User post-payment credit payment system based on big data

Abstract

The present invention discloses a user post-paid credit payment system based on big data. The post-paid credit platform integrates the user's cross-platform credit information and sets the basic credit limit. The dynamic adjustment mechanism adjusts the credit limit in real time according to the performance (addition or attenuation) of the payment time point in the deduction cycle. During the analysis period, user orders and interaction data are collected, consumption behavior characteristics are extracted, and user groups are divided by hierarchical clustering. The credit changes and real-time credit limits of each group are counted, and credit risk assessment and grading are performed to achieve precise and integrated credit analysis and management.

Description

Technical Field

[0001] The present invention relates to the field of payment system data analysis, and more specifically, to a user post-paid credit payment system based on big data. Background Art

[0002] With the rapid development of Internet finance, user post-paid credit payment systems have become an important part of the market. However, traditional credit assessment methods often rely on a single credit scoring model, which is difficult to fully and accurately reflect the credit status of users. At the same time, the traditional system lacks a dynamic adjustment mechanism for user credit limits, resulting in deficiencies in risk control and service efficiency. In addition, in existing credit analysis systems, credit analysis is often only performed on individual user information, which makes it impossible to quickly analyze and classify large-scale users, and it is difficult to perform hierarchical analysis and corresponding credit risk analysis on users from an overall and group perspective, resulting in low risk assessment capabilities and poor comprehensive credit analysis of large-scale user groups.

[0003] Therefore, developing a user post-paid credit payment system based on big data and intelligent algorithms is of great significance for improving the level of financial services and reducing credit risks. Summary of the invention

[0004] The present invention overcomes the defects of the prior art and proposes a user post-payment credit payment system based on big data.

[0005] The first aspect of the present invention provides a user post-paid credit payment method based on big data, comprising:

[0006] Obtain the user's credit information on other preset consumption platforms through the postpaid credit platform, and set the user's basic credit limit based on the credit information;

[0007] Set the deduction cycle. In each postpaid order of the user, obtain the user's payment time through the postpaid credit platform, determine whether the user's payment time is within the deduction cycle, and calculate the time difference with the deduction cycle. Perform credit addition or credit decay based on the time difference, dynamically adjust the basic credit limit, and obtain the real-time credit limit;

[0008] In one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data, extracts user features based on the user big data, obtains user consumption behavior features, uses user consumption behavior features as clustering basic data, and clusters and groups users based on hierarchical clustering to form multiple user groups;

[0009] During an analysis cycle, the number of credit additions and reductions and the real-time credit limits of all users in the user group are counted to form credit statistical data for the user group. Credit risk assessment is performed based on the credit statistical data, and credit levels are divided based on multiple user groups, with each user group corresponding to a credit level.

[0010] In this solution, the post-paid credit platform is used to obtain the user's credit information on other preset consumption platforms, and the user's basic credit limit is set based on the credit information, specifically:

[0011] Obtain user information through the postpaid credit platform;

[0012] Using user information, conduct a consumer credit query on at least one preset consumer platform to obtain corresponding credit information;

[0013] An initial comprehensive assessment of the user is conducted based on credit information to obtain a basic credit limit.

[0014] In this solution, the deduction cycle is set. In each postpaid order of the user, the user's payment time point is obtained through the postpaid credit platform, and it is determined whether the user's payment time point is within the deduction cycle, and the time difference with the deduction cycle is calculated. Credit addition or credit decay is performed based on the time difference, and the basic credit limit is dynamically adjusted to obtain the real-time credit limit, which is specifically:

[0015] Set up a deduction cycle within the postpaid credit platform;

[0016] In each postpaid order of the user, the user's payment time is obtained through the postpaid credit platform to determine whether the user's payment time is within the deduction cycle;

[0017] If yes, then calculate the time difference between the user's payment time and the deduction cycle, and use the time difference to add credit. The larger the time difference, the greater the credit addition value.

[0018] If not, the time difference between the user's payment time and the deduction cycle is calculated, and the credit decay is performed based on the time difference. The larger the time difference, the greater the credit decay value.

[0019] Dynamically adjust the basic credit limit based on credit bonus / credit decay, and get the real-time credit limit.

[0020] In this solution, within one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data, extracts user features based on the user big data, obtains user consumption behavior features, uses user consumption behavior features as clustering basic data, and clusters and groups users based on hierarchical clustering to form multiple user groups, specifically:

[0021] In one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data;

[0022] Perform data cleaning, outlier removal and standardization preprocessing on user big data;

[0023] Based on user big data, extract the feature information of user order data. The feature dimensions include order amount, order time, and order quantity, and obtain the user order features.

