User credit risk management method and system based on user behavior analysis

Abstract

The invention discloses a user credit risk management method and system based on user behavior analysis, and relates to the technical field of credit risk management. Related data of a target user are obtained through a first obtaining mode, basic information items are obtained, specific occasion data of the target user are obtained through a second obtaining mode, specific information items are obtained, the basic information items are used for judging basic credit of the target user, and the specific information items are used for judging feature credit of the target user. According to the invention, the credit risk is divided into three grades, namely good, normal and poor, and the basic credit of the target user and the specific judgment result of the behavior data in a certain occasion are subjected to grade judgment through the judgment method, so that the specific credit risk grade of the target user is output. The effect of performing credit risk management on the target user according to specific requirements is realized, and the credit risk management accuracy of the target user is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of credit risk management, and in particular to a user credit risk management method and system based on user behavior analysis. Background Art

[0002] Credit risk management refers to a series of strategies and measures taken to identify, evaluate and control the risk of default and potential losses caused by customers or counterparties. Its main goal is to reduce financial losses and improve the quality of credit decisions, identify the sources of credit risk, and analyze the various factors that affect credit risk. Cold chain logistics refers to the use of a series of facilities and technical means to control the temperature of temperature-sensitive goods throughout the logistics process to ensure that these goods are kept in a specific low-temperature environment during storage, transportation and distribution. In cold chain logistics, it is also necessary to judge and manage the customer's credit.

[0003] The user credit risk assessment method and system based on data analysis, with patent publication number CN117557361A, obtains basic information and historical credit behavior data of different users, and extracts non-time series attribute data and time series attribute data to form samples. The characteristics of the non-time series attribute data and the time series attribute data of the samples are learned through a multi-branch deep learning network, and the risk assessment model is trained to perform user credit risk assessment. The credit risk assessment can be performed by combining the non-time series attributes and time series attributes in the user's data, thereby improving the accuracy of credit risk assessment.

[0004] When the above-mentioned and similar technical solutions manage the credit risk of users based on their behaviors, different behaviors represent the credit level of users in different situations, and the multi-source heterogeneous information obtained from users can only present the overall credit level of users, but cannot clearly obtain the credit level of users in specific situations. For example, through the obtained multi-source heterogeneous information, it is known that the overall credit level of the user is poor, but in the final payment project, the user has never defaulted on the final payment. Therefore, it is impossible to accurately evaluate the credit level of the user in specific situations based on the obtained multi-source heterogeneous information, resulting in poor credit risk management of the user. Summary of the invention

[0005] The purpose of the present invention is to provide a user credit risk management method and system based on user behavior analysis to solve the problems raised in the above background technology.

[0006] To achieve the above objectives, the present invention provides the following technical solutions: a user credit risk management method and system based on user behavior analysis, comprising: Data fusion: The relevant data of the target user is obtained through the first acquisition method to obtain basic information items, and the specific occasion data of the target user is obtained through the second acquisition method to obtain specific information items. The basic information items are used to determine the basic credit of the target user, and the specific information items are used to determine the characteristic credit of the target user; Credit evaluation: Based on the basic information items, the evaluation is performed by a first evaluation method, and based on the specific information items, the evaluation is performed by a second evaluation method, and a first evaluation value and a second evaluation value are obtained respectively. The first evaluation value and the second evaluation value are used to determine the specific credit data of the target user; Grading: Credit risk is divided into at least two grades through the classification method, and at least two credit rating items are obtained; Grade determination: The first evaluation value and the second evaluation value are combined and determined by the determination method to obtain a combination result, and the credit risk assessment data of the target user is output based on the comparison relationship between the combination result and the credit grade item; The determination method comprises: Step 1: Ratio setting, setting the total impact ratio value and the allocation threshold, allocating the total impact ratio value based on the allocation threshold to obtain a first impact value and a second impact value, and the combination result of the first impact value and the second impact value is the total impact ratio value; Step 2: Impact comparison, based on the first impact value and the second impact value, the first evaluation value and the second evaluation value are compared respectively to obtain a basic risk value and a specific risk value; Step 3: Risk combination. Based on the combination of the basic risk value and the specific risk value, the risk combination item is obtained. The comparison relationship between the risk combination item and the credit rating item is obtained to obtain the actual credit rating of the target user.

