Credit card integration method, device, equipment, medium and program product

Abstract

The invention provides a credit card integration method which can be applied to the fields of big data and artificial intelligence. The credit card integration method comprises the steps of obtaining business feature data of a credit card to be integrated, and determining an initial probability that the type of the credit card to be integrated is a target type by adopting a fitting model based on the business feature data, the fitting model being obtained by training based on an evolutionary algorithm; client retention rate data generated after the previous credit card is integrated to the target credit card is obtained, the initial probability is corrected by adopting a weighting mode based on the client retention rate data, the final probability that the type of the credit card to be integrated is the target type is obtained, and the previous credit card is integrated based on a fitting model; and when the final probability is greater than a preset threshold, determining the type of the to-be-integrated credit card as a target type, and integrating the to-be-integrated credit card to the target credit card. The invention further provides a credit card integration device, credit card integration equipment, a medium and a program product.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of big data and artificial intelligence, in particular to a credit card integration method, device, equipment, medium and program product. BACKGROUND

[0002] In the current credit card business expansion process, the product diversification trend is significant. In order to meet the needs of the segmented scenarios, a large number of credit card products with highly similar functions have emerged in the market. However, from the actual operation of the business, among each type of similar products, only one product becomes the mainstream product due to high customer acceptance and strong market adaptability, while the remaining products gradually become marginal products due to low usage frequency and poor customer retention rate. If these redundant marginal products cannot be identified in a timely and accurate manner, and cannot be integrated through effective mechanisms, not only will they continue to occupy a large amount of data storage space in the system, but also will significantly increase the complexity of background data classification management, and increase the labor and time cost of system operation, thereby restricting the efficient operation of the credit card business. SUMMARY

[0003] In view of the above problems, the present application provides a credit card integration method, device, equipment, medium and program product for improving the identification accuracy of credit card marginal products and the utilization rate of data storage space.

[0004] According to a first aspect of the present application, a credit card integration method is provided, comprising: obtaining business feature data of a credit card to be integrated, determining an initial probability that the type of the credit card to be integrated is a target type based on the business feature data using a fitting model, the fitting model being trained based on an evolutionary algorithm; obtaining customer retention rate data generated after a past credit card is integrated into a target credit card, correcting the initial probability using a weighted manner based on the customer retention rate data to obtain a final probability that the type of the credit card to be integrated is the target type, the past credit card being integrated based on the fitting model; comparing the final probability with a preset threshold, and in the case that the final probability is greater than the preset threshold, determining the type of the credit card to be integrated as the target type, and integrating the credit card to be integrated into the target credit card.

[0005] According to an embodiment of the present application, obtaining the business feature data of the credit card to be integrated comprises: obtaining product rule feature data and customer activity data of the credit card to be integrated, the product rule feature data representing the product attractiveness of the credit card to be integrated from at least one dimension; and obtaining the business feature data after performing format standardization processing on the product rule feature data and the customer activity data.

[0006] According to an embodiment of the present application, obtaining the product rule feature data of the credit card to be integrated comprises: extracting a product interest calculation type, a product repayment sequence type and a product grace period from a rule library of the credit card to be integrated to obtain the product rule feature data.

[0007] According to an embodiment of this application, training a fitting model based on an evolutionary algorithm includes: randomly generating multiple equation individuals, substituting training data into each equation individual to obtain prediction results, where each equation individual represents the mapping relationship between business feature data of each dimension and the probability of the target type; obtaining business feature data of historical credit cards of the same type and the corresponding actual type labels, determining the matching degree between the prediction results and the actual type labels through a fitness function based on error correction, where the matching degree characterizes the performance of the equation individual; selecting equation individuals based on the matching degree, performing crossover and mutation on the selected equation individuals to obtain new equation individuals; repeating the steps of evaluating performance, selection, crossover, and mutation until an equation individual that meets the preset performance standard appears or the upper limit of the number of iterations is reached, and taking the equation individual with the best performance at the end of the iteration as the fitted model that has been trained.

[0008] According to the embodiments of this application, obtaining customer retention rate data after integrating past credit cards into the target credit card includes: obtaining the first customer retention rate data of the first past credit card in the previous integration process of the current integration; if integration was carried out before the previous integration process, obtaining the second customer retention rate data of the second past credit card before the previous integration process; and weighting and summing the first customer retention rate data and the second customer retention rate data to obtain the customer retention rate data.

[0009] According to the embodiments of this application, the initial probability is corrected by weighting to obtain the final probability that the type of credit card to be integrated is the target type. This includes: determining the first weighting coefficient corresponding to the customer retention rate data based on the difference between the customer retention rate data and the preset retention rate data; adding the product of the customer retention rate data and the first weighting coefficient to the product of the initial probability and the second weighting coefficient to obtain the final probability, wherein the sum of the second weighting coefficient and the first weighting coefficient is 1.

