User risk monitoring method, device, equipment and medium

Abstract

The invention relates to a user risk monitoring method and device, equipment and a medium, and relates to the field of big data processing, and the method comprises the steps: obtaining online transaction data and offline transaction data of a user, determining a transaction place and a first transaction amount of each offline transaction from the offline transaction data, determining a risk level and a second transaction amount of each transaction object from the online transaction data, determining a first behavior feature value based on the transaction place and the first transaction amount, and determining a second behavior feature value based on the risk level and the second transaction amount of the transaction object, and based on the first behavior feature value, the second behavior feature value and a preset dishonesty personnel library, determining a dishonesty risk probability of the user. According to the invention, the accuracy of user dishonesty risk monitoring and prediction can be improved.

Description

Technical Field

[0001] The present application relates to the field of big data processing, and in particular to a user risk monitoring method, device, equipment and medium. Background Art

[0002] With the development of the financial industry and social supply and demand relations, more and more loan products are gradually being enriched, and the number of individuals handling loan business is increasing. Due to the differences in property ownership and consumption habits of different users, the corresponding financial risks are also different. When the user's financial risk is too high, the user's ability to repay the loan may decline. That is, if the user has personal financial problems and is unable to repay on time, the user will stop repaying the loan, the loan will become bad, etc.

[0003] Currently, the default risk of a user is predicted only by comparing the user's offline transaction data and online transaction data with the offline transaction data and online transaction data of the defaulter. However, the accuracy of default risk prediction by only comparing the offline transaction data and online transaction data with the data of the defaulter is low. Summary of the invention

[0004] In order to improve the accuracy of user default risk monitoring and prediction, the present application provides a user risk monitoring method, device, equipment and medium.

[0005] In a first aspect, the present application provides a user risk monitoring method, which adopts the following technical solution: A user risk monitoring method, comprising: Acquire transaction data of the user, wherein the transaction data includes offline transaction data and online transaction data; Determine the transaction location and the first transaction amount of each offline transaction from the offline transaction data, and determine the risk level of each transaction object and the second transaction amount from the online transaction data; Marking the transaction location on a preset map, and determining the activity range of the user based on the transaction location; Determine the number of transactions at each trading location, and identify the target trading location with the most transactions; Determine the distance from the target transaction location to other transaction locations, and determine the average value of the distances; determining a first average of the first transaction amounts and a first sum of the first transaction amounts; Determine a first behavior characteristic value of the user based on the activity range, the distance average, the first average, the first sum, and respective corresponding coefficients; Determine a first number of transaction objects and a second number of target transaction objects whose risk levels reach a preset risk level; determining a first ratio of the second amount to the first amount, and determining an online transaction frequency based on a transaction time of each transaction object; Determine a second sum of the second transaction amounts of all target transaction objects, determine a third sum of all second transaction amounts, and determine a second ratio of the second sum to the third sum; Determine a second behavior characteristic value of the user based on the first ratio, the online transaction frequency, the second ratio, and respective corresponding coefficients; The default risk probability of the user is determined based on the first behavior characteristic value, the second behavior characteristic value and a preset defaulter database.

[0006] By adopting the above technical solution, the user's offline transaction data and online transaction data are obtained for subsequent analysis, and the transaction location and the first transaction amount in the offline transaction data are determined, and the risk level of the transaction object in the online transaction data and the second transaction amount are determined. After the transaction location is determined, it can be marked in the preset map and the user's activity range can be determined. The larger the user's activity range, the wider the user's offline transaction range, and the smaller the user's activity range, the smaller the user's offline transaction range. The user's activity space is relatively stable, and the number of transactions at each transaction location is determined, and the target transaction location with the largest number of transactions is determined. The target transaction location with the largest number of transactions indicates that the user appears at the transaction location more times, which can be used to represent the user's residence or a strongly associated location. The distance from the target transaction location to other transaction locations is determined, and then the distance average is calculated. The larger the distance average is, the larger the user's activity range is. The first average value of the first transaction amount is determined according to the first transaction amount of each offline transaction data. The larger the first average value is, the greater the amount of offline consumption of the user is. The first total is obtained by summing up all the first transaction amounts in the offline transaction data. The larger the first total is, the larger the number of offline transaction amounts of the user is. In summary, the user's activity range, the distance average, the first average value and the first total are all The important factors that affect the user's risk of default are more accurate in determining the first behavior characteristic value through the above four factors. The more the first number of online transaction objects, the more complex the user's online transactions are. The second number of target transaction objects that reach the preset risk level is determined. The more the second number of target transaction objects, the greater the online transaction risk of the user. The first ratio of the second number to the first number is calculated. The larger the first ratio, the greater the online transaction risk of the user, and the greater the corresponding risk of default. The online transaction frequency is determined according to the transaction time of each transaction object. The higher the online transaction frequency, the more frequent the user's online transaction activities are, and the greater the corresponding online financial risk is. The greater the user's default risk, the second total of the second transaction amounts of all target transaction objects is determined, the third total of all second transaction amounts is determined, and the second ratio of the second total to the third total is determined. Since the first ratio, online transaction frequency and the second ratio are all key factors characterizing the user's online transaction characteristics, it is more accurate to determine the second behavior characteristic value by comprehensively determining the above three factors. After determining the first behavior characteristic value and the second behavior characteristic value, the user's default risk probability is obtained by comparing these two characteristic values ​​with the defaulting persons who have already defaulted. This is more accurate than simply using offline transaction data and online transaction data to compare with the defaulting persons to judge the user's default risk.

[0007] In another possible implementation, determining the default risk probability of the user based on the first behavior feature value, the second behavior feature value, and a preset defaulter database includes: Determine a first sub-feature value for each offline transaction, where the first sub-feature value is a ratio of the distance to a first transaction amount; Acquire the transaction time of each offline transaction, and determine the first coordinate of each offline transaction based on the transaction time and the first sub-characteristic value; Determine a second sub-feature value for each online transaction, the feature value being a ratio of the risk level to a second transaction amount; Determine the second coordinate of each online transaction based on the transaction time and the second sub-characteristic value of each transaction object; Mapping the first coordinate and the second coordinate to a preset coordinate system and generating a line graph, wherein the horizontal axis of the preset coordinate system is the transaction time and the vertical axis is the characteristic value; Smoothing the line graph to obtain a smooth curve graph; Calculating the similarity between the smooth curve graph and the smooth curve graph of each dishonest person in the preset dishonest person database; If there is a target defaulter who reaches a preset similarity threshold, the default risk probability of the user is determined based on the similarity of the target defaulter and the target defaulter.

[0008] In another possible implementation, the determining the default risk probability of the user based on the similarity of the target defaulter and the target defaulter includes: Determine the average similarity value and the number of target dishonest persons; The default score of the user is determined based on the similarity average value, the number and the respective corresponding coefficients, and the default score represents the default risk probability of the user.

[0009] In another possible implementation, the method further includes: Determine a target preset probability interval in which the default risk probability lies from a plurality of preset probability intervals, each preset probability interval corresponding to a preset reminder frequency; Output reminder information according to the preset reminder frequency of the target preset probability interval.