[0007] Furthermore, the division method includes: S1: Value setting, setting the credit value and the cutting value, obtaining the credit item and the cutting item; S2: range cutting, based on the combination result of the cutting item and the credit item, obtain the cutting credit item, based on the combination result of the cutting credit item and the cutting item, obtain the secondary cutting item, repeat the steps to obtain at least two credit value data; S3: Range matching, setting the levels based on the credit value data to obtain the credit level items.

[0008] Furthermore, the target user's relevant data includes loan records and overdue record information, and the first acquisition method includes: M1: first threshold setting, setting the first time threshold to obtain the first time item; M2: First information acquisition, based on the first time item, the target user's loan record and overdue record data are acquired to obtain the interactive data item.

[0009] Furthermore, the first evaluation method includes: setting at least two first evaluation data to obtain first evaluation data items, obtaining at least three first evaluation ranges based on the first evaluation data items, setting first evaluation values ​​respectively based on the first evaluation ranges, matching the first evaluation ranges based on the interactive data items, and matching the corresponding first evaluation values ​​to obtain a first evaluation value.

[0010] Furthermore, the target user's relevant data includes a consumption amount, and the first acquisition method includes: N1: second threshold setting, setting the second time threshold to obtain the second time item; N2: Second information acquisition: acquiring the total consumption amount of the target user based on the second time item to obtain the consumption data item.

[0011] Furthermore, the first evaluation method includes: setting at least two second evaluation data to obtain second evaluation data items, obtaining at least three second evaluation ranges based on the second evaluation data items, setting second evaluation values ​​respectively based on the second evaluation ranges, matching the second evaluation ranges based on the consumption data items, and matching the corresponding second evaluation values ​​to obtain the first evaluation value.

[0012] Furthermore, the target user's specific occasion data includes the payment time of the final payment for cold chain logistics, and the second acquisition method includes: setting a third threshold, setting a third time threshold, obtaining a third time item, and obtaining the target user's cold chain logistics final payment time based on the third time item to obtain a payment time item.

[0013] Furthermore, the second evaluation method includes: setting at least two third evaluation data to obtain third evaluation data items, obtaining at least three third evaluation ranges based on the third evaluation data items, setting third evaluation values ​​respectively based on the third evaluation ranges, matching the third evaluation ranges based on the payment time items, and matching the corresponding third evaluation values ​​to obtain a second evaluation value.

[0014] Furthermore, a user credit risk management system based on user behavior analysis uses the user credit risk management method based on user behavior analysis described above.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The user credit risk management method and system based on user behavior analysis divides credit risk into three levels, namely good, normal and poor, and uses a judgment method to judge the level of the basic credit of the target user and the specific judgment results of the behavior data in a certain situation, and then outputs the specific credit risk level of the target user, thereby achieving the effect of credit risk management of the target user according to specific needs and improving the accuracy of credit risk management of the target user.

[0016] At the same time, the basic credit of the target user is judged by combining all aspects of the data, and the target user's specific occasion data represents the target user's behavioral data in a certain occasion. For example, when the project cooperated with the customer is cold chain logistics, the primary concern is the customer's payment of the down payment and the balance payment in the previous cold chain logistics process. Therefore, specific information items are used to determine the characteristic credit of the target user. The first evaluation method and the second evaluation method are respectively used to make specific judgments on the target user's basic credit and behavioral data in a certain occasion. According to specific needs, the target user's characteristic credit can be judged based on the target user's data in a specific occasion. Through the combined judgment of basic credit and characteristic credit, the accuracy of credit judgment on the target user is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of the relationship between data fusion, credit evaluation, grade classification and grade determination of the present invention; Figure 2 It is a schematic flow chart of the determination method of the present invention; Figure 3 A schematic diagram of the relationship between the total impact ratio value and the allocation threshold value of the present invention; Figure 4 is a schematic diagram of interactive data items of the present invention; Figure 5 A schematic diagram of the relationship between the consumption amount and the basic risk value of the present invention; Figure 6 It is a schematic diagram of the specific risk value relationship of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] When managing the credit risk of users, users will have different credit reactions to things in different situations. For example, when it is obtained through multi-source heterogeneous data that the target user is overdue and the transaction situation is poor, it is known that the credit level of the target user is poor. However, in specific situations, such as the user has never defaulted on the final payment during the cold chain logistics final payment process, in the cold chain logistics management alone, when the customer is evaluated as "poor credit level" based on the data obtained from multi-source heterogeneous data, the evaluation will obviously be inaccurate. The technical solution provided by this application divides credit risk into three levels, namely The system uses a judgment method to grade the target user's basic credit and the specific evaluation results of the behavioral data in a certain situation into good, normal and poor. Then, the specific credit risk level of the target user is output, achieving the effect of credit risk management of the target user according to specific needs, improving the accuracy of credit risk management of the target user, judging the basic credit of the target user by combining the multi-source heterogeneous data of the target user, and then judging the characteristic credit of the target user based on the data of the target user in a specific situation according to specific needs. The accuracy of credit judgment of the target user is improved through the combined judgment of basic credit and characteristic credit.