[0010] According to an embodiment of this application, integrating a credit card to be integrated into a target credit card includes: obtaining a list of business data for the credit cards to be integrated; and using a system batch processing tool to migrate the pre-processed business data list to the target credit card according to the data field mapping relationship between the credit cards to be integrated and the target credit card.

[0011] According to an embodiment of this application, the method further includes: sending credit card suspension information to a first target user before the credit card to be integrated is integrated into the target credit card; and sending credit card integration information to a second target user after the credit card to be integrated is integrated into the target credit card.

[0012] A second aspect of this application provides a credit card integration device, comprising: a fitting module for obtaining business characteristic data of the credit cards to be integrated, and determining an initial probability that the type of the credit card to be integrated is the target type based on the business characteristic data using a fitting model, wherein the fitting model is trained based on an evolutionary algorithm; a correction module for obtaining customer retention rate data generated after integrating past credit cards into the target credit card, and correcting the initial probability based on the customer retention rate data using a weighted method to obtain a final probability that the type of the credit card to be integrated is the target type, wherein past credit cards are integrated based on the fitting model; and an integration module for comparing the final probability with a preset threshold, and determining the type of the credit card to be integrated as the target type if the final probability is greater than the preset threshold, and integrating the credit card to be integrated into the target credit card.

[0013] A third aspect of this application provides an electronic device comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the method described above.

[0014] A fourth aspect of this application also provides a computer-readable storage medium having a computer program or instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0015] The fifth aspect of this application also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the steps of the above-described method. Attached Figure Description

[0016] The above-mentioned contents, other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0017] Figure 1 The illustrations depict application scenarios of credit card integration methods, apparatus, devices, media, and program products according to embodiments of this application.

[0018] Figure 2 A flowchart illustrating a credit card integration method according to an embodiment of this application is shown schematically;

[0019] Figure 3 The diagram illustrates a training flowchart of a fitting model employing feedback optimization factor correction according to an embodiment of this application.

[0020] Figure 4 A flowchart illustrating the process of integrating a credit card to be integrated into a target credit card according to an embodiment of this application is shown.

[0021] Figure 5A schematic diagram illustrating the structure of a credit card integration device according to an embodiment of this application is shown.

[0022] Figure 6 A block diagram schematically illustrates an electronic device suitable for implementing a credit card integration method according to an embodiment of this application. Detailed Implementation

[0023] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0024] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0025] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0026] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0027] In the technical solution of this application, the user information (including but not limited to user personal information, user image information, user device information, such as location information) and data (including but not limited to data used for analysis, stored data, and displayed data) involved are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with relevant laws, regulations, and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0028] In scenarios involving automated decision-making using personal information, the methods, devices, and systems provided in this application all offer users corresponding entry points for choosing to agree to or reject the automated decision-making results. If the user chooses to reject, the process proceeds to the expert decision-making stage. Here, "automated decision-making" refers to the activity of automatically analyzing and evaluating an individual's behavioral habits, interests, or economic, health, and credit status through computer programs, and then making a decision. Here, "expert decision-making" refers to the activity of making decisions by personnel who specialize in a particular field, possess specialized experience, knowledge, and skills, and have reached a certain level of professional expertise.

[0029] Figure 1 The illustration shows an application scenario diagram of the credit card integration method, apparatus, device, medium, and program product according to embodiments of this application.

[0030] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a bank staff terminal 101, a customer terminal 102, a credit card business system terminal 103, a communication network 104, and a credit card integration server 105. The communication network 104 is used as a medium to provide a communication link between the bank staff terminal 101, the customer terminal 102, the credit card business system terminal 103, and the credit card integration server 105, and may include various connection types such as wired communication links (e.g., leased network), wireless communication links (e.g., 4G / 5G mobile network), or fiber optic cables.

[0031] Bank staff can access the communication network 104 through bank staff terminals 101 (such as branch desktop computers or mobile office tablets) to interact with the credit card integration server 105, such as uploading business rule data of credit cards to be integrated, viewing edge product identification results, and monitoring integration progress. Customers can log in to the mobile banking application or online banking through customer terminals 102 (such as smartphones or home computers) to receive information such as credit card suspension notifications, integration completion prompts, and activation instructions for target credit cards. The credit card business system terminal 103 (such as the bank's core business system terminal) can synchronize customer activity data, historical transaction data, and past customer retention rate data of credit cards to be integrated to the credit card integration server 105 through the communication network 104, providing data support for integration decisions.