[0010] In another possible implementation, the method further includes: Obtaining the default risk probability of the user at every preset time interval and generating a default risk change graph; Determining a change trend of the user's default risk from the default risk change graph, where the change trend includes increasing, decreasing, and unchanged; Determining a reminder frequency adjustment value based on the change trend and the current preset reminder frequency; The current preset reminder frequency is adjusted based on the reminder frequency adjustment value.

[0011] In another possible implementation, determining the reminder frequency adjustment value based on the change trend and the current preset reminder frequency includes: If the change trend is rising or falling, determining the rising rate or falling rate, and determining the reminder frequency adjustment value based on the rising rate or falling rate, the current preset reminder frequency and the respective corresponding coefficients; If the change trend is unchanged, the reminder frequency adjustment value is determined to be 0.

[0012] In a second aspect, the present application provides a user risk monitoring device, which adopts the following technical solution: A user risk monitoring device, comprising: A first acquisition module, used to acquire transaction data of a user, wherein the transaction data includes offline transaction data and online transaction data; A first determination module is used to determine the transaction location and the first transaction amount of each offline transaction from the offline transaction data, and to determine the risk level of each transaction object and the second transaction amount from the online transaction data; A second determination module is used to mark the transaction location on a preset map and determine the activity range of the user based on the transaction location; The third determination module is used to determine the number of transactions at each transaction location and determine the target transaction location with the largest number of transactions; A fourth determination module is used to determine the distance from the target transaction location to other transaction locations, and determine the average value of the distance; A fifth determining module, configured to determine a first average value of the first transaction amount and a first sum of the first transaction amount; a sixth determination module, configured to determine a first behavior characteristic value of the user based on the activity range, the distance average, the first average, the first sum, and respective corresponding coefficients; A seventh determination module, configured to determine a first number of transaction objects and a second number of target transaction objects whose risk levels reach a preset risk level; an eighth determination module, configured to determine a first ratio of the second quantity to the first quantity, and determine an online transaction frequency based on a transaction time of each transaction object; a ninth determination module, configured to determine a second sum of the second transaction amounts of all target transaction objects, determine a third sum of all second transaction amounts, and determine a second ratio of the second sum to the third sum; a tenth determining module, configured to determine a second behavior characteristic value of the user based on the first ratio, the online transaction frequency, the second ratio, and respective corresponding coefficients; An eleventh determination module is used to determine the default risk probability of the user based on the first behavior characteristic value, the second behavior characteristic value and a preset defaulter database.

[0013] By adopting the above technical solution, the first acquisition module obtains the user's offline transaction data and online transaction data for subsequent analysis. The first determination module determines the transaction location and the first transaction amount in the offline transaction data, and determines the risk level of the transaction object in the online transaction data and the second transaction amount. After the second determination module determines the transaction location, it can be marked on the preset map and the user's activity range can be determined. The larger the user's activity range, the wider the user's offline transaction range, and the smaller the user's activity range, the smaller the user's offline transaction range, and the user's activity space is relatively stable. The third determination module determines the number of transactions at each transaction location and determines the target transaction location with the largest number of transactions. The target transaction location with the largest number of transactions The transaction location indicates that the user appears at the transaction location more often, which can be used to indicate the user's residence or a strongly associated location. The fourth determination module determines the distance from the target transaction location to other transaction locations, and then calculates the distance average. The larger the distance average, the larger the user's activity range. The fifth determination module determines the first average of the first transaction amount according to the first transaction amount of each offline transaction data. The larger the first average, the greater the amount of user offline consumption. The first total is obtained by summing up all the first transaction amounts in the offline transaction data. The larger the first total, the larger the number of offline transaction amounts of the user. In summary, the user's activity range, distance average, first average and first total are all factors that affect the user. The sixth determination module determines the first behavior characteristic value more accurately through the above four factors. The more the first number of online transaction objects, the more complex the user's online transactions. The seventh determination module determines the second number of target transaction objects that reach the preset risk level. The more the second number of target transaction objects, the greater the online transaction risk of the user. The eighth determination module calculates the first ratio of the second number to the first number. The larger the first ratio, the greater the online transaction risk of the user, and the greater the corresponding default risk. The online transaction frequency is determined according to the transaction time of each transaction object. The higher the online transaction frequency, the more frequent the user's online transaction activities, the greater the corresponding online financial risk, and the user's default. The greater the risk, the ninth determination module determines the second sum of the second transaction amounts of all target transaction objects, determines the third sum of all second transaction amounts, and determines the second ratio of the second sum to the third sum. Since the first ratio, online transaction frequency and the second ratio are all key factors characterizing the user's online transaction characteristics, the tenth determination module determines the second behavior characteristic value more accurately through the above three factors. After determining the first behavior characteristic value and the second behavior characteristic value, the eleventh determination module compares the two characteristic values ​​with the defaulters who have already defaulted to determine the user's default risk probability, which is more accurate than simply using offline transaction data and online transaction data to compare with defaulters to judge the user's default risk.

[0014] In another possible implementation, when the eleventh determination module determines the default risk probability of the user based on the first behavior feature value, the second behavior feature value, and a preset defaulter database, it is specifically used to: Determine a first sub-feature value for each offline transaction, where the first sub-feature value is a ratio of the distance to a first transaction amount; Acquire the transaction time of each offline transaction, and determine the first coordinate of each offline transaction based on the transaction time and the first sub-characteristic value; Determine a second sub-feature value for each online transaction, the feature value being a ratio of the risk level to a second transaction amount; Determine the second coordinate of each online transaction based on the transaction time and the second sub-characteristic value of each transaction object; Mapping the first coordinate and the second coordinate to a preset coordinate system and generating a line graph, wherein the horizontal axis of the preset coordinate system is the transaction time and the vertical axis is the characteristic value; Smoothing the line graph to obtain a smooth curve graph; Calculating the similarity between the smooth curve graph and the smooth curve graph of each dishonest person in the preset dishonest person database; If there is a target defaulter who reaches a preset similarity threshold, the default risk probability of the user is determined based on the similarity of the target defaulter and the target defaulter.

[0015] In another possible implementation, when the eleventh determination module determines the default risk probability of the user based on the similarity of the target defaulter and the target defaulter, it is specifically used to: Determine the average similarity value and the number of target dishonest persons; The default score of the user is determined based on the similarity average value, the number and the respective corresponding coefficients, and the default score represents the default risk probability of the user.

[0016] In another possible implementation, the user risk monitoring device further includes: An interval determination module, used to determine a target preset probability interval where the default risk probability is located from a plurality of preset probability intervals, each preset probability interval corresponding to a preset reminder frequency; The output module is used to output reminder information according to the preset reminder frequency of the target preset probability interval.

[0017] In another possible implementation, the user risk monitoring device further includes: A generating module, used to obtain the default risk probability of the user at every preset time interval and generate a default risk change graph; A trend determination module, used to determine the default risk change trend of the user from the default risk change graph, where the change trend includes rising, falling and unchanged; An adjustment value determination module, configured to determine a reminder frequency adjustment value based on the change trend and a current preset reminder frequency; An adjustment module is used to adjust the current preset reminder frequency based on the reminder frequency adjustment value.