[0024] Based on user big data, the consumption feature information of users is extracted. The feature dimensions include consumption history, consumption browsing information, and search evaluation information dimensions to obtain user consumption features;

[0025] Integrate user order characteristics and user consumption characteristics to form user consumption behavior characteristics;

[0026] The user consumption behavior characteristics are used as the clustering basic data, and the users are clustered and grouped through the hierarchical clustering method. The clustering process is as follows: initially, each user is regarded as a class, and the distance values ​​between classes are calculated based on the Manhattan distance. Then, all classes are merged cyclically according to the preset merging rules until the number of classes reaches the preset number.

[0027] The preset merging rule is based on the shortest distance method;

[0028] The users are grouped based on a preset number of classes to obtain multiple user groups, each of which corresponds to a class.

[0029] In this solution, the preset merging rules include:

[0030] Based on the shortest distance method, the distance value between two classes is analyzed. If the distance value is within the shortest distance range, the first condition is met;

[0031] Analyze the difference in real-time credit limits of users in the two categories, and if the difference is within a preset range, the second condition is met;

[0032] The first condition and the second condition are set as a preset merging rule.

[0033] In this solution, within one analysis cycle, the credit addition and attenuation times of all users in the user group and the real-time credit limit of the users are counted to form a credit statistical data of the user group. Credit risk assessment is performed based on the credit statistical data, and credit ratings are divided based on multiple user groups. Each user group corresponds to a credit rating, specifically:

[0034] In one analysis cycle, the credit addition and reduction times of all users in each user group are counted, and the ratio of the times is calculated. The maximum, minimum and average values ​​of the real-time credit limit of all users are calculated, and the statistical and calculated data are integrated to form credit statistical data;

[0035] Conduct credit risk assessment for each user group based on credit statistics and generate assessment data;

[0036] Based on a user group, the real-time credit limits of the corresponding users are averaged to obtain the group credit value;

[0037] Based on the group credit value, each user group is sorted, and the corresponding credit level is generated for each user group through the sorting results.

[0038] The second aspect of the present invention further provides a user post-paid credit payment system based on big data, the system comprising: a memory, a processor, the memory comprising a user post-paid credit payment program based on big data, the user post-paid credit payment program based on big data being executed by the processor to implement the following steps:

[0039] Obtain the user's credit information on other preset consumption platforms through the postpaid credit platform, and set the user's basic credit limit based on the credit information;

[0040] Set the deduction cycle. In each postpaid order of the user, obtain the user's payment time through the postpaid credit platform, determine whether the user's payment time is within the deduction cycle, and calculate the time difference with the deduction cycle. Perform credit addition or credit decay based on the time difference, dynamically adjust the basic credit limit, and obtain the real-time credit limit;

[0041] In one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data, extracts user features based on the user big data, obtains user consumption behavior features, uses user consumption behavior features as clustering basic data, and clusters and groups users based on hierarchical clustering to form multiple user groups;

[0042] During an analysis cycle, the number of credit additions and reductions and the real-time credit limits of all users in the user group are counted to form credit statistical data for the user group. Credit risk assessment is performed based on the credit statistical data, and credit levels are divided based on multiple user groups, with each user group corresponding to a credit level.

[0043] The third aspect of the present invention also provides a computer-readable storage medium, which includes a user post-paid credit payment program based on big data. When the user post-paid credit payment program based on big data is executed by a processor, the steps of the user post-paid credit payment method based on big data as described in any one of the above are implemented.

[0044] The present invention discloses a user post-paid credit payment system based on big data. The post-paid credit platform integrates the user's cross-platform credit information and sets the basic credit limit. The dynamic adjustment mechanism adjusts the credit limit in real time according to the performance (addition or attenuation) of the payment time point in the deduction cycle. During the analysis period, user orders and interaction data are collected, consumption behavior characteristics are extracted, and user groups are divided by hierarchical clustering. The credit changes and real-time credit limits of each group are counted, and credit risk assessment and grading are performed to achieve precise and integrated credit analysis and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A flow chart of a user post-paid credit payment method based on big data of the present invention is shown;

[0046] Figure 2 A block diagram of a user post-paid credit payment system based on big data of the present invention is shown. DETAILED DESCRIPTION

[0047] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0048] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0049] Figure 1 A flow chart of a user post-paid credit payment method based on big data of the present invention is shown.