[0032] The bank staff terminal 101, customer terminal 102, and credit card business system terminal 103 can be various electronic devices with data interaction and display functions, including but not limited to smartphones, tablets, laptops, desktop computers, and dedicated business terminals.

[0033] The credit card integration server 105 can be a dedicated server providing services such as credit card edge product identification, integration decision-making, and data migration, for example, a back-end processing server equipped with an evolutionary algorithm fitting model. This server can receive business characteristic data synchronized from the credit card business system terminal 103, calculate the initial and final probabilities that the credit card to be integrated is an edge product; generate integration instructions based on the probability determination results, push the integration task to the bank staff terminal 101, and simultaneously send relevant notifications to the customer terminal 102; it can also coordinate with the credit card business system terminal 103 to complete the batch migration of the credit card business data to be integrated to the target credit card, ensuring a closed-loop integration process.

[0034] It should be noted that the credit card integration method provided in this application embodiment can generally be executed by the credit card integration server 105, and correspondingly, the credit card integration device provided in this application embodiment can generally be set in the credit card integration server 105. The credit card integration method provided in this application embodiment can also be executed by a server cluster (such as a bank cloud business server cluster) that is different from the credit card integration server 105 and can communicate with the bank staff terminal 101, customer terminal 102, credit card business system terminal 103 and / or credit card integration server 105. Correspondingly, the credit card integration device provided in this application embodiment can also be set in the server cluster.

[0035] It should be understood that Figure 1 The number of terminal devices, communication networks, and servers shown is merely illustrative. Depending on the actual business deployment needs, there can be any number of terminal devices (such as adding more staff terminals at more outlets, customer terminals in different regions), communication networks (such as distinguishing between intranets and extranets), and servers (such as splitting data storage servers and model calculation servers).

[0036] The following will be based on Figure 1 The described scene, through Figures 2-4 The credit card integration method according to the embodiments of this application will be described in detail.

[0037] Figure 2 A flowchart illustrating a credit card integration method according to an embodiment of this application is shown.

[0038] like Figure 2 As shown, the credit card integration method of this embodiment includes operations S210 to S230, and the credit card integration method can be executed by server 105.

[0039] In operation S210, business characteristic data of the credit cards to be integrated is obtained. Based on the business characteristic data, a fitting model is used to determine the initial probability that the type of the credit card to be integrated is the target type. The fitting model is trained based on an evolutionary algorithm.

[0040] In operation S220, customer retention rate data is obtained after past credit cards are integrated into the target credit card. Based on the customer retention rate data, the initial probability is corrected using a weighted method to obtain the final probability that the type of the credit card to be integrated is the target type. Past credit cards are integrated based on a fitted model.

[0041] In operation S230, the final probability is compared with a preset threshold. If the final probability is greater than the preset threshold, the type of credit card to be integrated is determined as the target type, and the credit card to be integrated is integrated into the target credit card.

[0042] The bank's credit card business system can periodically or irregularly extract business characteristic data of credit cards to be integrated from the associated database. This business characteristic data can include various dimensions of features, reflecting the business situation of the credit cards to be integrated from different perspectives. Credit cards to be integrated can be redundant marginal credit card products in the bank's system, or specific types designated by the system. Marginal products refer to credit card products similar to mainstream products but with low customer activity rates. Before processing the business characteristic data in the fitting model, the feature data of different dimensions can be preprocessed, such as format standardization, to form standardized combinations of business characteristic data. Then, the business characteristic data is substituted into the fitting model trained based on an evolutionary algorithm to obtain the initial probability that the credit card is the target type (e.g., a marginal product), which can be any value between 0 and 1. The larger the value, the higher the probability that the credit card to be integrated is the target type. Next, the customer retention rate data of credit cards integrated based on this model is extracted, which can be denoted as λ. Customer retention rate data (or customer retention rate) refers to the probability that customers holding marginal products will continue to hold and effectively use the credit card product within the observation period after product integration. For example, based on the number of customers in group A (the product to be integrated before integration) and group A' (the integrated product) who continue to hold and effectively use credit card products within group A after a specific observation period following integration, the customer retention rate is the ratio of A' to A. Then, the customer retention rate λ is used as a feedback optimization factor for the fitted model, and weights are determined accordingly to correct the initial probability obtained from the fitted model. The weights in the correction can be set based on past customer retention rates or empirical values. The corrected final probability is then compared with a preset threshold to determine whether the credit card to be integrated can be integrated into the target credit card. The preset threshold can be determined based on the average customer retention rate of the target type of credit cards integrated historically.

