Credit card customer classification method and device

Abstract

The invention provides a credit card customer classification method and device, relates to the technical field of artificial intelligence, and can be applied to the technical field of finance or other technical fields. The credit card customer classification method comprises the steps of obtaining target credit card customer data, and performing customer screening on the target credit card customer data; inputting the screened target credit card customer data into a credit card customer classification model to obtain a credit card customer classification result; wherein the credit card customer classification model is obtained by training based on a historical credit card training data set, a plurality of historical credit card sub-training data sets and a corresponding credit card classification model. The credit card customer group can be optimized, and the quality of the credit card customers can be improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a credit card customer classification method and device. Background Art

[0002] At present, credit cards have become an important source of profit for the banking industry and an important means of developing customers. The competition in the credit card market is becoming increasingly fierce. How to do a good job in credit card business is the focus of major banks.

[0003] Credit card customers are key customer groups for banks. Research on each stage of their life cycle is a means to understand the important development changes and important behavioral characteristics of the customer group, and is also an important basis for banks to adjust their financial services according to customer needs. Activating new customers in the credit card customer group, awakening dormant customers, and subsequent quality improvement work are also key business issues in the credit card business.

[0004] With the development of the information age, major banks are currently undergoing digital transformation, and artificial intelligence technology has been widely used in various banking businesses. How to apply artificial intelligence technology to credit card business to promote the rapid development of credit card business will also be the focus of major banks in developing credit card business. Summary of the invention

[0005] The main purpose of the embodiments of the present invention is to provide a credit card customer classification method and device to optimize the credit card customer group and improve the quality of credit card customers.

[0006] In order to achieve the above object, an embodiment of the present invention provides a credit card customer classification method, comprising:

[0007] Acquire target credit card customer data, and perform customer screening on the target credit card customer data;

[0008] The screened target credit card customer data is input into a credit card customer classification model to obtain a credit card customer classification result; wherein the credit card customer classification model is trained based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification models.

[0009] In one embodiment, the steps of creating a credit card customer classification model include:

[0010] Dividing the historical credit card training data set into the plurality of historical credit card sub-training data sets;

[0011] Inputting the plurality of historical credit card sub-training data sets into the corresponding first credit card classification model to obtain corresponding first credit card data prediction results;

[0012] Inputting the features corresponding to the plurality of first credit card data prediction results and the historical credit card training data set into a second credit card classification model to obtain a credit card data prediction result;

[0013] Determine a loss function according to the credit card classification prediction result and the corresponding credit card actual result, and update the first credit card classification model and the second credit card classification model according to the loss function until the loss function converges;

[0014] The credit card customer classification model is created according to the first credit card classification model and the second credit card classification model after the loss function converges.

[0015] In one embodiment, it also includes:

[0016] Inputting the historical credit card training data set into a credit card feature model to obtain historical credit card features;

[0017] The historical credit card training data set is screened according to the comparison result between the historical credit card feature and the feature threshold to obtain the feature corresponding to the historical credit card training data set.

[0018] In one embodiment, inputting the plurality of first credit card data prediction results and features corresponding to the historical credit card training data set into a second credit card classification model to obtain the credit card data prediction results includes:

[0019] splicing the plurality of first credit card data prediction results and features corresponding to the historical credit card training data set to obtain a credit card feature splicing result;

[0020] The credit card feature concatenation result is input into the second credit card classification model to obtain the credit card data prediction result.

[0021] In one embodiment, dividing the historical credit card training data set into the plurality of historical credit card sub-training data sets comprises:

[0022] Randomly assigning data positions in the historical credit card training data set;

[0023] The randomly assigned data is divided into the multiple historical credit card sub-training data sets.

[0024] In one embodiment, screening the target credit card customer data includes:

[0025] Determine blacklist customer data based on the customer credit card knowledge graph and the target credit card customer data;

[0026] The target credit card customer data is screened according to the blacklist customer data.

[0027] In one embodiment, it also includes:

[0028] Determine credit card entity data and credit card relationship data based on original customer credit card data;

[0029] The customer credit card knowledge graph is constructed based on the credit card entity data and the credit card relationship data.