[0018] In another possible implementation, when the adjustment value determination module determines the reminder frequency adjustment value based on the change trend and the current preset reminder frequency, it is specifically used to: If the change trend is rising or falling, determining the rising rate or falling rate, and determining the reminder frequency adjustment value based on the rising rate or falling rate, the current preset reminder frequency and the respective corresponding coefficients; If the change trend is unchanged, the reminder frequency adjustment value is determined to be 0.

[0019] In a third aspect, the present application provides an electronic device, which adopts the following technical solution: An electronic device, comprising: at least one processor; Memory; At least one application, wherein at least one application is stored in a memory and configured to be executed by at least one processor, and at least one is configured to: execute a user risk monitoring method shown in any possible implementation of the first aspect.

[0020] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium, when the computer program is executed in a computer, causes the computer to execute a user risk monitoring method as described in any one of the first aspects.

[0021] In summary, the present application includes at least one of the following beneficial technical effects: Obtaining the user's offline transaction data and online transaction data is convenient for subsequent analysis, and determining the transaction location and the first transaction amount in the offline transaction data, determining the risk level of the transaction object in the online transaction data and the second transaction amount. After determining the transaction location, it can be marked in the preset map and the user's activity range can be determined. The larger the user's activity range, the wider the user's offline transaction range, and the smaller the user's activity range, the smaller the user's offline transaction range. The user's activity space is relatively stable, and the number of transactions at each transaction location is determined, and the target transaction location with the most transactions is determined. The target transaction location with the most transactions indicates that the user appears at the transaction location more times, which can be used to represent the user's residence or a strongly associated location. Determine the distance from the target transaction location to other transaction locations, and then calculate the average distance. The larger the average distance, the larger the user's activity range. Determine the first average value of the first transaction amount according to the first transaction amount of each offline transaction data. The larger the first average value, the greater the amount of user offline consumption. The sum of all first transaction amounts in the offline transaction data is obtained to obtain the first total. The larger the first total, the larger the number of offline transaction amounts of the user. In summary, the user's activity range, distance average, first average and first total are all factors that affect the user's dishonesty. The first behavior characteristic value is more accurately determined by comprehensively determining the above four factors. The more the first number of online transaction objects, the more complex the user's online transactions are. The second number of target transaction objects that reach the preset risk level is determined. The more the second number of target transaction objects, the greater the online transaction risk of the user. The first ratio of the second number to the first number is calculated. The larger the first ratio, the greater the online transaction risk of the user, and the greater the corresponding risk of default. The online transaction frequency is determined according to the transaction time of each transaction object. The higher the online transaction frequency, the more frequent the user's online transaction activities, the greater the corresponding online financial risk, and the user's default. The greater the risk, the second sum of the second transaction amounts of all target transaction objects is determined, the third sum of all second transaction amounts is determined, and the second ratio of the second sum to the third sum is determined. Since the first ratio, online transaction frequency and the second ratio are all key factors characterizing the user's online transaction characteristics, it is more accurate to determine the second behavior characteristic value by comprehensively determining the above three factors. After determining the first behavior characteristic value and the second behavior characteristic value, the user's default risk probability is obtained by comparing these two characteristic values ​​with the defaulters who have already defaulted. This is more accurate than simply using offline transaction data and online transaction data to compare with defaulters to judge the user's default risk. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flow chart of a user risk monitoring method according to an embodiment of the present application.

[0023] Figure 2It is a structural schematic diagram of a user risk monitoring device according to an embodiment of the present application.

[0024] Figure 3 It is a structural schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] The present application is further described in detail below in conjunction with the accompanying drawings.

[0026] After reading this specification, those skilled in the art may make non-creative modifications to this embodiment as needed, but such modifications are protected by patent law as long as they are within the scope of the claims of this application.

[0027] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0028] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0029] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0030] The embodiment of the present application provides a user risk monitoring method, which is executed by an electronic device, which can be a server or a terminal device, wherein the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to this. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, and the embodiment of the present application does not limit this. Figure 1 As shown, the method includes step S101, step S102, step S103, step S104, step S105, step S106, step S107, step S108, step S109, step S110, step S111 and step S112, wherein, S101, obtaining the user's transaction data.

[0031] Among them, transaction data includes offline transaction data and online transaction data.

[0032] For the embodiment of the present application, when a user uses a bank card or other savings means to consume or trade, each transaction record will be stored in the cloud server. Therefore, the user obtains the user's transaction record within a preset time period from the cloud server to understand the user's consumption and capital flow. There are online transaction data in the transaction record, such as online consumption and capital flow such as transfer and online shopping; there are also offline transaction data, such as offline store payment, shopping in a mall, etc. Specifically, the electronic device can extract the data of each transaction record. According to the transaction code (Transaction Code or Merchant Category Code, MCC) in the transaction record, each transaction will have an MCC code assigned by the issuing bank, which can indicate the type of transaction. For example, a retail store may have a different MCC code, and an online service may have another MCC code. Or determine whether the transaction record belongs to an online transaction or an offline transaction based on the extracted transaction method, such as online shopping payment is an online transaction, and scanning code payment is generally an offline transaction. Therefore, the electronic device can determine the online transaction data and offline transaction data based on the transaction record.

[0033] S102, determining the transaction location and the first transaction amount of each offline transaction from the offline transaction data, and determining the risk level of each transaction object and the second transaction amount from the online transaction data.

[0034] For the embodiment of the present application, since each transaction data has a certain format, after the electronic device determines the offline transaction data, it retrieves the required data object and extracts each offline transaction data to obtain the merchant name in each offline transaction data. Specifically, the electronic device can extract the merchant name, such as "xxx store", and then the electronic device calls the map software and searches to obtain the geographical location of "xxx store", that is, the transaction location. Similarly, the electronic device extracts the transaction amount from each offline transaction data, that is, the first transaction amount.

[0035] The electronic device extracts the transaction object in each online transaction data, such as the account name of the transferor, the name of the transfer company, etc. Different transaction objects correspond to different risk levels. For example, the risk level of personal transfers is different from that of corporate transfers. If there are other transaction objects, there may be other risk levels. The same electronic device can extract the second transaction amount of each online transaction data.

[0036] S103, marking the transaction location on a preset map, and determining the user's activity range based on the transaction location.

[0037] For the embodiment of the present application, after the electronic device determines the transaction location, it can be marked in the preset map, and the preset map can be a map of China or a world map. The electronic device uses a machine learning model (such as a neural network, a support vector machine, etc.) to classify the points according to the annotations in the preset map, and identifies which points are closer to the outline, and the identified outline is the activity range. The larger the user's activity range, the wider the user's offline transaction range and the higher the user's mobility. The smaller the user's activity range, the smaller the user's offline transaction range, the smaller the user's activity range, the more stable the user's activity space, the lower and more stable the user's mobility, and the lower the risk of offline transactions. After the electronic device determines the activity range of the user, it can determine the area of ​​the activity range.

[0038] S104, determining the number of transactions at each transaction location, and determining the target transaction location with the largest number of transactions.