[0050] like Figure 1 As shown, the first aspect of the present invention provides a user post-paid credit payment method based on big data, comprising:

[0051] S102, obtaining the user's credit information on other preset consumption platforms through the postpaid credit platform, and setting the user's basic credit limit based on the credit information;

[0052] S104, setting a deduction cycle. In each postpaid order of the user, the user payment time point is obtained through the postpaid credit platform, and it is determined whether the user payment time point is within the deduction cycle, and the time difference with the deduction cycle is calculated. Credit addition or credit decay is performed based on the time difference, and the basic credit limit is dynamically adjusted to obtain the real-time credit limit;

[0053] S106, within an analysis cycle, through the postpaid credit platform, collect the user's order data and platform interaction data to form user big data, extract user features based on the user big data, obtain user consumption behavior features, use the user consumption behavior features as clustering basic data and cluster and group the users based on hierarchical clustering to form multiple user groups;

[0054] S108, within one analysis cycle, the number of credit additions and reductions and the real-time credit limits of all users in the user group are counted to form a credit statistical data of the user group, credit risk assessment is performed based on the credit statistical data, and credit levels are divided based on multiple user groups, with each user group corresponding to a credit level.

[0055] According to an embodiment of the present invention, the post-paid credit platform is used to obtain the user's credit information on other preset consumption platforms, and the user's basic credit limit is set based on the credit information, specifically:

[0056] Obtain user information through the postpaid credit platform;

[0057] Using user information, conduct a consumer credit query on at least one preset consumer platform to obtain corresponding credit information;

[0058] An initial comprehensive assessment of the user is conducted based on credit information to obtain a basic credit limit.

[0059] It should be noted that the preset consumption platform is generally an e-commerce platform with consumer transactions. The credit information includes information such as the credit value on the corresponding platform, such as WeChat payment points. The larger the score, the larger the corresponding basic credit limit. When calculating the basic credit limit, the credit information of multiple preset consumption platforms can also be combined for average analysis.

[0060] When making a postpaid payment, based on the basic credit limit, the platform can set the number of postpaid orders or credit limit for the user.

[0061] According to an embodiment of the present invention, the deduction cycle is set. In each postpaid order of the user, the user payment time point is obtained through the postpaid credit platform, and it is determined whether the user payment time point is within the deduction cycle, and the time difference with the deduction cycle is calculated. Credit addition or credit decay is performed according to the time difference, and the basic credit limit is dynamically adjusted to obtain the real-time credit limit, which is specifically:

[0062] Set up a deduction cycle within the postpaid credit platform;

[0063] In each postpaid order of the user, the user's payment time is obtained through the postpaid credit platform to determine whether the user's payment time is within the deduction cycle;

[0064] If yes, then calculate the time difference between the user's payment time and the deduction cycle, and use the time difference to add credit. The larger the time difference, the greater the credit addition value.

[0065] If not, the time difference between the user's payment time and the deduction cycle is calculated, and the credit decay is performed based on the time difference. The larger the time difference, the greater the credit decay value.

[0066] Dynamically adjust the basic credit limit based on credit bonus / credit decay, and get the real-time credit limit.

[0067] It should be noted that the deduction period is a limited time value. After the user makes a post-paid payment, if it is within the deduction period, the user's credit will be increased. If it is after the deduction period, the user's credit will be reduced. The specific increase / decay value is determined by the size of the time difference. The basic credit limit is the user's initial basic value. After the user registers a platform account and performs post-consumption behavior, the real-time credit limit will be used to analyze the user's payment. The basic credit limit is stored in the database as the user's basic information;

[0068] The credit bonus / decay value calculation formula is as follows:

[0069] ;

[0070] Among them, S is the credit bonus / decay value, K is the correction coefficient, Payment time for users, The deduction period.

[0071] According to an embodiment of the present invention, within one analysis cycle, the user's order data and platform interaction data are collected through the postpaid credit platform to form user big data, user features are extracted based on the user big data, and user consumption behavior features are obtained. The user consumption behavior features are used as clustering basic data and users are clustered and grouped based on hierarchical clustering to form multiple user groups, specifically:

[0072] In one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data;

[0073] Perform data cleaning, outlier removal and standardization preprocessing on user big data;

[0074] Based on user big data, extract the feature information of user order data. The feature dimensions include order amount, order time, and order quantity, and obtain the user order features.

[0075] Based on user big data, the consumption feature information of users is extracted. The feature dimensions include consumption history, consumption browsing information, and search evaluation information dimensions to obtain user consumption features;

[0076] Integrate user order characteristics and user consumption characteristics to form user consumption behavior characteristics;

[0077] The user consumption behavior characteristics are used as the clustering basic data, and the users are clustered and grouped through the hierarchical clustering method. The clustering process is as follows: initially, each user is regarded as a class, and the distance values ​​between classes are calculated based on the Manhattan distance. Then, all classes are merged cyclically according to the preset merging rules until the number of classes reaches the preset number.