[0030] The embodiment of the present invention also provides a credit card customer classification device, comprising:

[0031] A customer screening module, used to obtain target credit card customer data and perform customer screening on the target credit card customer data;

[0032] The credit card customer classification module is used to input the screened target credit card customer data into the credit card customer classification model to obtain the credit card customer classification result; wherein, the credit card customer classification model is trained based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification models.

[0033] In one embodiment, the credit card customer classification device further comprises:

[0034] A training data set division module, used for dividing the historical credit card training data set into the plurality of historical credit card sub-training data sets;

[0035] A first credit card data prediction module, used for inputting the plurality of historical credit card sub-training data sets into a corresponding first credit card classification model to obtain a corresponding first credit card data prediction result;

[0036] A second credit card data prediction module, used for inputting the plurality of first credit card data prediction results and features corresponding to the historical credit card training data set into a second credit card classification model to obtain a credit card data prediction result;

[0037] A loss function module, used to determine a loss function according to the credit card classification prediction result and the corresponding credit card actual result, and update the first credit card classification model and the second credit card classification model according to the loss function until the loss function converges;

[0038] A credit card customer classification model creation module is used to create the credit card customer classification model based on the first credit card classification model and the second credit card classification model after the loss function converges.

[0039] In one embodiment, it also includes:

[0040] A historical credit card feature module, used to input the historical credit card training data set into a credit card feature model to obtain historical credit card features;

[0041] The screening module is used to screen the historical credit card training data set according to the comparison result between the historical credit card feature and the feature threshold, so as to obtain the feature corresponding to the historical credit card training data set.

[0042] In one embodiment, the second credit card data prediction module includes:

[0043] A splicing unit, used for splicing the plurality of first credit card data prediction results and features corresponding to the historical credit card training data set to obtain a credit card feature splicing result;

[0044] The credit card data prediction result unit is used to input the credit card feature splicing result into the second credit card classification model to obtain the credit card data prediction result.

[0045] In one embodiment, the training data set partitioning module includes:

[0046] A random unit, used for randomly assigning data positions in the historical credit card training data set;

[0047] The partitioning unit is used to partition the randomly allocated data into the plurality of historical credit card sub-training data sets.

[0048] In one embodiment, the customer screening module includes:

[0049] A blacklist customer unit, used to determine blacklist customer data based on the customer credit card knowledge graph and the target credit card customer data;

[0050] A customer screening unit is used to screen the target credit card customer data according to the blacklist customer data.

[0051] In one embodiment, it also includes:

[0052] An entity relationship determination module, used to determine credit card entity data and credit card relationship data based on original customer credit card data;

[0053] A customer credit card knowledge graph construction module is used to construct the customer credit card knowledge graph based on the credit card entity data and the credit card relationship data.

[0054] An embodiment of the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the credit card customer classification method are implemented.

[0055] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the credit card customer classification method are implemented.

[0056] The embodiment of the present invention also provides a computer program product, including a computer program / instruction, which implements the steps of the credit card customer classification method when the computer program / instruction is executed by a processor.

[0057] The credit card customer classification method and device of the embodiment of the present invention screens target credit card customer data, obtains a credit card customer classification model based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification model training, inputs the screened target credit card customer data into the credit card customer classification model, and obtains a credit card customer classification result, which can optimize the credit card customer group and improve the quality of credit card customers. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1 This is a credit card full life cycle process analysis diagram in an embodiment of the present invention;

[0060] Figure 2 is a flow chart of a credit card customer classification method according to an embodiment of the present invention;

[0061] Figure 3 is a flow chart of a credit card customer classification method in another embodiment of the present invention;

[0062] Figure 4 is a flow chart of S101 in an embodiment of the present invention;

[0063] Figure 5 is a flow chart of constructing a customer credit card knowledge graph in an embodiment of the present invention;

[0064] Figure 6 is a flow chart of creating a credit card customer classification model in an embodiment of the present invention;

[0065] Figure 7 is a flow chart of S401 in an embodiment of the present invention;

[0066] Figure 8 is a flow chart of obtaining features corresponding to a historical credit card training data set in an embodiment of the present invention;

[0067] Fig. 9 is a flow chart of S403 in an embodiment of the present invention;