[0039] For the embodiment of the present application, the electronic device counts the number of transactions at each transaction location based on offline transaction data, thereby determining the number of transactions at each transaction location and determining the target transaction location with the most transactions. The target transaction location with the most transactions indicates that the user appears at the transaction location more often, indicating that the user consumes more frequently at the transaction location and it is more convenient for the user to go to the transaction location, which can be used to represent the user's residence or a strongly associated location.

[0040] S105, determining the distance from the target transaction location to other transaction locations, and determining the average distance of the distances.

[0041] For the embodiment of the present application, after the electronic device determines the target transaction location, it can calculate the distance from the target transaction location to other transaction locations according to the distance formula between two points, and then calculate the average distance using the average calculation formula. The larger the average distance, the larger the user's activity range, which also means that the risk of offline transactions is greater.

[0042] S106, determining a first average value of the first transaction amount and a first sum of the first transaction amount.

[0043] For the embodiment of the present application, the electronic device determines the first average value of the first transaction amount based on the first transaction amount of each offline transaction data and the average value calculation formula. The larger the first average value, the greater the amount of offline consumption of the user, and the greater the risk of default of the user. The electronic device sums all the first transaction amounts in the offline transaction data to obtain the first sum. The larger the first sum, the greater the number of offline transaction amounts of the user, and the greater the risk of default.

[0044] S107, determining a first behavior characteristic value of the user based on the activity range, the distance average, the first average, the first sum, and the respective corresponding coefficients.

[0045] For the embodiments of the present application, in summary, the user's activity range, distance average, first average and first sum are all important factors that affect the user's offline transaction risk, and the above four factors have different degrees of influence on offline transaction risk. The staff can set different coefficients for the activity range, distance average, first average and first sum in advance and store them in the electronic device. After the electronic device determines the activity range, distance average, first average and first sum, it can determine the user's first behavior feature value by weighted calculation based on the corresponding coefficients. It is more accurate to comprehensively determine the first behavior feature value through the above four factors.

[0046] S108, determining a first number of transaction objects and a second number of target transaction objects whose risk levels reach a preset risk level.

[0047] For the embodiment of the present application, the greater the first number of transaction objects, the more complex the user's online transactions are, and the larger the corresponding second behavior characteristic value is. Different types of transaction objects have different risk levels. For example, the risk level of a transaction object that is an enterprise is lower than the risk level of a transaction object that is an individual. The risk level of a transaction object that has had transaction violations is higher than the risk level corresponding to a transaction without transaction violations. The risk level of an overseas transaction object is higher than the risk level of a domestic transaction object. Therefore, the staff can set different risk levels according to different transaction objects. The preset risk level serves as a dividing point for whether the transaction risk is too high. After the electronic device determines each online transaction object, it compares the risk level of each transaction object with the preset risk level, thereby determining the target transaction object that reaches the preset risk level.

[0048] S109, determining a first ratio of the second quantity to the first quantity, and determining an online transaction frequency based on the transaction time of each transaction object.

[0049] For the embodiment of the present application, the more the second number of target transaction objects, the greater the online transaction risk of the user. Therefore, the electronic device calculates the first ratio of the second number to the first number. The larger the first ratio, the greater the online transaction risk of the user. Each transaction generated by the user corresponds to a transaction time. Therefore, the electronic device can extract the transaction time from the online transaction data, and then the electronic device determines the time interval between two adjacent online transactions in chronological order. The online transaction frequency can be determined based on the time interval and the preset time period. The higher the online transaction frequency, the more frequent the user's online transaction activities are, and the greater the risk of the user's default.

[0050] S110, determining a second sum of the second transaction amounts of all target transaction objects, determining a third sum of all second transaction amounts, and determining a second ratio of the second sum to the third sum.

[0051] For the embodiment of the present application, the second transaction amounts of all target transaction objects of the electronic device are summed to obtain a second total. The larger the second total, the more transaction amounts the user has with transaction objects with higher risk levels, and the higher the risk of user default. The electronic device sums the second transaction amounts of all online transaction data to obtain the total amount of all online transactions of the user within a preset time period, i.e., the third total. After the electronic device determines the second total and the third total, it can determine a second ratio of the second total to the third total. The larger the second ratio, the larger the proportion of the amount of online transactions conducted by the user with transaction objects with higher risk levels, and the higher the risk of user default.

[0052] S111, determining a second behavior characteristic value of the user based on the first ratio, the online transaction frequency, the second ratio, and their corresponding coefficients.

[0053] For the embodiment of the present application, since the first ratio, online transaction frequency and second ratio are all key factors that characterize the online transaction risk of the user, and the above three have different degrees of influence on the online transaction risk. The staff can set different coefficients for the first ratio, online transaction frequency and second ratio in advance and store them in the electronic device. After the electronic device determines the first ratio, online transaction frequency and second ratio, it can perform weighted calculation according to the corresponding coefficients to determine the second behavior characteristic value of the user. It is more accurate to comprehensively determine the second behavior characteristic value through the above three factors.

[0054] S112, determining the default risk probability of the user based on the first behavior characteristic value, the second behavior characteristic value, and a preset defaulter database.

[0055] For the embodiment of the present application, after the electronic device determines the first behavior characteristic value and the second behavior characteristic value in the above manner, the user's default risk probability is obtained by comparing the two characteristic values ​​with the defaulters in the preset defaulter database, which is more accurate than simply using offline transaction data and online transaction data to compare with the offline transaction data and online transaction data of the defaulters.

[0056] In a possible implementation of the embodiment of the present application, in step S112, determining the default risk probability of the user based on the first behavior feature value, the second behavior feature value, and a preset defaulter database, specifically includes step S1121 (not shown in the figure), step S1122 (not shown in the figure), step S1123 (not shown in the figure), step S1124 (not shown in the figure), step S1125 (not shown in the figure), step S1126 (not shown in the figure), step S1127 (not shown in the figure), and step S1128 (not shown in the figure), wherein: S1121, determining the first sub-feature value of each offline transaction.

[0057] The first sub-characteristic value is the ratio of the distance to the first transaction amount.

[0058] For the embodiment of the present application, the electronic device extracts the first transaction amount of each offline transaction of the user and the distance between the offline transaction location and the target transaction location. The distance and the first transaction amount are both key features of an offline transaction. Therefore, the electronic device determines the ratio of the distance to the first transaction amount and uses the ratio as a unique characteristic value corresponding to an offline transaction, namely, the first sub-characteristic value.

[0059] S1122, obtaining the transaction time of each offline transaction, and determining the first coordinate of each offline transaction based on the transaction time and the first sub-eigenvalue.

[0060] For the embodiment of the present application, each offline transaction corresponds to a time point when the transaction occurs, so the electronic device can determine the transaction time of each offline transaction from the offline transaction record. The electronic device can determine the first coordinate of each offline transaction based on the transaction time and the first sub-feature value of each offline transaction.

[0061] S1123, determining a second sub-characteristic value for each online transaction.

[0062] The characteristic value is the ratio of the risk level to the second transaction amount.

[0063] For the embodiment of the present application, the electronic device extracts the second transaction amount of each online transaction of the user and the risk level of the transaction object of each online transaction. The risk level of the transaction object and the second transaction amount are both key features of an online transaction. Therefore, the electronic device determines the ratio of the risk level of the transaction object of each online transaction to the second transaction amount, and uses the ratio as a unique characteristic value corresponding to an online transaction, that is, the second sub-characteristic value.