[0078] The preset merging rules are based on the shortest distance method;

[0079] The users are grouped based on a preset number of classes to obtain multiple user groups, each of which corresponds to a class.

[0080] It should be noted that the user consumption behavior characteristics include the behavior characteristics of each platform user, and calculating the distance value between classes specifically means calculating the behavior characteristic distance value between users.

[0081] The preset number is generally 5 to 10.

[0082] According to an embodiment of the present invention, the preset merging rule includes:

[0083] Based on the shortest distance method, the distance value between two classes is analyzed. If the distance value is within the shortest distance range, the first condition is met;

[0084] Analyze the difference in real-time credit limits of users in the two categories, and if the difference is within a preset range, the second condition is met;

[0085] The first condition and the second condition are set as a preset merging rule.

[0086] It should be noted that in the preset merging rules, clustering and merging operations can be performed based only on the shortest distance method. At this time, the second condition, that is, clustering based on user credit limit, will not be considered, and the clustering results tend to cluster based on user consumption habits. Furthermore, the preset merging rules can also include rule settings that meet the second condition, that is, adding real-time credit limit as the basis for grouping. At this time, in the clustering results, users with similar consumption habits and credit are clustered and grouped at the same time, effectively providing data support for subsequent credit risk assessment and credit rating classification. When there are multiple users in a class, the difference in real-time credit limit can be determined based on the mean difference.

[0087] According to the embodiment of the present invention, within one analysis cycle, the credit addition and attenuation times of all users in the user group and the real-time credit limit of the users are counted to form credit statistical data of a user group, credit risk assessment is performed based on the credit statistical data, and credit ratings are divided based on multiple user groups, each user group corresponds to a credit rating, specifically:

[0088] In one analysis cycle, the credit addition and reduction times of all users in each user group are counted, and the ratio of the times is calculated. The maximum, minimum and average values ​​of the real-time credit limit of all users are calculated, and the statistical and calculated data are integrated to form credit statistical data;

[0089] Conduct credit risk assessment for each user group based on credit statistics and generate assessment data;

[0090] Based on a user group, the real-time credit limits of the corresponding users are averaged to obtain the group credit value;

[0091] Based on the group credit value, each user group is sorted, and the corresponding credit level is generated for each user group through the sorting results.

[0092] It should be noted that the ratio of the times is the ratio of the credit bonus times to the credit decay times among all users in the group. The larger the ratio, the better the credit situation of the group. The evaluation data is comprehensive risk evaluation data for the user group, and each user group includes corresponding evaluation data. The corresponding credit level is generated for each user group through the sorting results, and the higher the level, the higher the ranking.

[0093] In the existing credit analysis system, credit analysis is often only performed on individual user information, which makes it impossible to quickly analyze and classify large-scale users, and it is difficult to perform stratified analysis and corresponding credit risk analysis on users from an overall and group perspective, resulting in low risk assessment capabilities and poor comprehensive credit analysis results for large-scale user groups. In addition, the existing technology also lacks user group classification and risk assessment for post-consumer groups, making it difficult to grasp the overall credit trend of the post-consumer group on the platform.

[0094] In the present invention, hierarchical clustering is performed on users, and users are divided into groups based on consumption behavior characteristics and corresponding real-time credit limits. The characteristics of post-consumption payment users are analyzed through the platform, and users with similar consumption habits and credit conditions can be distinguished and divided into groups. From the perspective of user groups, credit risk assessment and group credit rating analysis are performed, which reduces the analysis error caused by large individual differences, and through group credit ratings, comprehensive credit assessment of different groups can be performed.

[0095] Especially in the credit platform analysis process of large-scale users, the present invention can realize fast and accurate overall credit trend analysis and group trend analysis, provide effective data support for the credit payment platform, and facilitate the platform to mine user consumption characteristics and effectively regulate post-consumption patterns. In addition, in different analysis cycles, the user group can be updated and analyzed by updating user big data in real time, and the credit trends of different user groups can be updated in real time.

[0096] According to an embodiment of the present invention, it also includes:

[0097] Get a real-time analysis user and mark it as the current user;

[0098] Mark the user group to which the current user belongs as the current group;

[0099] In N analysis cycles, the post-consumption orders of the current user are counted, the ratio of the credit addition times to the decay times is calculated, and N current credit ratios are obtained. The ratio of the credit addition times to the decay times of the current group is obtained, and N group credit ratios are obtained.