[0068] Fig.10 is a schematic diagram of a customer credit card knowledge graph in an embodiment of the present invention;

[0069] Fig.11 is a schematic diagram of an integrated learning model in one embodiment of the present invention;

[0070] Fig.12 is a structural block diagram of a credit card customer classification device in an embodiment of the present invention;

[0071] Fig.13 A schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0073] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a system, device, apparatus, method or computer program product. Therefore, the present disclosure may be specifically implemented in the following forms, namely: complete hardware, complete software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0074] The acquisition, storage, use, and processing of data in the technical solution of the present invention are in compliance with the relevant provisions of national laws and regulations. The user information in the embodiments of this application is obtained through legal and compliant channels, and the acquisition, storage, use, and processing of user information are authorized and agreed by the customer.

[0075] The present invention mainly adopts machine learning and knowledge graph technology in artificial intelligence, among which knowledge graph technology will be used to screen and eliminate high-risk customers in the full customer base, and optimize the credit card customer base. Machine learning is used to build models, and the clustering model in machine learning clusters new credit card customers who have applied for but not activated their cards, customers who have activated their cards but have not consumed, customers who have activated their cards but have consumed little, and customers who have made low contributions to the bank into related groups, and analyzes the characteristic behaviors of the related groups; at the same time, a machine learning classification prediction model is used to make corresponding classification predictions for high-willing customers and low-willing customers in the group, and the two models are combined, so that subsequent marketing strategies can be formulated more comprehensively, and the success rate of model implementation can be improved.

[0076] The present invention provides a full-process mechanism for activating new customers, awakening dormant customers, and improving the quality of stable customers based on the life cycle of a credit card. By building a potential customer mining model, high-quality and potential customers can be identified among a large number of customers who have not applied for a credit card; by building a dormant customer awakening model, existing customers can be activated, which helps to increase the credit card activation ratio and improve the overall quality of the credit card customer base. Finally, combined with the model results, marketing strategies for activating new credit card customers and awakening dormant customers and improving the quality of credit cards can be formulated in layers and stages.

[0077] The life cycle of a credit card customer starts with the customer applying for a card and ends with the customer canceling the account, including the entire process of application, activation, use and account cancellation. The present invention defines customers at different stages of the credit card life cycle as follows:

[0078] ① Potential customers: Credit card potential customers mainly refer to customers who have not applied for local bank credit cards, but hold credit cards from other banks or do not hold credit cards, and have certain consumption needs. This type of customers mainly takes credit card recommendation as the main marketing target.

[0079] ② New customers: This type of customers are mainly those who have applied for a credit card but have not yet activated it.

[0080] ③ Dormant customers: These customers are those who have activated their credit cards but have not swiped them, or those who have swiped their credit cards but rarely. The main marketing target for these customers is credit card activation and consumption.

[0081] ④Active customers: Active customers refer to customers who use credit cards normally and have no overdue or illegal use of credit cards. The main marketing purpose for this type of customers is to increase assets and product coverage.

[0082] ⑤ Lost customers: One type is customers who have closed their accounts. For such customers, we need to investigate the reasons for closing their accounts and retain them in a timely manner. The other type refers to customers who have abnormal credit card repayments and may deteriorate. For such customers, we need to predict the probability of deterioration, remind them in advance, and take measures such as reducing or freezing their credit cards to prevent asset quality from declining.

[0083] In the entire credit card life cycle, different stages contain many important events. Customers at different stages have different consumption behaviors, which also affect the revenue of the credit card business in different ways. We establish different data models for different stages in the cycle to solve the key problems in each stage. The present invention is described in detail below with reference to the accompanying drawings.

[0084] Figure 1 It is a process analysis diagram of the entire life cycle of a credit card in an embodiment of the present invention. Figure 2It is a flow chart of the credit card customer classification method in an embodiment of the present invention. Figure 3 FIG. 1 is a flow chart of a method for classifying credit card customers in another embodiment of the present invention. Figure 1-Figure 3 As shown, the credit card customer classification methods include:

[0085] S101: Acquire target credit card customer data, and perform customer screening on the target credit card customer data.

[0086] In one embodiment, before executing S101, the process further includes: extracting the target customer's customer code, and supplementing relevant information according to the customer's customer code.