[0064] S1124, determining the second coordinate of each online transaction based on the transaction time and the second sub-feature value of each transaction object.

[0065] For the embodiment of the present application, the electronic device can determine the second coordinate of each online transaction based on the time of each online transaction and the second sub-feature value. After the electronic device determines the first coordinate and the second coordinate, since the first coordinate and the second coordinate are both the corresponding relationship between the transaction time and the sub-feature value, the electronic device can perform a fusion analysis of the offline transaction and the online transaction based on the first coordinate and the second coordinate.

[0066] S1125, mapping the first coordinate and the second coordinate to a preset coordinate system and generating a line graph.

[0067] Among them, the horizontal axis of the preset coordinate system is the transaction time, and the vertical axis is the characteristic value.

[0068] For the embodiment of the present application, the electronic device maps each first coordinate and each second coordinate to a preset coordinate system, thereby integrating offline transactions and online transactions into the same coordinate system. The electronic device connects the first coordinate and the second coordinate in sequence to generate a line graph, which represents the specific trends and performances of all the user's offline transactions and online transactions.

[0069] S1126, smoothing the line graph to obtain a smooth curve graph.

[0070] For the embodiments of the present application, the electronic device may smooth the discount graph by interpolation, Fourier transform, wavelet transform, etc., or may smooth the line graph by an origin plug-in to obtain a smooth curve graph.

[0071] S1127, calculating the similarity between the smooth curve graph and the smooth curve graph of each dishonest person in the preset dishonest person database.

[0072] For the embodiment of the present application, each dishonest person in the preset dishonest person database also corresponds to a smooth curve graph obtained based on their offline transactions and online transactions. Similarly, the smooth curve graph of the dishonest person can also be determined in the above manner. The electronic device can input the smooth curve graph of the user and the smooth curve graph of each dishonest person into the trained network model for similarity calculation. First, the smooth curve graph of the user and the smooth curve graph of each dishonest person are converted into vectors, and then the network model obtains the similarity between the user and each dishonest person by calculating the cosine distance, Euclidean distance, etc. of the vector. The higher the similarity between the smooth curve graph of the user and a dishonest person, the closer the transaction performance of the user is to the transaction performance of the dishonest person, and the greater the probability of dishonesty risk of the user.

[0073] S1128, if there is a target defaulter who reaches a preset similarity threshold, the default risk probability of the user is determined based on the similarity of the target defaulter and the target defaulter.

[0074] For the embodiment of the present application, a preset similarity threshold is used as a dividing point for whether the similarity reaches a higher standard. The electronic device compares the similarity corresponding to each defaulter with the preset similarity threshold to determine whether there is a target defaulter who reaches the preset similarity threshold. If there is a target defaulter, the electronic device performs a comprehensive analysis based on the similarity of the target defaulter and the attributes of the target defaulter to accurately determine the default risk probability of the user.

[0075] In a possible implementation of the embodiment of the present application, in step S1128, determining the default risk probability of the user based on the similarity of the target defaulter and the target defaulter specifically includes step 1 and step 2, wherein: Step 1: Determine the average similarity value and the number of target defaulters.

[0076] Step 2: Determine the user's dishonesty score based on the similarity average, quantity, and respective corresponding coefficients.

[0077] Among them, the default score represents the user's default risk probability.

[0078] For the embodiment of the present application, after the electronic device determines the target dishonest person, the number of the target dishonest person is counted to obtain the number of the target dishonest person. The more the number of the target dishonest persons, the higher the general similarity between the transaction performance of the user and the dishonest person, and the greater the probability of dishonest risk of the user. The electronic device calculates the average value of the similarity of the universal target dishonest person through the average value calculation formula to obtain the similarity average value. The larger the similarity average value, the higher the overall similarity between the user and these target dishonest persons, and the greater the probability of dishonest risk of the user. In summary, the similarity average value and the number of target dishonest persons are both important factors affecting the probability of dishonest risk of the user. Therefore, the staff sets different coefficients for the similarity average value and the number of target dishonest persons and stores them in the local storage medium of the electronic device. After the electronic device determines the similarity average value and the number of target dishonest persons, the corresponding coefficients are called for weighted calculation to obtain the dishonest score of the user. The dishonest score is the probability of dishonest risk of the user. The dishonest risk probability of the user obtained by the electronic device through comprehensive analysis of the similarity average value and the number of target dishonest persons is more accurate.

[0079] In a possible implementation manner of the embodiment of the present application, the method further includes step three and step four, wherein: Step three: determine the target preset probability interval where the default risk probability lies from multiple preset probability intervals.

[0080] Each preset probability interval corresponds to a preset reminder frequency.

[0081] Step 4: Output reminder information according to the preset reminder frequency of the target preset probability interval.

[0082] For the embodiment of the present application, after the electronic device determines the default risk probability of the user, it compares the default risk probability with each preset probability interval, thereby determining the target preset probability interval where the default risk probability is located, each preset probability interval corresponds to a preset reminder frequency, and the preset reminder frequency corresponding to each preset probability interval is set by the staff. The greater the probability in the preset probability interval, the higher the corresponding preset reminder frequency. Therefore, the electronic device determines the target preset probability interval where the default risk probability of the user is located, and outputs reminder information according to the preset reminder frequency corresponding to the target preset probability interval, so that the user can know his own default risk probability more timely. The higher the default risk probability, the higher the corresponding reminder frequency. High-frequency reminders to users enable users to take measures against their own default risks, thereby reducing the user's default risk. The reminder information can be a text message reminder message sent by the electronic device to the user's terminal device such as a mobile phone, "Your default risk is xxx, please pay attention", or a text message "Your default risk is xxx, please pay attention" pushed to the user's terminal device through an APP.

[0083] In a possible implementation manner of the embodiment of the present application, the method further includes step five, step six, step seven and step eight, wherein: Step 5: Obtain the user's default risk probability at every preset time interval and generate a default risk change graph.

[0084] For the embodiment of the present application, after the electronic device calculates the default risk probability of the user each time, the default risk probability is stored each time, so as to facilitate the electronic device to obtain, and the preset time interval can be three days, five days, or other time periods. Taking three days as an example, the electronic device obtains the default risk probability of the user every three days, and maps the default risk probability to the coordinate system to generate a default risk change graph.

[0085] Step six, determine the user's default risk change trend from the default risk change graph.

[0086] Among them, the changing trends include rising, falling and unchanged.

[0087] For the embodiment of the present application, the electronic device connects the coordinate points in the default risk change graph to obtain a default risk line graph, and then performs linear fitting on the line graph to obtain a linear function, and then the electronic device determines the slope of the linear function, which characterizes the default risk change trend of the user. A positive slope indicates that the default risk change trend is increasing, a negative slope indicates that the default risk change trend is decreasing, and a slope of 0 indicates that the default risk change trend is unchanged.

[0088] Step seven: determining a reminder frequency adjustment value based on the change trend and the current preset reminder frequency.