[0100] Taking N current credit ratios as dependent variables and time as independent variable, the numerical prediction of the next analysis period is performed based on linear regression to obtain the current prediction value;

[0101] Taking the credit ratios of N groups as dependent variables and time as independent variables, the numerical prediction of the next analysis period is carried out based on linear regression to obtain the group prediction value;

[0102] If the current prediction value is less than the group prediction value, the current user is determined to be a credit risk user, and the credit status of the current user is sent to the user terminal through the postpaid credit platform.

[0103] It should be noted that each credit ratio is obtained through statistical analysis within an analysis period.

[0104] Through the embodiments of the present invention, it is possible to perform credit prediction analysis on a certain user and the user group to which he belongs, compare the predicted credit ratio of the current user with the de-grouped predicted value, effectively perform reasonable analysis and comparison on the credit trend of the user, give reasonable reminders to the user, and give reasonable risk warnings to the platform.

[0105] The current prediction value reflects the credit changes of individual users, and the group prediction value reflects the credit changes of the user group.

[0106] Figure 2 A block diagram of a user post-paid credit payment system based on big data of the present invention is shown.

[0107] The second aspect of the present invention further provides a user post-paid credit payment system 2 based on big data, the system comprising: a memory 21, a processor 22, the memory comprising a user post-paid credit payment program based on big data, the user post-paid credit payment program based on big data being executed by the processor to implement the following steps:

[0108] Obtain the user's credit information on other preset consumption platforms through the postpaid credit platform, and set the user's basic credit limit based on the credit information;

[0109] Set the deduction cycle. In each postpaid order of the user, obtain the user's payment time through the postpaid credit platform, determine whether the user's payment time is within the deduction cycle, and calculate the time difference with the deduction cycle. Perform credit addition or credit decay based on the time difference, dynamically adjust the basic credit limit, and obtain the real-time credit limit;

[0110] In one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data, extracts user features based on the user big data, obtains user consumption behavior features, uses user consumption behavior features as clustering basic data, and clusters and groups users based on hierarchical clustering to form multiple user groups;

[0111] During an analysis cycle, the number of credit additions and reductions and the real-time credit limits of all users in the user group are counted to form credit statistical data for the user group. Credit risk assessment is performed based on the credit statistical data, and credit levels are divided based on multiple user groups, with each user group corresponding to a credit level.

[0112] According to an embodiment of the present invention, the post-paid credit platform is used to obtain the user's credit information on other preset consumption platforms, and the user's basic credit limit is set based on the credit information, specifically:

[0113] Obtain user information through the postpaid credit platform;

[0114] Using user information, conduct a consumer credit query on at least one preset consumer platform to obtain corresponding credit information;

[0115] An initial comprehensive assessment of the user is conducted based on credit information to obtain a basic credit limit.

[0116] It should be noted that the preset consumption platform is generally an e-commerce platform with consumer transactions. The credit information includes information such as the credit value on the corresponding platform, such as WeChat payment points. The larger the score, the larger the corresponding basic credit limit. When calculating the basic credit limit, the credit information of multiple preset consumption platforms can also be combined for average analysis.

[0117] When making a postpaid payment, based on the basic credit limit, the platform can set the number of postpaid orders or credit limit for the user.

[0118] According to an embodiment of the present invention, the deduction cycle is set. In each postpaid order of the user, the user payment time point is obtained through the postpaid credit platform, and it is determined whether the user payment time point is within the deduction cycle, and the time difference with the deduction cycle is calculated. Credit addition or credit decay is performed according to the time difference, and the basic credit limit is dynamically adjusted to obtain the real-time credit limit, which is specifically:

[0119] Set up a deduction cycle within the postpaid credit platform;

[0120] In each postpaid order of the user, the user's payment time is obtained through the postpaid credit platform to determine whether the user's payment time is within the deduction cycle;

[0121] If yes, then calculate the time difference between the user's payment time and the deduction cycle, and use the time difference to add credit. The larger the time difference, the greater the credit addition value.

[0122] If not, the time difference between the user's payment time and the deduction cycle is calculated, and the credit decay is performed based on the time difference. The larger the time difference, the greater the credit decay value.

[0123] Dynamically adjust the basic credit limit based on credit bonus / credit decay, and get the real-time credit limit.