[0043] This embodiment acquires multi-dimensional business characteristic data of the credit cards to be integrated, calculates the initial probability based on the fitting model trained by the evolutionary algorithm, and then corrects it with the customer retention rate data of previous integrations to obtain the final probability, forming an adaptive and self-optimizing intelligent identification mechanism. This replaces the traditional single static attribute identification method, which can more accurately determine whether a credit card is a marginal product and reduce false positives and false negatives. Through the judgment mechanism of comparing the final probability with a preset threshold, when the probability meets the condition, the credit card to be integrated is automatically identified as a marginal product and integrated into the target credit card without manual intervention. This realizes an automated closed loop from identification to integration of marginal products, avoids the long-term accumulation of redundant marginal products, effectively releases system data storage space, improves data storage control and utilization efficiency, and reduces data management and maintenance costs.

[0044] In some embodiments, product rule feature data and customer activity data of the credit cards to be integrated can be obtained. The product rule feature data characterizes the product attractiveness of the credit cards to be integrated from at least one dimension. Then, after standardizing the format of the product rule feature data and customer activity data, business feature data is obtained.

[0045] As an example, product rule feature data can be extracted from credit card product rules. This data includes core rules influencing customer choices and can reflect the competitiveness of a credit card from the perspective of product attractiveness. Customer activity refers to the probability that a credit card holder will use the card within a specific period. Customer activity data can be extracted from a customer behavior database, for example, by statistically analyzing indicators such as the frequency of card use and the percentage of spending over the past M days; the value can be between 0-100%. Then, the business feature data of these two dimensions undergoes format standardization. For example, non-numerical indicators in the product rule feature data are mapped to 0-1 values, and customer activity data is normalized, such as converting 0-100% activity values ​​into 0-1 range values. This ultimately forms multi-dimensional business feature data that can be directly input into the fitting model, ensuring the accuracy and consistency of the model calculations.

[0046] This embodiment acquires business feature data from two dimensions: product rule features and customer activity, and standardizes the data input to the model to ensure that the data format of the input model is uniform and can be directly calculated, thus providing a data foundation for the accuracy of the initial probability.

[0047] In some embodiments, product interest calculation type, product repayment order type, and product grace period can be extracted from the rule base of the credit card to be integrated to obtain product rule feature data.

[0048] Product interest calculation type refers to the method of calculating interest on outstanding credit card balances after the interest-free period has expired. It can be divided into full penalty interest and partial penalty interest. Full penalty interest calculates interest on the full amount due from the billing date to the repayment date, plus interest on the outstanding balance from the due date of the previous billing cycle to the current billing date. Partial penalty interest calculates interest on the outstanding balance from the due date of the previous billing cycle to the current billing date. Repayment order type refers to the order in which overdraft transactions are deducted from the repayment account. Generally, for accounts overdue for less than 90 days, interest and fees are deducted first, then principal; for accounts overdue for 90 days or more, principal is deducted first, then interest and fees. Grace period refers to a certain period (N days) after the due date when repayment is still considered normal repayment, without penalty fees or interest. N days is the grace period.

[0049] As an example, when extracting product rule feature data from the proprietary rule base of the credit cards to be integrated, the bank's self-developed rule extraction tool can be used to filter key information according to preset data dimensions. For example, first, locate the "Product Interest Calculation Type" field to filter out the interest calculation method for outstanding payments after the interest-free period; second, find the "Repayment Order Type" configuration module to determine the order of overdraft transaction reduction (interest first, then principal or principal first, then interest) for different overdue stages (within 90 days / over 90 days); finally, retrieve the "Grace Period Setting" parameter to obtain the number of days (e.g., 3 days, 5 days) that are still considered normal repayments after the due date. During the extraction process, the system can automatically verify data integrity. If a field is missing, it will be filled with the mode of the same type of credit card field to ensure that the product rule feature data is complete and valid.

[0050] This embodiment refines the specific dimensions of product rule features into interest calculation type, repayment order, and grace period. These dimensions directly affect the product attractiveness of credit cards. Targeted extraction can enhance the correlation between features and target types, and further improve the accuracy of model judgment.

[0051] Evolutionary algorithms are based on Darwinian evolutionary theory, using fitness as a basis and drawing on natural selection mechanisms such as survival of the fittest and natural heredity. They achieve an iterative, random search process for population restructuring by performing evolutionary operator operations on individuals within the population. This paper incorporates a feedback optimization factor correction mechanism into the evolutionary algorithm; the feedback optimization factor is derived from customer retention rate. Figure 3 The diagram illustrates a training flowchart of a fitting model employing feedback optimization factor correction according to an embodiment of this application.

[0052] like Figure 3As shown, during model training, parameters can be set first, such as the probabilities of evolutionary operators like mutation and crossover, population size (the number of individuals participating in training), maximum number of iterations N, and termination conditions (the number of iterations reaches N or the fitness value of the best individual in the population is less than a certain threshold). Each parameter can be set according to different scenarios. Parameter settings also include a feedback optimization factor λ. The feedback optimization factor λ can be derived from past customer retention rates, which will be further explained in the examples below.

