communication log is updated, there are 50 initial risk numbers determined. By comparison, 20 potential risk numbers are found. These 20 potential risk numbers may develop into target risk numbers in the future. According to the corresponding potential risk levels, targeted risk prevention and control strategies can be formulated to improve the efficiency of risk prevention and control.
[0133] Among them, the updated call network map can be analyzed, and the first associated number and the second associated number of each number can be determined, so as to determine the target risk number that meets the target call feature requirements among the numbers to be identified, and the risk call features of the target risk number in the communication log; finally, by analyzing the risk call features of the target risk number in the communication log, the step of determining the complaint group number that has a risk association relationship with the target risk number can be achieved, so as to achieve the purpose of regularly updating the target risk number and the complaint group number as the communication log is continuously updated. Furthermore, a risk analysis report that changes over time can be generated based on the update situation, providing data-based decision support for specifying targeted risk management strategies, such as monitoring high-risk numbers.
[0134] The method provided in the embodiments of the present application can be applied to multiple fields and scenarios such as finance, claims services, and e-commerce. Especially in the current complex and changeable financial ecological environment, there is a highly organized and professional black-gray industrial chain hidden behind the illegal agency complaint organization, which not only causes serious infringement on the legitimate rights and interests of financial consumers, but also seriously disrupts the normal order of the financial market and undermines the construction of the social credibility system. The present application can accurately identify and block illegal agency complaints, protect the rights and interests of financial consumers, and maintain market order.
[0135] The risk identification method based on complaining individuals and groups provided in the embodiment of the present application collects and analyzes communication logs and a complaint number resource library to construct a call network map, and then identifies and evaluates the risk level and correlation of the numbers in the communication log through the complaint numbers, thereby effectively improving the ability to identify and manage potential risk complaints; then, by analyzing the first associated number and the second associated number of each number in the call network map, the target risk number that meets the target call feature requirements among the numbers to be identified, as well as the risk call characteristics of the target risk number in the communication log can be determined, that is, by analyzing the call characteristics of the associated numbers, the high-risk numbers that meet the requirements are determined, thereby improving the accuracy of identifying risk numbers; finally, by analyzing the risk call characteristics of the target risk number in the communication log, the means of determining the complaint group number that has a risk-related relationship with the target risk number is determined, and the user group corresponding to the number that is closely connected with the high-risk number and frequently calls can be obtained, thereby improving the efficiency of identifying complaining individuals and groups, and providing a more accurate analysis basis for formulating targeted risk prevention and control strategies. At the same time, it enables financial institutions to identify the behavioral patterns and hidden patterns of illegal agency complaint organizations (risk groups), monitor and evaluate the risk level of complaint behaviors in real time, issue early warnings in a timely manner, and formulate targeted risk prevention and control strategies to effectively prevent and control financial risks and public opinion crises caused by illegal agency complaints.
[0136] Figure 2 A flowchart of another risk identification method based on complaining individuals and gangs provided in an embodiment of the present application. Figure 2 As shown, the method may include:
[0137] S201. Build a complaint number resource library.
[0138] Among them, complaint numbers are collected through various known channels (such as industry supervision), the collected complaint numbers are classified, and a number database is established to achieve standardized management of basic data.
[0139] S202: Call feature extraction and risk quantification.
[0140] Among them, advanced data mining technology is used to screen the target numbers with call records with the numbers in the complaint number database from the massive communication logs, and the multi-dimensional characteristics such as the call frequency, duration, number of days of call, call direction and connection status in the specified time period are recorded in detail. Furthermore, a dynamic risk rating mechanism is introduced to default the number with no association to zero risk level, and the remaining numbers are divided into risk levels ranging from 1 to 5 based on the frequency of calls with the complaint number and other call characteristics, to ensure the refinement of risk assessment.
[0141] S203: Construct a call network map.