[0124] It should be noted that the deduction period is a limited time value. After the user makes a post-paid payment, if it is within the deduction period, the user's credit will be increased. If it is after the deduction period, the user's credit will be reduced. The specific increase / decay value is determined by the size of the time difference. The basic credit limit is the user's initial basic value. After the user registers a platform account and performs post-consumption behavior, the real-time credit limit will be used to analyze the user's payment. The basic credit limit is stored in the database as the user's basic information;

[0125] The credit bonus / decay value calculation formula is as follows:

[0126] ;

[0127] Among them, S is the credit bonus / decay value, K is the correction coefficient, Payment time for users, The deduction period.

[0128] According to an embodiment of the present invention, within one analysis cycle, the user's order data and platform interaction data are collected through the postpaid credit platform to form user big data, user features are extracted based on the user big data, and user consumption behavior features are obtained. The user consumption behavior features are used as clustering basic data and users are clustered and grouped based on hierarchical clustering to form multiple user groups, specifically:

[0129] In one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data;

[0130] Perform data cleaning, outlier removal and standardization preprocessing on user big data;

[0131] Based on user big data, extract the feature information of user order data. The feature dimensions include order amount, order time, and order quantity, and obtain the user order features.

[0132] Based on user big data, the consumption feature information of users is extracted. The feature dimensions include consumption history, consumption browsing information, and search evaluation information dimensions to obtain user consumption features;

[0133] Integrate user order characteristics and user consumption characteristics to form user consumption behavior characteristics;

[0134] The user consumption behavior characteristics are used as the clustering basic data, and the users are clustered and grouped through the hierarchical clustering method. The clustering process is as follows: initially, each user is regarded as a class, and the distance values ​​between classes are calculated based on the Manhattan distance. Then, all classes are merged cyclically according to the preset merging rules until the number of classes reaches the preset number.

[0135] The preset merging rule is based on the shortest distance method;

[0136] The users are grouped based on a preset number of classes to obtain multiple user groups, each of which corresponds to a class.

[0137] It should be noted that the user consumption behavior characteristics include the behavior characteristics of each platform user, and calculating the distance value between classes specifically means calculating the behavior characteristic distance value between users.

[0138] The preset number is generally 5 to 10.

[0139] According to an embodiment of the present invention, the preset merging rule includes:

[0140] Based on the shortest distance method, the distance value between two classes is analyzed. If the distance value is within the shortest distance range, the first condition is met;

[0141] Analyze the difference in real-time credit limits of users in the two categories, and if the difference is within a preset range, the second condition is met;

[0142] The first condition and the second condition are set as a preset merging rule.

[0143] It should be noted that in the preset merging rules, clustering and merging operations can be performed based only on the shortest distance method. At this time, the second condition, that is, clustering based on user credit limit, will not be considered, and the clustering results tend to cluster based on user consumption habits. Furthermore, the preset merging rules can also include rule settings that meet the second condition, that is, adding real-time credit limit as the basis for grouping. At this time, in the clustering results, users with similar consumption habits and credit are clustered and grouped at the same time, effectively providing data support for subsequent credit risk assessment and credit rating classification. When there are multiple users in a class, the difference in real-time credit limit can be determined based on the mean difference.

[0144] According to the embodiment of the present invention, within one analysis cycle, the credit addition and attenuation times of all users in the user group and the real-time credit limit of the users are counted to form credit statistical data of a user group, credit risk assessment is performed based on the credit statistical data, and credit ratings are divided based on multiple user groups, each user group corresponds to a credit rating, specifically:

[0145] In one analysis cycle, the credit addition and reduction times of all users in each user group are counted, and the ratio of the times is calculated. The maximum, minimum and average values ​​of the real-time credit limit of all users are calculated, and the statistical and calculated data are integrated to form credit statistical data;

[0146] Conduct credit risk assessment for each user group based on credit statistics and generate assessment data;

[0147] Based on a user group, the real-time credit limits of the corresponding users are averaged to obtain the group credit value;

[0148] Based on the group credit value, each user group is sorted, and the corresponding credit level is generated for each user group through the sorting results.

[0149] It should be noted that the ratio of the times is the ratio of the credit bonus times to the credit decay times among all users in the group. The larger the ratio, the better the credit situation of the group. The evaluation data is comprehensive risk evaluation data for the user group, and each user group includes corresponding evaluation data. The corresponding credit level is generated for each user group through the sorting results, and the higher the level, the higher the ranking.

[0150] In the existing credit analysis system, credit analysis is often only performed on individual user information, which makes it impossible to quickly analyze and classify large-scale users, and it is difficult to perform stratified analysis and corresponding credit risk analysis on users from an overall and group perspective, resulting in low risk assessment capabilities and poor comprehensive credit analysis results for large-scale user groups. In addition, the existing technology also lacks user group classification and risk assessment for post-consumer groups, making it difficult to grasp the overall credit trend of the post-consumer group on the platform.