[0053] Next, the population is initialized. Multiple equation individuals are randomly generated, and the training data is substituted into each equation individual to obtain the prediction results. Each equation individual represents the mapping relationship between business feature data of each dimension and the probability of the target type. For example, each equation individual can be an equation composed of overdue days (x), interest type (y), customer qualification (u), repayment rate (v), and selectable operators (such as addition, subtraction, multiplication, division, trigonometric functions, and powers). The length of each independent individual is consistent, such as (u / v)1 / 2-x*c. Consistent length means that the sum of operators and variables in the equation is consistent.

[0054] Next, individuals are selected based on their matching degree. Historical business characteristic data of similar credit cards and their corresponding actual type labels are obtained. The matching degree between the predicted results and the actual type labels is determined using a fitness function based on error correction. The matching degree characterizes the performance of an individual in the equation. For example, the individual fitness function can be expressed as: F(x) = 1 / ((|f(x)-1|) 2 +1), where f(x) is the individual result of the population equation, that is, the probability of conforming to the same type of marginal product. The larger F(x) is, the better the fitness of the individual, that is, the higher the matching degree.

[0055] Next, operator crossover and mutation are performed. Individuals are selected based on their matching degree. These selected individuals are then subjected to crossover and mutation to obtain new individuals, thus determining the winning individuals. For example, during crossover, two individuals can be randomly selected to exchange some genes (such as combinations of operators and variables), and during mutation, the operators or variables of the individuals can be randomly adjusted.

[0056] The iterative process is then repeated. The steps of evaluating performance, screening, crossover, and mutation are repeated until an individual equation meets the preset performance criteria or the maximum number of iterations is reached. The individual equation with the best performance at the end of the iteration is then taken as the completed fitted model. It should be noted that the fitted model at this point is the initial, uncorrected model.

[0057] Next, corrections are made. The initial-state model is corrected using feedback optimization factors to obtain the final-state model. After evolutionary training, the optimal individual is selected, and then the equation is reversed according to the binary tree encoding rules. For example, the initial-state fitting model f(x,y,u,v) for judging edge credit card products of the same type is derived based on influencing factors such as product interest type (x), product repayment order type (y), product grace period (u), and customer activity (v). To improve model accuracy, the customer retention rate after the initial product integration is further introduced as a feedback optimization factor λ for the fitting model, forming the final-state fitting model F(x,y,u,v), where F(x,y,u,v) is the weighted result of f(x,y,u,v) and λ.

[0058] This embodiment completes the training of the initial fitting model through the processes of generating individuals, evaluating matching degree, screening, cross-mutation, and iterative optimization. The initial model is then corrected by feedback optimization factors to obtain the final fitting model, thereby improving the accuracy of model recognition.

[0059] Please continue reading. Figure 3 The feedback optimization factor can be obtained from two customer retention rates. In some embodiments, the first customer retention rate data of the first past credit card in the previous integration process can be obtained, and the first customer retention rate λ1 can be calculated. If an integration was performed before the previous integration process, the second customer retention rate data of the second past credit card before the previous integration process can be obtained, and the second customer retention rate λ2 can be calculated. Then, the first customer retention rate data and the second customer retention rate data are weighted and summed to obtain the customer retention rate data, that is, the feedback optimization factor λ, which can be expressed as λ=ελ1+(1-ε)λ2, where ε is the weight factor, ε∈(0,1).

[0060] In some embodiments, if no other rounds of integration have been performed prior to the previous integration process, the first customer retention rate λ1 can be directly used as λ. That is, ε=1.

[0061] In some embodiments, the second customer retention rate data of the second past credit card prior to the last integration process can be the average of customer retention rate data from multiple rounds of integration.

[0062] This embodiment weights and merges customer retention rate data from multiple previous integrations, taking into account both recent integration effects and historical experience, making the retention data more representative and improving the reliability of initial probability correction.

[0063] In some embodiments, a first weighting coefficient corresponding to the customer retention rate data can be determined based on the difference between the customer retention rate data and the preset retention rate data; the final probability is obtained by adding the product of the customer retention rate data and the first weighting coefficient to the product of the initial probability and the second weighting coefficient, and the sum of the second weighting coefficient and the first weighting coefficient is 1.

[0064] For example, the final-state fitting model described above can be expressed as: F(x,y,u,v)=c*λ+(1-c)*f(x,y,u,v), where c is the first weighting coefficient and 1-c is the second weighting coefficient. The higher the first weighting coefficient, the closer the customer retention effect is to the expectation.