[0020] like Figure 1-Figure 2As shown, the present invention provides a technical solution: a user credit risk management method and system based on user behavior analysis, including: data fusion: obtaining relevant data of the target user through a first acquisition method to obtain basic information items, obtaining specific occasion data of the target user through a second acquisition method to obtain specific information items, the basic information items are used to determine the basic credit of the target user, and the specific information items are used to determine the characteristic credit of the target user; credit judgment: based on the basic information items, judging by a first judgment method, based on the specific information items, judging by a second judgment method, respectively obtaining a first judgment value and a second judgment value, respectively, the first judgment value and the second judgment value are used to determine the specific credit data of the target user; grade division: dividing the credit risk into at least two grades by a division method, obtaining at least two credit grade items; grade judgment Determination: The first judgment value and the second judgment value are combined and judged by a judgment method to obtain a combination result, and the credit risk assessment data of the target user is output based on the comparison relationship between the combination result and the credit rating item; the judgment method includes: step one: matching ratio setting, setting the total impact matching ratio value and the allocation threshold, and allocating the total impact matching ratio value based on the allocation threshold to obtain the first impact value and the second impact value, and the combination result of the first impact value and the second impact value is the total impact matching value; step two: impact comparison, based on the first impact value and the second impact value, respectively compare the first judgment value and the second judgment value to obtain a basic risk value and a specific risk value; step three: risk combination, based on the combination result of the basic risk value and the specific risk value, obtain a risk combination item, obtain the comparison relationship between the risk combination item and the credit rating item, and obtain the actual credit rating of the target user.

[0021] It should be noted that in the case of combined multi-source heterogeneous data, the target user's relevant data represents the user's consumption record, payment method, transaction frequency and amount, the user's past borrowing record, repayment habits, overdue record, the user's clicks on the website or application, browsing behavior, dynamic updates and other data. The basic credit of the target user is judged by combining all aspects of the data, and the target user's specific occasion data represents the target user's behavior data in a certain occasion. For example, when the project cooperated with the customer is cold chain logistics, the primary concern is the customer's payment of the down payment and the balance payment in the previous cold chain logistics process. Therefore, specific information items are used to determine the characteristic credit of the target user. The first evaluation method and the second evaluation method are respectively used to evaluate the target user's basic credit. The system can make a specific judgment on the target user's basic credit and the behavioral data in a certain situation, and divide the credit risk into three levels, namely good, normal and poor. The judgment method is used to judge the level of the target user's basic credit and the specific judgment results of the behavioral data in a certain situation, and then output the specific credit risk level of the target user, so as to achieve the effect of credit risk management of the target user according to specific needs, improve the accuracy of credit risk management of the target user, judge the basic credit of the target user by combining the multi-source heterogeneous data of the target user, and then judge the characteristic credit of the target user according to the data of the target user in a specific situation according to specific needs, and improve the accuracy of credit judgment of the target user by combining the basic credit and characteristic credit.