[0151] In the present invention, hierarchical clustering is performed on users, and users are divided into groups based on consumption behavior characteristics and corresponding real-time credit limits. The characteristics of post-consumption payment users are analyzed through the platform, and users with similar consumption habits and credit conditions can be distinguished and divided into groups. From the perspective of user groups, credit risk assessment and group credit rating analysis are performed, which reduces the analysis error caused by large individual differences, and through group credit ratings, comprehensive credit assessment of different groups can be performed.

[0152] Especially in the credit platform analysis process of large-scale users, the present invention can realize fast and accurate overall credit trend analysis and group trend analysis, provide effective data support for the credit payment platform, and facilitate the platform to mine user consumption characteristics and effectively regulate post-consumption patterns. In addition, in different analysis cycles, the user group can be updated and analyzed by updating user big data in real time, and the credit trends of different user groups can be updated in real time.

[0153] The third aspect of the present invention also provides a computer-readable storage medium, which includes a user post-paid credit payment program based on big data. When the user post-paid credit payment program based on big data is executed by a processor, the steps of the user post-paid credit payment method based on big data as described in any one of the above are implemented.

[0154] The present invention discloses a user post-paid credit payment system based on big data. The post-paid credit platform integrates the user's cross-platform credit information and sets the basic credit limit. The dynamic adjustment mechanism adjusts the credit limit in real time according to the performance (addition or attenuation) of the payment time point in the deduction cycle. During the analysis period, user orders and interaction data are collected, consumption behavior characteristics are extracted, and user groups are divided by hierarchical clustering. The credit changes and real-time credit limits of each group are counted, and credit risk assessment and grading are performed to achieve precise and integrated credit analysis and management.

[0155] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0156] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0157] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0158] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0159] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0160] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A user post-paid credit payment method based on big data, characterized in that: include: Obtain the user's credit information on other preset consumption platforms through the postpaid credit platform, and set the user's basic credit limit based on the credit information; Set the deduction cycle. In each postpaid order of the user, obtain the user's payment time through the postpaid credit platform, determine whether the user's payment time is within the deduction cycle, and calculate the time difference with the deduction cycle. Perform credit addition or credit decay based on the time difference, dynamically adjust the basic credit limit, and obtain the real-time credit limit; In one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data, extracts user features based on the user big data, obtains user consumption behavior features, uses user consumption behavior features as clustering basic data, and clusters and groups users based on hierarchical clustering to form multiple user groups; In one analysis cycle, the credit addition and reduction times of all users in the user group and the real-time credit limit of users are counted to form a credit statistical data of the user group. Credit risk assessment is performed based on the credit statistical data, and credit ratings are divided based on multiple user groups, with each user group corresponding to a credit rating. In one analysis cycle, the credit addition and attenuation times of all users in the user group and the real-time credit limit of the users are counted to form a credit statistical data of the user group, and credit risk assessment is performed based on the credit statistical data. Credit ratings are divided based on multiple user groups, and each user group corresponds to a credit rating, specifically: In one analysis cycle, the credit addition and reduction times of all users in each user group are counted, and the ratio of the times is calculated. The maximum, minimum and average values ​​of the real-time credit limit of all users are calculated, and the statistical and calculated data are integrated to form credit statistical data; Conduct credit risk assessment for each user group based on credit statistics and generate assessment data; Based on a user group, the real-time credit limits of the corresponding users are averaged to obtain the group credit value; Based on the group credit value, each user group is sorted, and the corresponding credit level is generated for each user group according to the sorting results; The ratio of the times is the ratio of the credit addition times to the credit decay times among all users in the group.

2. A user post-paid credit payment method based on big data according to claim 1, characterized in that: The post-paid credit platform is used to obtain the user's credit information on other preset consumption platforms, and the user's basic credit limit is set based on the credit information, specifically: Obtain user information through postpaid credit platforms; Using user information, conduct a consumer credit query on at least one preset consumer platform to obtain corresponding credit information; An initial comprehensive assessment of the user is conducted based on credit information to obtain a basic credit limit.