[0142] In one example, complex network theory is used to construct a multi-dimensional call network map with mobile phone numbers as nodes and call records as edges. The node attributes include not only the number itself, but also the risk level of S202. The weight of the edge takes into account the number of calls, duration and days, forming an information-rich visual risk distribution model. Furthermore, different edge weights are set according to different call characteristics between numbers to construct different types of call network maps. For example, call network map 1 can be constructed using the call frequency between number A and number B; call network map 2 can be constructed using the call duration between number A and number B; call network map 3 can be constructed using the call days between number A and number B. Call network map 1, call network map 2, and call network map 3 form a multi-dimensional call network map. Different maps can be used to identify risk numbers with different characteristics, and risk numbers can be analyzed from a multi-dimensional perspective to determine risk intermediaries, thereby improving the accuracy of identification.
[0143] S204, risk intermediary and gang identification.
[0144] Among them, through deep network analysis, by identifying the first-degree and second-degree related users in the graph, a reasonable risk threshold is automatically set based on the algorithm to mark those common high-risk contacts as "risk intermediaries". At the same time, the algorithm clusters those user groups that are closely connected and frequently talk to risk intermediaries, defining them as "risk gangs". First-degree related users can refer to users who are directly connected to the target user, here refers to the nodes directly connected to the target node in the call network graph. Second-degree related users can refer to users who have no direct contact with the target user, but are indirectly connected through first-degree related users, that is, there is a first-degree related user node between the target node and the call network graph.
[0145] S205. Prediction of potential complainants.
[0146] Among them, based on the call network analysis, users with significant call characteristics (such as high frequency calls in a short period of time) with risk intermediaries can be proactively identified and preset as "potential complaint risk objects", and their risk levels can be dynamically adjusted according to their call patterns, providing a strategic basis for early intervention and resolution of potential complaint events. For example, the call pattern can refer to specific call record requirements, such as the quantile requirements of call frequency: the total call frequency of the known risk group is 100 times, then when the call frequency between the number and the risk intermediary is 90-100 times (excluding 100 times), the quantile is 90 points; when the call frequency between the number and the risk intermediary is 80-90 times (excluding 90 times), the quantile is 80 points; when the call frequency between the number and the risk intermediary is 70-80 times (excluding 80 times), the quantile is 70 points, and so on, determine the quantile of the number, and analyze the quantiles of these numbers to determine the potential complaint risk objects.
[0147] The risk identification method based on complaining individuals and gangs provided in the embodiment of the present application divides users into different risk levels according to call records, thereby realizing rapid identification and classification of risks; constructs a call network containing user risk level information, intuitively displays risk associations, and provides visualization tools for in-depth analysis, so as to accurately identify and block illegal agency complaints, protect the rights and interests of financial consumers, and maintain market order; uses call data and risk propagation models to deeply explore the associated individuals around the gang and their potential risks, predicts the complaining users that may be generated in the future, and provides a scientific basis for early intervention.
[0148] Figure 3 A specific flow chart of another risk identification method based on complaining individuals and gangs provided for this application is as follows: Figure 3 As shown, in this embodiment Figure 2 Based on the embodiment, a risk identification method based on complaining individuals and gangs is described in detail, and the method includes:
[0149] (1) Methods for constructing a complaint number resource library may include: searching for classified complaint numbers; establishing databases: industry supervision, consumer rights protection, and public complaints.
[0150] (2) Methods for extracting call features and quantifying risks may include: screening call records; recording multi-dimensional features: frequency, duration, number of days, direction, status; dynamic complaint risk rating: 0-5 levels.
[0151] (3) The method of constructing a call network graph may include: constructing a network graph: mobile phone numbers are nodes; setting node attributes: risk level; edge weight calculation: number of calls, duration, and number of days.
[0152] (4) Methods for identifying risk intermediaries and risk groups may include: call network analysis; identifying first- and second-degree related users; and labeling risk intermediaries and risk groups.
[0153] (5) Methods for predicting potential complainants may include: analyzing call patterns; identifying potential complaint risk objects; and dynamically adjusting risk levels.