[0089] For the embodiment of the present application, after the electronic device determines the changing trend, since the changing trend of the user's risk of default is dynamically changing, the current preset reminder frequency needs to be adjusted. The electronic device determines a reminder frequency adjustment value according to the specific circumstances of the changing trend and the current preset reminder frequency.

[0090] Step eight, adjusting the current preset reminder frequency based on the reminder frequency adjustment value.

[0091] For the embodiment of the present application, after the electronic device determines the reminder frequency adjustment value, it can adjust the current preset reminder frequency according to the adjustment value, so that the reminder frequency is more in line with the user's current situation. If the change trend is rising, it means that the reminder frequency needs to be increased. Therefore, after the electronic device determines the reminder frequency adjustment value, it can add the reminder frequency adjustment value to the current preset reminder frequency. If the change trend is falling, it means that the user's risk of default is reduced and a higher reminder frequency is no longer needed. Therefore, after the electronic device determines the reminder frequency adjustment value, it can subtract the reminder frequency adjustment value from the current preset reminder frequency. If the change trend is unchanged, the electronic device does not adjust the current preset reminder frequency.

[0092] In a possible implementation of the embodiment of the present application, in step 7, the reminder frequency adjustment value is determined based on the change trend and the current preset reminder frequency, specifically including step Sa (not shown in the figure) and step Sb (not shown in the figure), wherein: Sa, if the change trend is rising or falling, the rising rate or the falling rate is determined, and the reminder frequency adjustment value is determined based on the rising rate or the falling rate, the current preset reminder frequency and the respective corresponding coefficients.

[0093] Sb, if the change trend is unchanged, the reminder frequency adjustment value is determined to be 0.

[0094] For the embodiment of the present application, if the trend of change is rising or falling, it means that the current preset reminder frequency needs to be adjusted, and the electronic device determines the slope of the linear function in step six, and the slope represents the rising rate or the falling rate. The rising rate or the falling rate plus the current preset reminder frequency are specific factors that affect the size of the reminder frequency adjustment value. Therefore, the staff sets the corresponding coefficients for the rising rate or the falling rate and the current preset reminder frequency and stores them in the electronic device. Then, the electronic device determines the rising rate or the falling rate and the current preset reminder frequency, and then calls the corresponding coefficients for weighted calculation to obtain the reminder frequency adjustment value. If the trend of change is unchanged, the reminder frequency adjustment value is determined to be 0, that is, the current preset reminder frequency is not adjusted. It is more accurate to comprehensively analyze and determine the reminder frequency adjustment value through the rising rate or the falling rate and the current preset reminder frequency.

[0095] In a possible implementation of the embodiment of the present application, step S106 further includes step S107 (not shown in the figure) and step S108 (not shown in the figure), wherein: S107, determining a reminder frequency based on the financial risk level.

[0096] S108: Output reminder information according to the reminder frequency.

[0097] For the embodiment of the present application, the staff can set different reminder frequencies for different financial risk levels in advance and store them in the electronic device. It is necessary to know that the higher the financial risk level, the higher the corresponding reminder frequency. If three financial risk levels are set, namely level one, level two and level three, the reminder frequency corresponding to level one is once a month, the reminder frequency corresponding to level two is twice a month, and the reminder frequency corresponding to level three is three times a month. After the electronic device determines the financial risk level of the user, it can find the corresponding reminder frequency according to the financial risk level. Assuming that the user's financial risk level is level three, it means that the user's repayment risk is relatively large and it is necessary to remind the user to repay on time more frequently. Then the electronic device determines that the user's reminder frequency is three times a month, and sends reminder information to the user's terminal device such as a mobile phone, a tablet computer, etc. at a frequency of three times a month. It can be a text message of "Please repay on time" or a reminder message of "Please repay on time" pushed through an APP. Determine the appropriate reminder frequency by the financial risk level, and output the reminder information according to the reminder frequency, thereby improving the reminder effect for the user.

[0098] The above embodiment introduces a user risk monitoring method from the perspective of method flow, and the following embodiment introduces a user risk monitoring device from the perspective of a virtual module or a virtual unit. Please refer to the following embodiment for details.

[0099] The present application embodiment provides a user risk monitoring device 20, such as Figure 2 As shown, a user risk monitoring device 20 may specifically include: The first acquisition module 201 is used to acquire the transaction data of the user, where the transaction data includes offline transaction data and online transaction data; A first determination module 202 is used to determine the transaction location and the first transaction amount of each offline transaction from the offline transaction data, and to determine the risk level of each transaction object and the second transaction amount from the online transaction data; The second determination module 203 is used to mark the transaction location on a preset map and determine the user's activity range based on the transaction location; The third determination module 204 is used to determine the number of transactions at each transaction location, and determine the target transaction location with the largest number of transactions; The fourth determination module 205 is used to determine the distance from the target transaction location to other transaction locations, and determine the average distance of the distances; A fifth determining module 206, configured to determine a first average value of the first transaction amount and a first sum of the first transaction amount; A sixth determination module 207, configured to determine a first behavior characteristic value of the user based on the activity range, the distance average, the first average, the first sum, and the respective corresponding coefficients; A seventh determination module 208, configured to determine a first number of transaction objects and a second number of target transaction objects whose risk levels reach a preset risk level; An eighth determination module 209, configured to determine a first ratio of the second quantity to the first quantity, and determine an online transaction frequency based on a transaction time of each transaction object; A ninth determination module 210, configured to determine a second sum of the second transaction amounts of all target transaction objects, determine a third sum of all second transaction amounts, and determine a second ratio of the second sum to the third sum; A tenth determining module 211, configured to determine a second behavior characteristic value of the user based on the first ratio, the online transaction frequency, the second ratio, and the respective corresponding coefficients; The eleventh determination module 212 is used to determine the default risk probability of the user based on the first behavior characteristic value, the second behavior characteristic value and a preset defaulter database.

[0100] The embodiment of the present application discloses a user risk monitoring device 20, wherein the first acquisition module 201 acquires the user's offline transaction data and online transaction data for subsequent analysis, the first determination module 202 determines the transaction location and the first transaction amount in the offline transaction data, and determines the risk level of the transaction object in the online transaction data and the second transaction amount, the second determination module 203 determines the transaction location and can mark it on a preset map and determine the user's activity range after determining the transaction location, the larger the user's activity range is, the wider the user's offline transaction range is, the smaller the user's activity range is, the smaller the user's offline transaction range is, and the user's activity space is relatively stable, the third determination module 204 determines the number of transactions at each transaction location and determines The target transaction location with the most transactions is determined, and the target transaction location with the most transactions indicates that the user appears at the transaction location more often, which can be used to indicate the user's residence or a strongly associated location. The fourth determination module 205 determines the distance from the target transaction location to other transaction locations, and then calculates the average distance value. The larger the average distance value, the larger the user's activity range. The fifth determination module 206 determines the first average value of the first transaction amount according to the first transaction amount of each offline transaction data. The larger the first average value, the larger the amount flow of the user's offline consumption. All the first transaction amounts in the offline transaction data are summed to obtain a first total. The larger the first total, the larger the number of offline transaction amounts of the user. In summary, the user The activity range, the average distance, the first average and the first sum are all important factors that affect the user's default risk. The sixth determination module 207 determines the first behavior characteristic value more accurately through the above four factors. The more the first number of online transaction objects, the more complex the user's online transaction. The seventh determination module 208 determines the second number of target transaction objects that reach the preset risk level. The more the second number of target transaction objects, the greater the online transaction risk of the user. The eighth determination module 209 calculates the first ratio of the second number to the first number. The larger the first ratio, the greater the online transaction risk of the user, and the greater the corresponding default risk. The online transaction frequency is determined according to the transaction time of each transaction object. The higher the transaction frequency, the more frequent the user's online transaction activities, the greater the corresponding online financial risk, and the greater the user's risk of default. The ninth determination module 210 determines the second sum of the second transaction amounts of all target transaction objects, determines the third sum of all second transaction amounts, and determines the second ratio of the second sum to the third sum. Since the first ratio, the online transaction frequency, and the second ratio are all key factors that characterize the user's online transaction characteristics, the tenth determination module 211 determines the second behavior characteristic value more accurately through the above three factors. After determining the first behavior characteristic value and the second behavior characteristic value, the eleventh determination module 212 compares the two characteristic values ​​with the defaulting persons who have already defaulted to determine the user's default risk probability.Compared with simply using offline transaction data and online transaction data to compare with dishonest persons to judge the dishonest risk of users, it is more accurate.