3. A user post-paid credit payment method based on big data according to claim 2, characterized in that: The deduction cycle is set. In each postpaid order of the user, the user payment time point is obtained through the postpaid credit platform, and it is determined whether the user payment time point is within the deduction cycle, and the time difference with the deduction cycle is calculated. Credit addition or credit decay is performed based on the time difference, and the basic credit limit is dynamically adjusted to obtain the real-time credit limit. Specifically, Set up a deduction cycle within the postpaid credit platform; In each postpaid order of the user, the user's payment time is obtained through the postpaid credit platform to determine whether the user's payment time is within the deduction cycle; If yes, then calculate the time difference between the user's payment time and the deduction cycle, and use the time difference to add credit. The larger the time difference, the greater the credit addition value. If not, the time difference between the user's payment time and the deduction cycle is calculated, and the credit decay is performed based on the time difference. The larger the time difference, the greater the credit decay value. Dynamically adjust the basic credit limit based on credit bonus / credit decay, and get the real-time credit limit.

4. A user post-paid credit payment method based on big data according to claim 3, characterized in that: In one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data, extracts user features based on the user big data, obtains user consumption behavior features, uses user consumption behavior features as clustering basic data, and clusters and groups users based on hierarchical clustering to form multiple user groups, specifically: In one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data; Perform data cleaning, outlier removal and standardization preprocessing on user big data; Based on user big data, extract the feature information of user order data. The feature dimensions include order amount, order time, and order quantity, and obtain the user order features. Based on user big data, the consumption feature information of users is extracted. The feature dimensions include consumption history, consumption browsing information, and search evaluation information dimensions to obtain user consumption features; Integrate user order characteristics and user consumption characteristics to form user consumption behavior characteristics; The user consumption behavior characteristics are used as the clustering basic data, and the users are clustered and grouped through the hierarchical clustering method. The clustering process is as follows: initially, each user is regarded as a class, and the distance values ​​between classes are calculated based on the Manhattan distance. Then, all classes are merged cyclically according to the preset merging rules until the number of classes reaches the preset number. The preset merging rules are based on the shortest distance method; The users are grouped based on a preset number of classes to obtain multiple user groups, each of which corresponds to a class.

5. A user post-paid credit payment method based on big data according to claim 4, characterized in that: The preset merging rules include: Based on the shortest distance method, the distance value between two classes is analyzed. If the distance value is within the shortest distance range, the first condition is met; Analyze the difference in real-time credit limits of users in the two categories, and if the difference is within a preset range, the second condition is met; The first condition and the second condition are set as a preset merging rule.

6. A user post-paid credit payment system based on big data, characterized in that: The system includes: a memory and a processor, wherein the memory includes a user post-paid credit payment program based on big data, and when the user post-paid credit payment program based on big data is executed by the processor, the following steps are implemented: Obtain the user's credit information on other preset consumption platforms through the postpaid credit platform, and set the user's basic credit limit based on the credit information; Set the deduction cycle. In each postpaid order of the user, obtain the user's payment time through the postpaid credit platform, determine whether the user's payment time is within the deduction cycle, and calculate the time difference with the deduction cycle. Perform credit addition or credit decay based on the time difference, dynamically adjust the basic credit limit, and obtain the real-time credit limit; In one analysis cycle, the postpaid credit platform collects user order data and platform interaction data to form user big data, extracts user features based on the user big data, obtains user consumption behavior features, uses user consumption behavior features as clustering basic data, and clusters and groups users based on hierarchical clustering to form multiple user groups; In one analysis cycle, the credit addition and reduction times of all users in the user group and the real-time credit limit of users are counted to form a credit statistical data of the user group. Credit risk assessment is performed based on the credit statistical data, and credit ratings are divided based on multiple user groups, with each user group corresponding to a credit rating. In one analysis cycle, the credit addition and attenuation times of all users in the user group and the real-time credit limit of the users are counted to form a credit statistical data of the user group, and credit risk assessment is performed based on the credit statistical data. Credit ratings are divided based on multiple user groups, and each user group corresponds to a credit rating, specifically: In one analysis cycle, the credit addition and reduction times of all users in each user group are counted, and the ratio of the times is calculated. The maximum, minimum and average values ​​of the real-time credit limit of all users are calculated, and the statistical and calculated data are integrated to form credit statistical data; Conduct credit risk assessment for each user group based on credit statistics and generate assessment data; Based on a user group, the real-time credit limits of the corresponding users are averaged to obtain the group credit value; Based on the group credit value, each user group is sorted, and the corresponding credit level is generated for each user group according to the sorting results; The ratio of the times is the ratio of the credit addition times to the credit decay times among all users in the group.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a user post-paid credit payment program based on big data. When the user post-paid credit payment program based on big data is executed by a processor, the steps of the user post-paid credit payment method based on big data as described in any one of claims 1 to 5 are implemented.

Images