[0154] The risk identification method based on complaining individuals and groups provided in the embodiment of the present application reveals the hidden correlation pattern of illegal agency complaint organizations (i.e., risk groups) by analyzing multi-dimensional data such as the call relationship between complainants and the time correlation of complaint behaviors, thereby improving the accuracy of identifying risk groups, thereby providing financial institutions with more accurate risk prevention and control measures, effectively preventing and controlling financial risks and public opinion crises caused by illegal agency complaints, and promoting the healthy development of the financial market.
[0155] Figure 4 The schematic diagram of the structure of the risk identification device based on complaining individuals and gangs provided in this application is as follows: Figure 4 As shown, the risk identification device 40 based on complaining individuals and gangs provided in this embodiment includes:
[0156] The acquisition module 401 is used to acquire the call network graph of the communication log and the complaint number in the complaint number resource library, wherein the nodes in the call network graph are determined according to the numbers in the communication log, and the edges in the call network graph are determined according to the target call characteristics between the numbers;
[0157] The first determination module 402 is used to determine the number to be identified and the risk level of the number to be identified according to the complaint number, wherein the number to be identified is a number having a call feature relationship with the complaint number, and the risk level is determined according to the risk complaint level and call feature of the complaint number;
[0158] A second determination module 403 is used to determine, according to the call network map, a first associated number having a first side connection with the number to be identified and a second associated number having a second side connection, wherein the first side connection and the second side connection are different;
[0159] The third determination module 404 is used to determine the target risk number that meets the target call feature requirements among the numbers to be identified, and the risk call features of the target risk number in the communication log according to the risk level, the first associated number and the second associated number;
[0160] The fourth determination module 405 is used to determine the complaint group number having a risk association relationship with the target risk number according to the risk call characteristics of the target risk number in the communication log.
[0161] In a possible implementation, the acquisition module 401 may also be used to:
[0162] Obtain communication logs and complaint numbers from the complaint number resource library;
[0163] Determine the characteristics of calls between the complaint number and other numbers in the communication log;
[0164] According to the influence of call features on the risk association between the complaint number and other numbers in the communication log, the call features are screened and processed to determine the target call features;
[0165] The numbers in the communication log are used as nodes and the target call features between the numbers are used as edges to construct a call network graph.
[0166] In a possible implementation, the first determining module 402 may also be used to:
[0167] According to the complaint number, the numbers in the communication log are selected to obtain the numbers to be identified that have a call feature relationship with the complaint number;
[0168] According to the preset call time period, determine the call characteristics between the complaint number and the number to be identified in the communication log, where the call characteristics include call frequency, duration, number of days of call, call direction, and connection status;
[0169] Based on the call characteristics and the preset risk weights of the call characteristics, the risk level index of the number to be identified and the risk level of the number to be identified corresponding to the risk level index are determined, wherein the risk weight of call frequency is greater than the risk weight of duration, and the risk weight of duration is greater than the risk weight of call days.
[0170] In a possible implementation, the first determining module 402 may also be used to:
[0171] Determine the channels for obtaining complaint numbers, including industry supervision channels, consumer rights protection channels, and public complaint channels;
[0172] Determine the risk complaint level of the complaint number based on the way the complaint number is obtained;
[0173] According to the preset risk weights of the call features, the call features are weighted and summed to obtain the risk level index of the number to be identified;
[0174] The risk level of the number to be identified is determined from a preset number risk level range according to the risk complaint level of the complaint number and the risk level index of the number to be identified.
[0175] In a possible implementation, the first determining module 402 may also be used to:
[0176] Determine the node attributes in the call network graph according to the risk level of the number to be identified;
[0177] The call network map is displayed on a visual display interface based on node attributes, wherein the visual display interface provides a visual tool for further analysis so that the user can identify nodes of different risk levels in the visual display interface.
[0178] In a possible implementation, the third determining module 404 may also be used to:
[0179] Analyze the target call feature of the number to be identified according to the first associated number and the second associated number to obtain target call feature distribution information;
[0180] Determine the target call feature requirements based on the target call feature distribution information and the risk level of the number to be identified in the call network map;
[0181] According to the target call feature requirements, determine the target risk number from the numbers to be identified;
[0182] According to the risk level of the target risk number, determine the risk call characteristics of the target risk number in the communication log.