[0101] In a possible implementation of the embodiment of the present application, when the eleventh determination module 212 determines the default risk probability of the user based on the first behavior feature value, the second behavior feature value, and the preset defaulter database, it is specifically used to: Determine a first sub-feature value for each offline transaction, where the first sub-feature value is a ratio of the distance to the first transaction amount; Obtaining the transaction time of each offline transaction, and determining the first coordinate of each offline transaction based on the transaction time and the first sub-eigenvalue; Determine a second sub-feature value for each online transaction, where the feature value is a ratio of the risk level to the second transaction amount; Determine the second coordinate of each online transaction based on the transaction time and the second sub-characteristic value of each transaction object; Mapping the first coordinate and the second coordinate to a preset coordinate system and generating a line graph, wherein the horizontal axis of the preset coordinate system is the transaction time and the vertical axis is the characteristic value; Smoothing the line graph to obtain a smooth curve graph; Calculate the similarity between the smooth curve graph and the smooth curve graph of each dishonest person in the preset dishonest person database; If there is a target defaulter who reaches a preset similarity threshold, the default risk probability of the user is determined based on the similarity of the target defaulter and the target defaulter.

[0102] In a possible implementation of the embodiment of the present application, when the eleventh determination module 212 determines the default risk probability of the user based on the similarity of the target defaulter and the target defaulter, it is specifically used to: Determine the average similarity and the number of target defaulters; The user's dishonesty score is determined based on the similarity average value, number and respective corresponding coefficients, and the dishonesty score represents the dishonesty risk probability of the user.

[0103] In a possible implementation of the embodiment of the present application, a user risk monitoring device 20 further includes: An interval determination module is used to determine a target preset probability interval where the default risk probability is located from a plurality of preset probability intervals, each preset probability interval corresponding to a preset reminder frequency; The output module is used to output reminder information according to the preset reminder frequency of the target preset probability interval.

[0104] In a possible implementation of the embodiment of the present application, a user risk monitoring device 20 further includes: A generation module, used to obtain the default risk probability of the user at every preset time interval and generate a default risk change graph; A trend determination module is used to determine the change trend of the user's default risk from the default risk change graph, and the change trend includes rising, falling and unchanged; An adjustment value determination module, used to determine the reminder frequency adjustment value based on the change trend and the current preset reminder frequency; The adjustment module is used to adjust the current preset reminder frequency based on the reminder frequency adjustment value.

[0105] In a possible implementation manner of the embodiment of the present application, when the adjustment value determination module determines the reminder frequency adjustment value based on the change trend and the current preset reminder frequency, it is specifically used to: If the change trend is rising or falling, then determine the rising rate or falling rate, and determine the reminder frequency adjustment value based on the rising rate or falling rate, the current preset reminder frequency, and the corresponding coefficients; If the change trend is unchanged, the reminder frequency adjustment value is determined to be 0.

[0106] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the user risk monitoring device 20 described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0107] An electronic device is provided in an embodiment of the present application, such as Figure 3 As shown, Figure 3 The electronic device 30 shown includes: a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, such as through a bus 302. Optionally, the electronic device 30 may also include a transceiver 304. It should be noted that in actual applications, the transceiver 304 is not limited to one, and the structure of the electronic device 30 does not constitute a limitation on the embodiments of the present application.

[0108] The processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0109] The bus 302 may include a path to transmit information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but it does not mean that there is only one bus or only one type of bus.

[0110] The memory 303 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0111] The memory 303 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the contents shown in the above method embodiment.

[0112] The electronic devices include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and fixed terminals such as digital TVs, desktop computers, etc. It can also be a server, etc. Figure 3 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0113] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment. Compared with the related art, in the embodiment of the present application, the offline transaction data and online transaction data of the user are obtained to facilitate subsequent analysis, and the transaction location and the first transaction amount in the offline transaction data are determined, and the risk level and the second transaction amount of the transaction object in the online transaction data are determined. After the transaction location is determined, it can be marked in the preset map and the user's activity range can be determined. The larger the user's activity range, the wider the user's offline transaction range, and the smaller the user's activity range, the smaller the user's offline transaction range. The user's activity space is relatively stable, the number of transactions at each transaction location is determined, and the target transaction location with the largest number of transactions is determined. The target transaction location with the largest number of transactions indicates that the user appears at the transaction location more times, which can be used to represent the user's residence or a strongly associated location. The distance from the target transaction location to other transaction locations is determined, and then the distance average is calculated. The larger the distance average is, the larger the user's activity range is. The first average value of the first transaction amount is determined according to the first transaction amount of each offline transaction data. The larger the first average value is, the greater the amount of offline consumption of the user is. The first total is obtained by summing up all the first transaction amounts in the offline transaction data. The larger the first total is, the larger the number of offline transaction amounts of the user is. In summary, the user's activity range, the distance average, the first average value and the first total are all It is an important factor affecting the user's risk of default. The first behavior characteristic value is more accurately determined by comprehensively determining the above four factors. The more the first number of online transaction objects, the more complex the user's online transactions are. The second number of target transaction objects that reach the preset risk level is determined. The more the second number of target transaction objects, the greater the user's online transaction risk is. The first ratio of the second number to the first number is calculated. The larger the first ratio is, the greater the user's online transaction risk is, and the corresponding default risk is greater. The online transaction frequency is determined according to the transaction time of each transaction object. The higher the online transaction frequency is, the more frequent the user's online transaction activities are, and the corresponding online financial risk is greater. The greater the user's risk of default, the second total of the second transaction amounts of all target transaction objects is determined, the third total of all second transaction amounts is determined, and the second ratio of the second total to the third total is determined. Since the first ratio, online transaction frequency and the second ratio are all key factors characterizing the user's online transaction characteristics, it is more accurate to determine the second behavior characteristic value by comprehensively determining the above three factors. After determining the first behavior characteristic value and the second behavior characteristic value, the user's default risk probability is obtained by comparing these two characteristic values ​​with the defaulting persons who have already defaulted. This is more accurate than simply using offline transaction data and online transaction data to compare with the defaulting persons to judge the user's default risk.