[0022] like Figure 3 As shown, it should be noted that the total impact ratio value and the allocation threshold are set. The total impact ratio value is 100%, and the allocation threshold is 0.3. At this time, the first impact value and the second impact value are 30% and 70% respectively. The first evaluation value and the second evaluation value are compared and matched according to the first impact value and the second impact value to obtain the basic risk value and the specific risk value. The combination of the basic risk value and the specific risk value is compared with the credit risk level to obtain the specific credit risk level of the target user.

[0023] The division method includes: S1: numerical setting, setting the credit value and the cutting value to obtain the credit item and the cutting item; S2: range cutting, based on the combination result of the cutting item and the credit item, obtaining the cutting credit item, based on the combination result of the cutting credit item and the cutting item, obtaining the secondary cutting item, repeating the steps to obtain at least two credit value data; S3: range matching, setting the level respectively based on the credit value data to obtain the credit level item.

[0024] It should be noted that the set credit value is 100 and the cut value is 20. Since there are three credit risk levels, namely good, normal and poor, the cut credit item is 80 and the secondary cut item is 60, resulting in three credit value ranges, namely 100-80, 80-60, and below 60, corresponding to good, normal and poor respectively.

[0025] The relevant data of the target user includes loan records and overdue record information, and the first acquisition method includes: M1: first threshold setting, setting the first time threshold to obtain the first time item; M2: first information acquisition, obtaining the loan records and overdue record data of the target user based on the first time item to obtain the interactive data item.

[0026] It should be noted that the first time threshold is set to one year, and the target user's loan record and overdue record data within one year are obtained to obtain the interactive data item.

[0027] The first evaluation method includes: setting at least two first evaluation data to obtain first evaluation data items, obtaining at least three first evaluation ranges based on the first evaluation data items, setting first evaluation values ​​respectively based on the first evaluation ranges, matching the first evaluation ranges based on the interactive data items, and matching the corresponding first evaluation values ​​to obtain a first evaluation value.

[0028] It should be noted that the first evaluation data set is the target user's loan record and overdue record number, which are 3 and 6 times respectively. The three first evaluation ranges obtained are 0-3 times, 3-6 times and more than 6 times. The set first evaluation values ​​are 100, 80 and 60 respectively, corresponding to the three first evaluation ranges.

[0029] like Figure 4 As shown in the specific implementation process, the number of loans of a certain user in a year is obtained to be 3, and the overdue record is 4. At this time, the interactive data item of the target user is 7 times. The credit value data of the target user obtained through matching is 60. Since the first impact value is 30%, the basic risk value of the target user is obtained at this time. .

[0030] The relevant data of the target user includes the consumption amount, and the first acquisition method includes: N1: second threshold setting, setting the second time threshold to obtain the second time item; N2: second information acquisition, obtaining the total consumption amount of the target user based on the second time item to obtain the consumption data item.

[0031] It should be noted that the second time threshold is set to one year, and the total consumption amount data of the target user within one year is obtained to obtain the consumption data item.

[0032] The first evaluation method includes: setting at least two second evaluation data to obtain second evaluation data items, obtaining at least three second evaluation ranges based on the second evaluation data items, setting second evaluation values ​​respectively based on the second evaluation ranges, matching the second evaluation ranges based on the consumption data items, and matching the corresponding second evaluation values ​​to obtain the first evaluation value.

[0033] It should be noted that the second evaluation data set is the total consumption amount of the target users, which are 200,000 and 400,000 respectively. The three second evaluation ranges obtained are above 400,000, 200,000-400,000 and 0-200,000 respectively. The set second evaluation values ​​are 100, 80 and 60 respectively, corresponding to the three second evaluation ranges.

[0034] like Figure 5 As shown in the specific implementation process, the consumption amount of a user in one year is 330,000, and the credit value data of the target user obtained through matching is 80. Since the first impact value is 30%, the basic risk value of the target user is obtained at this time. .

[0035] Embodiment 1 The target user's specific occasion data includes the payment time of the final payment for cold chain logistics. The second acquisition method includes: setting a third threshold, setting a third time threshold, obtaining a third time item, and obtaining the target user's cold chain logistics final payment time based on the third time item to obtain a payment time item.

[0036] It should be noted that the third time threshold is set to one year, and the final payment time of the target user in the cold chain logistics within one year is obtained to obtain the payment time item.