[0065] This embodiment uses a first weighted coefficient to associate customer retention rate data and a second weighted coefficient to associate the initial probability to achieve probability correction, balancing the impact of model prediction and actual retention effect, so that the final probability is more in line with business objectives and achieves accurate integration under high retention rate.

[0066] Figure 4 A flowchart illustrating the process of integrating a credit card to be integrated into a target credit card according to an embodiment of this application is shown.

[0067] like Figure 4 As shown, in some embodiments, during the process of integrating the credit card to be integrated into the target credit card, a list of business data for the credit card to be integrated can be obtained first. Then, the parameters of the credit card to be integrated are invalidated. Then, using the system's batch processing tool, the pre-processed business data list is migrated to the target credit card according to the data field mapping relationship between the credit card to be integrated and the target credit card.

[0068] As an example, when integrating a credit card to be integrated into a target credit card, data extraction tools such as Extract-Transform-Load (ETL) can be used to filter the business data list of the credit card to be integrated from the business database. The business data list can include basic customer information such as name, ID number, and registered mobile phone number. It can also include core account information such as credit limit, outstanding balance, points balance, and activated value-added services. Furthermore, it can include transaction history information such as consumption / repayment records and overdue records for the past year. The data list is then preprocessed, including but not limited to removing duplicate customer records, completing missing contact information, and verifying data consistency. Then, the bank's batch processing tools are used to migrate the data according to preset data field mapping relationships. For example, the "points balance" of the credit card to be integrated is mapped 1:1 to the target credit card's points account, "outstanding balance" is associated with the target credit card's repayment statement, and "value-added services" are simultaneously activated for the target credit card. After the migration is complete, data integrity is automatically verified to ensure a high migration success rate.

[0069] This embodiment integrates credit cards by acquiring a business data list, batch migration, and field mapping, ensuring that the business data of the credit cards to be integrated are completely and accurately migrated to the target credit cards, avoiding data loss or disorder, and ensuring the continuity of the customer's card usage experience.

[0070] Please continue reading. Figure 4 In some embodiments, before integrating the credit card to be integrated into the target credit card, the issuance of the credit card to be integrated is stopped, and a credit card issuance stoppage message is sent to the first target user. After integrating the credit card to be integrated into the target credit card, a credit card integration message is sent to the second target user.

[0071] For example, before credit card integration begins, a suspension notice can be sent to the first target user (such as the holder of the credit card to be integrated) via the bank's SMS platform. This notice includes, but is not limited to, the name of the credit card to be integrated, the reason for suspension, the suspension period, and the subsequent integration plan. After credit card integration is completed, integration information can be sent to the second target user (who may be the same as the first target user) via push notifications and / or SMS notifications through the mobile banking application. This information includes, but is not limited to, the target credit card activation link, the original card data migration status, the target card's core benefits, and customer service contact channels. The system automatically records the sending status after notification. For customers whose notifications were not received, a follow-up call can be made with customer service representatives.

[0072] This embodiment addresses the risk of complaints or customer churn caused by information asymmetry by synchronizing customer information before and after integration, while also strengthening business compliance and improving customer satisfaction.

[0073] Based on the above-described credit card integration method, this application also provides a credit card integration device. The following will combine... Figure 5 The device is described in detail.

[0074] Figure 5 A schematic block diagram of a credit card integration device according to an embodiment of this application is shown.

[0075] like Figure 5 As shown, the credit card integration device 500 of this embodiment includes a fitting module 510, a correction module 520 and an integration module 530.

[0076] The fitting module 510 can be used to obtain business feature data of the credit cards to be integrated. Based on the business feature data, a fitting model is used to determine the initial probability that the type of the credit card to be integrated is the target type. The fitting model is trained based on an evolutionary algorithm. In one embodiment, the fitting module 510 can be used to perform the operation S210 described above, which will not be repeated here.

[0077] The correction module 520 can be used to obtain customer retention rate data generated after integrating past credit cards into the target credit card. Based on the customer retention rate data, the initial probability is corrected using a weighted method to obtain the final probability that the type of the credit card to be integrated is the target type. The past credit cards are integrated based on a fitted model. In one embodiment, the correction module 520 can be used to perform the operation S220 described above, which will not be repeated here.

[0078] The integration module 530 can be used to compare the final probability with a preset threshold. If the final probability is greater than the preset threshold, the type of the credit card to be integrated is determined as the target type, and the credit card to be integrated is integrated into the target credit card. In one embodiment, the integration module 530 can be used to perform the operation S230 described above, which will not be repeated here.

[0079] According to an embodiment of this application, the fitting module 510 can be used to obtain product rule feature data and customer activity data of the credit card to be integrated. The product rule feature data characterizes the product attractiveness of the credit card to be integrated from at least one dimension. After the product rule feature data and customer activity data are processed for format standardization, business feature data is obtained.