[0087] Table 1

[0088]

[0089]

[0090]

[0091] Table 1 is a schematic table of credit card customer data. As shown in Table 1, the present invention extracts the full amount of target customer groups and supplements relevant information. The information is extracted from the relevant information table of the bank's data lake and data warehouse.

[0092] Figure 4 is a flow chart of S101 in an embodiment of the present invention. Figure 4 As shown, S101 includes:

[0093] S201: Determine blacklist customer data based on the customer credit card knowledge graph and the target credit card customer data.

[0094] Knowledge graph is a major underlying technology of artificial intelligence. It is a semantic network that depicts the relationship between entities. It has its own semantics, logical meanings and rules. It describes the relationship between things in the form of "triplets", that is, a set of "entity × relationship × attribute". Knowledge graph structures and visualizes knowledge information in the nonlinear world, assisting humans in reasoning, prediction, classification, etc. A knowledge graph often contains multiple types of entities and relationships. By constructing a knowledge graph, we can mine the multi-degree relationship between entities.

[0095] The knowledge graph is used to evaluate the credit card customer's personal asset status, assets in the bank, loan repayment status, social relationships and other information, and high-risk customers are screened out and eliminated. The remaining low-risk customers proceed to the next step.

[0096] Figure 5 is a flow chart of constructing a customer credit card knowledge graph in an embodiment of the present invention. Figure 5 As shown, the credit card customer classification method also includes:

[0097] S301: Determine credit card entity data and credit card relationship data based on original customer credit card data.

[0098] Table 2

[0099]

[0100] Table 2 is a table of credit card entities and credit card relationship data. As shown in Table 2, the credit card entity data includes credit cards (card numbers), credit card customers, and blacklist customers. The credit card relationship data includes fund transactions, blacklist customer associated customers, and blacklist customer holding credit cards.

[0101] S302: Construct the customer credit card knowledge graph based on the credit card entity data and the credit card relationship data.

[0102] Fig.10 Schematic diagram of customer credit card knowledge graph in an embodiment of the present invention. Fig.10 As shown, the above-determined entities and relationships are used to construct a knowledge graph, find all customers associated with blacklisted customers and customers who have transactions with credit cards held by blacklisted customers, and integrate these customers to form a blacklisted customer data list. The customers in the box are blacklisted customer data.

[0103] S202: Screening the target credit card customer data according to the blacklist customer data.

[0104] S102: Input the screened target credit card customer data into a credit card customer classification model to obtain a credit card customer classification result.

[0105] The credit card customer classification model is trained based on a historical credit card training data set, a plurality of historical credit card sub-training data sets and a corresponding credit card classification model.

[0106] The credit card classification model of the present invention is a fusion model, which may include a logistic regression model, a decision tree model, a support vector machine model, a random forest model, an XGBoost model, and a LightGBM model.

[0107] In the prior art, the data set used for the second-stage model training is composed of the output values ​​of the model training sets in the first stage. Generally, there are as many input features as there are primary learners. Although this method can effectively avoid model overfitting and improve model effects, the use of a new combined training set may result in the loss of some information from the original training set.

[0108] Fig.11 FIG. 1 is a schematic diagram of an integrated learning model in one embodiment of the present invention. Fig.11As shown, the present invention divides the historical credit card training data set into a training set and a test set, selects 80% of the data set as the training set, and selects 20% of the data set as the test set. Different training data is used, that is, different models use different training sets, forming training set 1, training set 2, training set 3, and so on to training set n.

[0109] In order to maintain the characteristics of the original data set information, the present invention further performs feature screening on the original data set (historical credit card training data set) and sets a feature importance threshold, and selects features that exceed the feature importance threshold to form a new credit card data set, which is then used as part of the secondary learner training, that is, the secondary learner's training set is obtained by combining the original credit card training set and the new data set formed by the stacking model in the first stage.

[0110] Figure 6 : is a flow chart of creating a credit card customer classification model in an embodiment of the present invention. Figure 6 As shown, the steps to create a credit card customer classification model include:

[0111] S401: Divide the historical credit card training data set into the multiple historical credit card sub-training data sets.