[0114] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0115] The above description is only a partial implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A user risk monitoring method, characterized in that: include: Acquire transaction data of the user, wherein the transaction data includes offline transaction data and online transaction data; Determine the transaction location and the first transaction amount of each offline transaction from the offline transaction data, and determine the risk level of each transaction object and the second transaction amount from the online transaction data; Marking the transaction location on a preset map, and determining the activity range of the user based on the transaction location; Determine the number of transactions at each trading location, and identify the target trading location with the most transactions; Determine the distance from the target transaction location to other transaction locations, and determine the average value of the distances; determining a first average of the first transaction amounts and a first sum of the first transaction amounts; Determine a first behavior characteristic value of the user based on the activity range, the distance average, the first average, the first sum, and respective corresponding coefficients; Determine a first number of transaction objects and a second number of target transaction objects whose risk levels reach a preset risk level; determining a first ratio of the second amount to the first amount, and determining an online transaction frequency based on a transaction time of each transaction object; Determine a second sum of the second transaction amounts of all target transaction objects, determine a third sum of all second transaction amounts, and determine a second ratio of the second sum to the third sum; Determine a second behavior characteristic value of the user based on the first ratio, the online transaction frequency, the second ratio, and respective corresponding coefficients; The default risk probability of the user is determined based on the first behavior characteristic value, the second behavior characteristic value and a preset defaulter database.

2. A user risk monitoring method according to claim 1, characterized in that: The determining the default risk probability of the user based on the first behavior feature value, the second behavior feature value and a preset defaulter database includes: Determine a first sub-feature value for each offline transaction, where the first sub-feature value is a ratio of the distance to a first transaction amount; Acquire the transaction time of each offline transaction, and determine the first coordinate of each offline transaction based on the transaction time and the first sub-characteristic value; Determine a second sub-feature value for each online transaction, the feature value being a ratio of the risk level to a second transaction amount; Determine the second coordinate of each online transaction based on the transaction time and the second sub-characteristic value of each transaction object; Mapping the first coordinate and the second coordinate to a preset coordinate system and generating a line graph, wherein the horizontal axis of the preset coordinate system is the transaction time and the vertical axis is the characteristic value; Smoothing the line graph to obtain a smooth curve graph; Calculating the similarity between the smooth curve graph and the smooth curve graph of each dishonest person in the preset dishonest person database; If there is a target defaulter who reaches a preset similarity threshold, the default risk probability of the user is determined based on the similarity of the target defaulter and the target defaulter.

3. A user risk monitoring method according to claim 2, characterized in that: The determining of the default risk probability of the user based on the similarity of the target defaulter and the target defaulter includes: Determine the average similarity value and the number of target dishonest persons; The default score of the user is determined based on the similarity average value, the number and the respective corresponding coefficients, and the default score represents the default risk probability of the user.

4. A user risk monitoring method according to claim 1, characterized in that: The method further comprises: Determine a target preset probability interval in which the default risk probability lies from a plurality of preset probability intervals, each preset probability interval corresponding to a preset reminder frequency; Output reminder information according to the preset reminder frequency of the target preset probability interval.

5. A user risk monitoring method according to claim 4, characterized in that: The method further comprises: Obtaining the default risk probability of the user at every preset time interval and generating a default risk change graph; Determining a change trend of the user's default risk from the default risk change graph, where the change trend includes increasing, decreasing, and unchanged; Determining a reminder frequency adjustment value based on the change trend and the current preset reminder frequency; The current preset reminder frequency is adjusted based on the reminder frequency adjustment value.

6. A user risk monitoring method according to claim 5, characterized in that: The determining the reminder frequency adjustment value based on the change trend and the current preset reminder frequency includes: If the change trend is rising or falling, determining the rising rate or falling rate, and determining the reminder frequency adjustment value based on the rising rate or falling rate, the current preset reminder frequency and the respective corresponding coefficients; If the change trend is unchanged, the reminder frequency adjustment value is determined to be 0.

7. A user risk monitoring device, characterized in that: include: A first acquisition module, used to acquire transaction data of a user, wherein the transaction data includes offline transaction data and online transaction data; A first determination module is used to determine the transaction location and the first transaction amount of each offline transaction from the offline transaction data, and to determine the risk level of each transaction object and the second transaction amount from the online transaction data; A second determination module is used to mark the transaction location on a preset map and determine the activity range of the user based on the transaction location; The third determination module is used to determine the number of transactions at each transaction location and determine the target transaction location with the largest number of transactions; A fourth determination module is used to determine the distance from the target transaction location to other transaction locations, and determine the average value of the distance; A fifth determining module, configured to determine a first average value of the first transaction amount and a first sum of the first transaction amount; a sixth determination module, configured to determine a first behavior characteristic value of the user based on the activity range, the distance average, the first average, the first sum, and respective corresponding coefficients; A seventh determination module, configured to determine a first number of transaction objects and a second number of target transaction objects whose risk levels reach a preset risk level; an eighth determination module, configured to determine a first ratio of the second quantity to the first quantity, and determine an online transaction frequency based on a transaction time of each transaction object; a ninth determination module, configured to determine a second sum of the second transaction amounts of all target transaction objects, determine a third sum of all second transaction amounts, and determine a second ratio of the second sum to the third sum; a tenth determining module, configured to determine a second behavior characteristic value of the user based on the first ratio, the online transaction frequency, the second ratio, and respective corresponding coefficients; An eleventh determination module is used to determine the default risk probability of the user based on the first behavior characteristic value, the second behavior characteristic value and a preset defaulter database.

8. A user risk monitoring device according to claim 7, characterized in that: When the eleventh determination module determines the default risk probability of the user based on the first behavior characteristic value, the second behavior characteristic value and the preset defaulter database, it is specifically used to: Determine a first sub-feature value for each offline transaction, where the first sub-feature value is a ratio of the distance to a first transaction amount; Acquire the transaction time of each offline transaction, and determine the first coordinate of each offline transaction based on the transaction time and the first sub-characteristic value; Determine a second sub-feature value for each online transaction, the feature value being a ratio of the risk level to a second transaction amount; Determine the second coordinate of each online transaction based on the transaction time and the second sub-characteristic value of each transaction object; Mapping the first coordinate and the second coordinate to a preset coordinate system and generating a line graph, wherein the horizontal axis of the preset coordinate system is the transaction time and the vertical axis is the characteristic value; Smoothing the line graph to obtain a smooth curve graph; Calculating the similarity between the smooth curve graph and the smooth curve graph of each dishonest person in the preset dishonest person database; If there is a target defaulter who reaches a preset similarity threshold, the default risk probability of the user is determined based on the similarity of the target defaulter and the target defaulter.

9. An electronic device, characterized in that: It includes: at least one processor; Memory; At least one application, wherein the at least one application is stored in the memory and is configured to be executed by the at least one processor, and the at least one application is used to execute a user risk monitoring method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed in a computer, the computer is caused to execute a user risk monitoring method as described in any one of claims 1 to 6.

Images