[0080] According to an embodiment of this application, the fitting module 510 can be used to extract product interest type, product repayment order type and product grace period from the rule base of the credit card to be integrated, and obtain product rule feature data.

[0081] According to an embodiment of this application, the credit card integration device 500 includes a training module (not shown). The training module is used to train a fitting model based on an evolutionary algorithm, including: randomly generating multiple equation individuals, substituting training data into each equation individual to obtain prediction results, where each equation individual represents the mapping relationship between business feature data of each dimension and the probability of the target type; obtaining business feature data of historical credit cards of the same type and the corresponding actual type labels, determining the matching degree between the prediction results and the actual type labels through a fitness function based on error correction, where the matching degree characterizes the performance of the equation individual; selecting equation individuals based on the matching degree, performing crossover and mutation on the selected equation individuals to obtain new equation individuals; repeating the steps of evaluating performance, selection, crossover, and mutation until an equation individual that meets the preset performance standard appears or the upper limit of the number of iterations is reached, and taking the equation individual with the best performance at the end of the iteration as the fitted model that has been trained.

[0082] According to an embodiment of this application, the correction module 520 can be used to obtain the first customer retention rate data of the first past credit card in the previous integration process during the current integration; if an integration was performed before the previous integration process, it can obtain the second customer retention rate data of the second past credit card before the previous integration process; and it can perform a weighted summation of the first customer retention rate data and the second customer retention rate data to obtain the customer retention rate data.

[0083] According to an embodiment of this application, the correction module 520 can be used to determine the first weighting coefficient corresponding to the customer retention rate data based on the difference data between the customer retention rate data and the preset retention rate data; and to add the product of the customer retention rate data and the first weighting coefficient to the product of the initial probability and the second weighting coefficient to obtain the final probability, wherein the sum of the second weighting coefficient and the first weighting coefficient is 1.

[0084] According to an embodiment of this application, the integration module 530 can be used to obtain a list of business data of the credit cards to be integrated; and through the system batch processing tool, the pre-processed business data list is migrated to the target credit card according to the data field mapping relationship between the credit cards to be integrated and the target credit card.

[0085] According to an embodiment of this application, the credit card integration device 500 includes a sending module (not shown). The sending module is used to send credit card suspension information to a first target user before integrating the credit card to be integrated into the target credit card; and to send credit card integration information to a second target user after integrating the credit card to be integrated into the target credit card.

[0086] According to embodiments of this application, any multiple modules described above can be combined into one module, or any one of these modules can be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules can be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this application, at least one of the above modules can be at least partially implemented as hardware circuitry, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a System-on-Chip, a System-on-Substrate, a System-on-Package, an Application-Specific Integrated Circuit (ASIC), or implemented in hardware or firmware through any other reasonable means of integrating or packaging circuitry, or implemented in any one of software, hardware, and firmware methods, or in a suitable combination of any of these. Alternatively, at least one of the above modules can be at least partially implemented as a computer program module, which, when run, can perform corresponding functions.

[0087] Figure 6 A block diagram schematically illustrates an electronic device suitable for implementing a credit card integration method according to an embodiment of this application.

[0088] like Figure 6 As shown, an electronic device 600 according to an embodiment of this application includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage portion 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of this application.

[0089] RAM 603 stores various programs and data required for the operation of electronic device 600. Processor 601, ROM 602, and RAM 603 are interconnected via bus 604. Processor 601 executes various operations of the method flow according to embodiments of this application by executing programs in ROM 602 and / or RAM 603. It should be noted that programs may also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 may also execute various operations of the method flow according to embodiments of this application by executing programs stored in one or more memories.

[0090] According to embodiments of this application, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to a bus 604. The electronic device 600 may also include one or more of the following components connected to the input / output (I / O) interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output (I / O) interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 610 as needed so that computer programs read from it can be installed into the storage section 608 as needed.

[0091] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of this application.

[0092] According to embodiments of this application, the computer-readable storage medium can be a non-volatile computer-readable storage medium, such as including but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this application, the computer-readable storage medium may include ROM 602 and / or RAM 603 and / or one or more memories other than ROM 602 and RAM 603 described above.

[0093] Embodiments of this application also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code is used to enable the computer system to implement the credit card integration method provided in the embodiments of this application.