[0183] In a possible implementation, the fourth determining module 405 may also be used to:
[0184] Determine the call closeness index between the target risk number and other numbers based on the risk call characteristics of the target risk number in the communication log;
[0185] According to the call density index, the clustering relationship map between the target risk number and other numbers is obtained;
[0186] According to the preset risk association requirements, the cluster relationship map is analyzed to determine the complaint group numbers that have a risk association relationship with the target risk number.
[0187] In a possible implementation, the fourth determining module 405 may also be used to:
[0188] Get updated communication logs;
[0189] According to the target risk number, the number in the updated communication log is selected to obtain the initial risk number that meets the preset call characteristic frequency requirement with the target risk number;
[0190] Screen the initial risk numbers to identify potential risk numbers that are different from the complaint group numbers;
[0191] Determine the potential risk level of the potential risk number based on the call characteristics between the target risk number and the potential risk number;
[0192] According to the updated communication log and the potential risk level of the potential risk number, the call network map is updated to obtain an updated call network map.
[0193] The risk identification device based on complaining individuals and groups provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and this embodiment will not be described in detail here.
[0194] Figure 5 This is a schematic diagram of the structure of the electronic device provided in this application. Figure 5 As shown, the electronic device 50 provided in this embodiment includes: at least one processor 501 and a memory 502. Optionally, the device 50 also includes a communication component 503. The processor 501, the memory 502 and the communication component 503 are connected via a bus 504.
[0195] In a specific implementation process, at least one processor 501 executes the computer-executable instructions stored in the memory 502, so that at least one processor 501 executes the above method.
[0196] The specific implementation process of the processor 501 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.
[0197] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.
[0198] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.
[0199] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.
[0200] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.
[0201] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.
[0202] The above-mentioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.
[0203] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.
[0204] The division of units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.
[0205] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0206] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0207] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.
[0208] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.
[0209] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.
Claims
1. A risk identification method based on complaining individuals and groups, characterized in that: include: Obtaining a call network graph of the communication log and complaint numbers in a complaint number resource library, wherein the nodes in the call network graph are determined according to the numbers in the communication log, and the edges in the call network graph are determined according to target call features between the numbers; According to the complaint number, determine the number to be identified among the numbers and the risk level of the number to be identified, wherein the number to be identified is a number among the numbers that has a call feature relationship with the complaint number, and the risk level is determined according to the risk complaint level of the complaint number and the call feature; Determine, according to the call network graph, a first associated number having a first side connection with the number to be identified, and a second associated number having a second side connection, wherein the first side connection and the second side connection are different; Determine, according to the risk level, the first associated number, and the second associated number, a target risk number among the to-be-identified numbers that meets the target call feature requirement, and a risk call feature of the target risk number in the communication log; According to the risky call characteristics of the target risk number in the communication log, a complaint group number having a risk association relationship with the target risk number is determined.
2. The method according to claim 1, characterized in that: The acquisition of the call network map of the communication log and the complaint number in the complaint number resource library includes: Obtain communication logs and complaint numbers from the complaint number resource library; Determining call characteristics between the complaint number and other numbers in the communication log; According to the influence of the call features on the risk association between the complaint number and other numbers in the communication log, the call features are screened to determine the target call features; The numbers in the communication log are used as nodes and the target call features between the numbers are used as edges to construct a call network graph.
3. The method according to claim 1, characterized in that The step of determining, according to the complaint number, the number to be identified among the numbers and the risk level of the number to be identified includes: According to the complaint number, the numbers in the communication log are selected to obtain the numbers to be identified that have a call feature relationship with the complaint number; Determine the call characteristics between the complaint number and the number to be identified in the communication log according to the preset call time period, wherein the call characteristics include call frequency, duration, number of days of call, call direction, and connection status; According to the call characteristics and the preset risk weights of the call characteristics, the risk level index of the number to be identified and the risk level of the number to be identified corresponding to the risk level index are determined, wherein the risk weight of call frequency is greater than the risk weight of duration, and the risk weight of duration is greater than the risk weight of call days.