[0037] The second evaluation method includes: setting at least two third evaluation data to obtain third evaluation data items, obtaining at least three third evaluation ranges based on the third evaluation data items, setting third evaluation values ​​respectively based on the third evaluation ranges, matching the third evaluation ranges based on the payment time items, and matching the corresponding third evaluation values ​​to obtain a second evaluation value.

[0038] It should be noted that the third assessment data set is the number of defaults in the final payment of the target users, which are 3 times and 6 times respectively, resulting in three third assessment ranges, namely 0-3 times, 3-6 times and more than 6 times, and the set third assessment values ​​are 100, 80 and 60, respectively corresponding to the three third assessment ranges.

[0039] like Figure 6 As shown in the specific implementation process, the number of defaults of a certain user in one year is 0, and the credit value data of the target user obtained through matching is 100. Since the first impact value is 70%, the specific risk value of the target user is obtained at this time. .

[0040] In the specific implementation process, it is obtained that a certain user has borrowed twice in a year and has 0 overdue records. At this time, the target user's interactive data item is 2 times. Through matching, the first evaluation value of the target user is 100. Since the first impact value is 30%, the basic risk value of the target user is obtained at this time. At the same time, the target user was found to have failed to pay on time 4 times during the cold chain final payment process within one year. The second evaluation value of the target user was 80 through matching. Since the first impact value was 70%, the specific risk value of the target user was By comparing the combined result of the basic risk value and the specific risk value with the credit risk level, it is found that the specific credit risk level of the target user is good.

[0041] Embodiment 2 It should be noted that the target user's specific occasion data also includes the number of complaints received in cold chain logistics. The second acquisition method includes: setting a third threshold, setting a third time threshold, obtaining a third time item, and obtaining the number of cold chain logistics complaints of the target user based on the third time item to obtain a complaint number item.

[0042] It should be noted that the third time threshold is set to one year, and the number of complaints made by the target user in cold chain logistics within one year is obtained to obtain the number of complaints item.

[0043] The second evaluation method includes: setting at least two third evaluation data to obtain third evaluation data items, obtaining at least three third evaluation ranges based on the third evaluation data items, setting third evaluation values ​​respectively based on the third evaluation ranges, matching the third evaluation ranges based on the number of complaints, and matching the corresponding third evaluation values ​​to obtain a second evaluation value.

[0044] It should be noted that the third evaluation data set is the number of complaints received by the target users, which are 3 times and 6 times respectively, resulting in three third evaluation ranges, namely 0-3 times, 3-6 times and more than 6 times. The set third evaluation values ​​are 100, 80 and 60 respectively, corresponding to the three third evaluation ranges.

[0045] In the specific implementation process, the number of complaints against a certain user in a year is obtained to be 4. Through matching, the credit value data of the target user is obtained to be 80. Since the first impact value is 70%, the specific risk value of the target user is obtained at this time. .

[0046] In the specific implementation process, it is obtained that a certain user has borrowed 4 times in a year and has 3 overdue records. At this time, the target user's interactive data items are 7 times. Through matching, the first evaluation value of the target user is 60. Since the first impact value is 30%, the basic risk value of the target user is obtained at this time. At the same time, the target user is found to have received 0 complaints in the cold chain logistics within one year. The second evaluation value of the target user is 100 through matching. Since the first impact value is 70%, the specific risk value of the target user is By comparing the combined result of the basic risk value and the specific risk value with the credit risk level, it is found that the specific credit risk level of the target user is good.

[0047] The basic credit of the target user is judged by combining the multi-source heterogeneous data of the target user, and then according to specific needs, the characteristic credit of the target user is judged based on the data of the target user in specific situations. The accuracy of the credit judgment of the target user is improved through the combined judgment of basic credit and characteristic credit, and the characteristic credit is judged by at least the above two methods.

[0048] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is limited by the attached embodiments and their equivalents.