[0094] When the computer program is executed by the processor 601, it performs the functions defined in the system / apparatus of this application embodiment. According to the embodiments of this application, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0095] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via the communication section 609, and / or installed from the removable medium 611. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0096] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, it performs the functions defined in the system of this application embodiment. According to the embodiments of this application, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0097] According to embodiments of this application, program code for executing the computer programs provided in the embodiments of this application can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0098] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0099] Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments of this application can be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

Claims

1. A method for integrating credit cards, characterized in that, The method includes: Obtain business characteristic data of the credit cards to be integrated, and based on the business characteristic data, use a fitting model to determine the initial probability that the type of the credit cards to be integrated is the target type. The fitting model is trained based on an evolutionary algorithm. Customer retention rate data generated after integrating past credit cards into the target credit card is obtained. Based on the customer retention rate data, the initial probability is corrected using a weighted method to obtain the final probability that the type of the credit card to be integrated is the target type. The past credit cards are integrated based on the fitted model. The final probability is compared with a preset threshold. If the final probability is greater than the preset threshold, the type of the credit card to be integrated is determined as the target type, and the credit card to be integrated is integrated into the target credit card.

2. The credit card integration method according to claim 1, characterized in that, The business characteristic data of the credit cards to be integrated includes: Obtain product rule feature data and customer activity data of the credit card to be integrated, wherein the product rule feature data characterizes the product attractiveness of the credit card to be integrated from at least one dimension; After standardizing the format of the product rule feature data and the customer activity data, the business feature data is obtained.

3. The credit card integration method according to claim 2, characterized in that, The product rule feature data obtained for the credit card to be integrated includes: Extract the product interest calculation type, product repayment order type, and product grace period from the rule base of the credit card to be integrated to obtain the product rule feature data.

4. The credit card integration method according to claim 1, characterized in that, Training the fitted model based on the evolutionary algorithm includes: Multiple individual equations are randomly generated. The training data is substituted into each individual equation to obtain the prediction result. Each individual equation represents the mapping relationship between business feature data of each dimension and the probability of target type. Obtain historical business characteristic data of similar credit cards and corresponding actual type labels, and determine the matching degree between the prediction result and the actual type label through an error-corrected fitness function. The matching degree characterizes the performance of the individual equation. Based on the matching degree, individuals of the equation are selected, and crossover and mutation are performed on the selected individuals of the equation to obtain new individuals of the equation. Repeat the steps of evaluating performance, screening, crossover, and mutation until an individual equation that meets the preset performance criteria is found or the maximum number of iterations is reached. The individual equation with the best performance at the end of the iteration is taken as the fitted model that has been trained.

5. The credit card integration method according to claim 1, characterized in that, The customer retention rate data generated after integrating previous credit cards into the target credit card includes: Obtain the first customer retention rate data of the first credit card in the previous integration process during the current integration; If an integration was performed prior to the previous integration process, obtain the second customer retention rate data of the second past credit card prior to the previous integration process; The customer retention rate data is obtained by weighted summing of the first customer retention rate data and the second customer retention rate data.

6. The credit card integration method according to claim 1, characterized in that, The step of correcting the initial probability using a weighted method to obtain the final probability that the type of the credit card to be integrated is the target type includes: Based on the difference between the customer retention rate data and the preset retention rate data, a first weighting coefficient corresponding to the customer retention rate data is determined; The final probability is obtained by adding the product of the customer retention rate data and the first weighting coefficient to the product of the initial probability and the second weighting coefficient, where the sum of the second weighting coefficient and the first weighting coefficient is 1.

7. The credit card integration method according to claim 1, characterized in that, The process of integrating the credit card to be integrated into the target credit card includes: Obtain the business data list of the credit cards to be integrated; Using the system's batch processing tool, the pre-processed business data list is migrated to the target credit card according to the data field mapping relationship between the credit card to be integrated and the target credit card.

8. The credit card integration method according to claim 1 or 7, characterized in that, The method further includes: Before the credit card to be integrated is integrated into the target credit card, a credit card issuance suspension information is sent to the first target user; After the credit card to be integrated is integrated into the target credit card, the credit card integration information is sent to the second target user.

9. A credit card integration device, characterized in that, The device includes: The fitting module is used to obtain the business feature data of the credit cards to be integrated, and based on the business feature data, a fitting model is used to determine the initial probability that the type of the credit cards to be integrated is the target type. The fitting model is trained based on an evolutionary algorithm. The correction module is used to obtain customer retention rate data after integrating past credit cards into the target credit card, and to correct the initial probability based on the customer retention rate data using a weighted method to obtain the final probability that the type of the credit card to be integrated is the target type. The past credit cards are integrated based on the fitted model. An integration module is used to compare the final probability with a preset threshold, and if the final probability is greater than the preset threshold, determine the type of the credit card to be integrated as the target type, and integrate the credit card to be integrated into the target credit card.

10. An electronic device, comprising: One or more processors; Memory, used to store one or more computer programs. The characteristic feature is that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 8.

12. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method according to any one of claims 1 to 8.

Images