4. The method according to claim 3, characterized in that The determining, according to the call feature and the preset risk weight of the call feature, the risk level index of the number to be identified and the risk level of the number to be identified corresponding to the risk level index comprises: Determine the way to obtain the complaint number, wherein the way to obtain the complaint number includes industry supervision way, consumer rights protection way, and public complaint way; Determine the risk complaint level of the complaint number according to the way the complaint number is obtained; According to the preset risk weights of the call features, weighted summation processing is performed on the call features to obtain a risk level index of the number to be identified; According to the risk complaint level of the complaint number and the risk level index of the number to be identified, the risk level of the number to be identified is determined from a preset number risk level range.
5. The method according to claim 1, characterized in that After determining the numbers to be identified among the numbers and the risk levels of the numbers to be identified according to the complaint number, the method further includes: Determining node attributes in the call network graph according to the risk level of the number to be identified; The call network map is displayed on a visual display interface based on the node attributes, wherein the visual display interface provides a visual tool for further analysis so that the user can identify nodes of different risk levels in the visual display interface.
6. The method according to claim 1, characterized in that The determining, according to the risk level, the first associated number, and the second associated number, a target risk number among the to-be-identified numbers that meets the target call feature requirement, and the risk call feature of the target risk number in the communication log, comprises: Analyzing target call features of the number to be identified according to the first associated number and the second associated number to obtain target call feature distribution information; Determining target call feature requirements according to the target call feature distribution information and the risk level of the number to be identified in the call network map; According to the target call feature requirements, determining a target risk number from the numbers to be identified; According to the risk level of the target risk number, the risk call characteristics of the target risk number in the communication log are determined.
7. The method according to claim 1, characterized in that The determining, according to the risk call characteristics of the target risk number in the communication log, a complaint group number having a risk association relationship with the target risk number includes: Determine the call closeness index between the target risk number and other numbers according to the risk call characteristics of the target risk number in the communication log; According to the call density index, a clustering relationship map between the target risk number and other numbers is obtained; According to the preset risk association requirements, the cluster relationship map is analyzed to determine the complaint group number that has a risk association relationship with the target risk number.
8. The method according to claim 1, characterized in that After determining the complaint group number having a risk association relationship with the target risk number according to the risk call characteristics of the target risk number in the communication log, the method further includes: Get updated communication logs; According to the target risk number, the number in the updated communication log is selected to obtain an initial risk number that meets the preset call characteristic frequency requirement with the target risk number; Screening the initial risk numbers to determine potential risk numbers that are different from the complaint group numbers; Determining the potential risk level of the potential risk number according to the call characteristics between the target risk number and the potential risk number; The call network map is updated according to the updated communication log and the potential risk level of the potential risk number to obtain an updated call network map.
9. A risk identification device based on complaining individuals and groups, characterized in that: include: An acquisition module, used to acquire a call network graph of a communication log and complaint numbers in a complaint number resource library, wherein nodes in the call network graph are determined according to numbers in the communication log, and edges in the call network graph are determined according to target call features between the numbers; A first determination module is used to determine, according to the complaint number, a number to be identified among the numbers and a risk level of the number to be identified, wherein the number to be identified is a number among the numbers that has a call feature relationship with the complaint number, and the risk level is determined according to the risk complaint level of the complaint number and the call feature; A second determination module is used to determine, according to the call network map, a first associated number having a first side connection with the number to be identified and a second associated number having a second side connection, wherein the first side connection and the second side connection are different; A third determination module is used to determine a target risk number that meets the target call feature requirements among the to-be-identified numbers, and a risk call feature of the target risk number in the communication log according to the risk level, the first associated number, and the second associated number; The fourth determination module is used to determine the complaint group number having a risk association relationship with the target risk number according to the risk call characteristics of the target risk number in the communication log.
10. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 8.
11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 8 when executed by a processor.
12. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.