Claims

1. A user credit risk management method based on user behavior analysis, characterized in that: include: Data fusion: The relevant data of the target user is obtained through the first acquisition method to obtain basic information items, and the specific occasion data of the target user is obtained through the second acquisition method to obtain specific information items. The basic information items are used to determine the basic credit of the target user, and the specific information items are used to determine the characteristic credit of the target user; Credit evaluation: Based on the basic information items, the evaluation is performed by a first evaluation method, and based on the specific information items, the evaluation is performed by a second evaluation method, and a first evaluation value and a second evaluation value are obtained respectively. The first evaluation value and the second evaluation value are used to determine the specific credit data of the target user; Grading: Credit risk is divided into at least two grades through the classification method, and at least two credit rating items are obtained; Grade determination: The first evaluation value and the second evaluation value are combined and determined by the determination method to obtain a combination result, and the credit risk assessment data of the target user is output based on the comparison relationship between the combination result and the credit grade item; The determination method comprises: Step 1: Ratio setting, setting the total impact ratio value and the allocation threshold, allocating the total impact ratio value based on the allocation threshold to obtain a first impact value and a second impact value, and the combination result of the first impact value and the second impact value is the total impact ratio value; Step 2: Impact comparison, based on the first impact value and the second impact value, the first evaluation value and the second evaluation value are compared respectively to obtain a basic risk value and a specific risk value; Step 3: Risk combination. Based on the combination of the basic risk value and the specific risk value, the risk combination item is obtained. The comparison relationship between the risk combination item and the credit rating item is obtained to obtain the actual credit rating of the target user.

2. The user credit risk management method based on user behavior analysis according to claim 1, characterized in that: The division method comprises: S1: Value setting, setting the credit value and the cutting value, and obtaining the credit item and the cutting item; S2: range cutting, based on the combination result of the cutting item and the credit item, obtain the cutting credit item, based on the combination result of the cutting credit item and the cutting item, obtain the secondary cutting item, repeat the steps to obtain at least two credit value data; S3: Range matching, setting the levels based on the credit value data to obtain the credit level items.

3. The user credit risk management method based on user behavior analysis according to claim 1, characterized in that: The target user's relevant data includes loan records and overdue record information, and the first acquisition method includes: M1: first threshold setting, setting the first time threshold to obtain the first time item; M2: First information acquisition, based on the first time item, the target user's loan record and overdue record data are acquired to obtain the interactive data item.

4. The user credit risk management method based on user behavior analysis according to claim 3 is characterized by: The first evaluation method includes: setting at least two first evaluation data to obtain first evaluation data items, obtaining at least three first evaluation ranges based on the first evaluation data items, setting first evaluation values ​​respectively based on the first evaluation ranges, matching the first evaluation ranges based on the interactive data items, and matching the corresponding first evaluation values ​​to obtain a first evaluation value.

5. The user credit risk management method based on user behavior analysis according to claim 1 is characterized by: The target user's relevant data includes the consumption amount, and the first acquisition method includes: N1: second threshold setting, setting the second time threshold to obtain the second time item; N2: Second information acquisition: acquiring the total consumption amount of the target user based on the second time item to obtain the consumption data item.

6. A user credit risk management method based on user behavior analysis according to claim 5, characterized in that: The first evaluation method includes: setting at least two second evaluation data to obtain second evaluation data items, obtaining at least three second evaluation ranges based on the second evaluation data items, setting second evaluation values ​​respectively based on the second evaluation ranges, matching the second evaluation ranges based on the consumption data items, and matching the corresponding second evaluation values ​​to obtain the first evaluation value.

7. The user credit risk management method based on user behavior analysis according to claim 1, characterized in that: The target user's specific occasion data includes the payment time of the final payment for cold chain logistics. The second acquisition method includes: setting a third threshold, setting a third time threshold, obtaining a third time item, and obtaining the target user's cold chain logistics final payment time based on the third time item to obtain a payment time item.

8. The user credit risk management method based on user behavior analysis according to claim 6 is characterized by: The second evaluation method includes: setting at least two third evaluation data to obtain third evaluation data items, obtaining at least three third evaluation ranges based on the third evaluation data items, setting third evaluation values ​​respectively based on the third evaluation ranges, matching the third evaluation ranges based on the payment time items, and matching the corresponding third evaluation values ​​to obtain a second evaluation value.

9. A user credit risk management system based on user behavior analysis, characterized in that: A user credit risk management method based on user behavior analysis as described in any one of claims 1-8 is used.

Images