[0112] Figure 7 4 is a flow chart of S401 in an embodiment of the present invention. Figure 7 As shown, S401 includes:

[0113] S501: Randomly assign data positions in the historical credit card training data set.

[0114] S502: Divide the randomly allocated data into the multiple historical credit card sub-training data sets.

[0115] In the specific implementation, the historical credit card training data set is divided into two groups of data, training set and test set, at a ratio of 80% and 20%. There are five historical credit card sub-training data sets, namely training set 1, training set 2, training set 3, training set 4 and training set 5.

[0116] S402: Input the plurality of historical credit card sub-training data sets into a corresponding first credit card classification model to obtain a corresponding first credit card data prediction result.

[0117] In specific implementation, the present invention will select XGBoost, LightGBM and random forest as the first credit card classification model. Taking random forest as an example, the present invention sequentially selects 4 of the 5 training sets after division, and then uses the 5-fold cross-validation method to train the model. Finally, the prediction is performed on the test set to obtain the result, thereby obtaining 5 prediction results (first credit card data prediction results) trained by the random forest model on the training set and 1 prediction value B1 on the test set. The 5 prediction results are then overlapped and spliced ​​vertically into a new A1 feature. The LgihtGBM model and the random forest model also use the same method to perform model prediction to generate new prediction features.

[0118] Each cross-validation of the present invention will make predictions based on the model generated by the training data, and these prediction values ​​will eventually be spliced ​​together as the training set of the second-layer model (the second credit card classification model). At the same time, after each cross-validation, the original test set of the data set will be predicted, and finally the arithmetic average of the prediction values ​​of each part will be taken as the test set of the second-layer model. After that, the matrix obtained by merging the training set prediction values ​​of the first-layer model (the first credit card classification model) in parallel is used as the training set, and the predictions of the first-layer model and the matrix obtained by merging in parallel are used as the test set, which are substituted into the second-layer model, and then further trained based on them to obtain the final prediction results.

[0119] S403: Input the features corresponding to the multiple first credit card data prediction results and the historical credit card training data set into a second credit card classification model to obtain a credit card data prediction result.

[0120] Figure 8 Flowchart of obtaining the features corresponding to the historical credit card training data set in an embodiment of the present invention. Figure 8 As shown, the credit card customer classification method also includes:

[0121] S601: Input the historical credit card training data set into a credit card feature model to obtain historical credit card features.

[0122] S602: Filter the historical credit card training data set according to the comparison result between the historical credit card feature and the feature threshold, and obtain the feature corresponding to the historical credit card training data set.

[0123] In specific implementation, the present invention uses a random forest model to perform feature importance screening on a historical credit card training data set, sets a feature importance threshold, and selects features exceeding the feature importance threshold to form a subset of the historical credit card training data set as feature A4.

[0124] Fig. 9 This is a flow chart of S403 in an embodiment of the present invention. Fig. 9As shown, S403 includes:

[0125] S701: Concatenate the plurality of first credit card data prediction results with features corresponding to the historical credit card training data set to obtain a credit card feature concatenation result.

[0126] S702: Input the credit card feature concatenation result into the second credit card classification model to obtain the credit card data prediction result.

[0127] S404: Determine a loss function according to the credit card classification prediction result and the corresponding credit card actual result, and update the first credit card classification model and the second credit card classification model according to the loss function until the loss function converges.

[0128] S405: Creating the credit card customer classification model according to the first credit card classification model and the second credit card classification model after the loss function converges.

[0129] Among them, after the three first credit card classification models are trained and the important features are selected, new features A1, A2, A3 and A4 will be formed, and then the LR model (Logistic model) will be used as the second credit card classification model for training. Using the trained LR model training test set, three new features B1, B2 and B3 will be generated on the three basic learning models of the first part of the model (the first credit card classification model) and feature B4 obtained by feature importance screening through random forests. The LR model will predict these features to obtain classification prediction results.

[0130] Since the performance of the first credit card classification models varies, in order to better obtain effective information in the first credit card classification model, the results output by each model in the first layer can be weighted averaged based on the performance of each model in the first layer, and the weight is set to the KS value (Lorenz curve) output by each model.

[0131] Table 3

[0132] True value: 1 True value: 0 Prediction value: 1 True Positive (TP) False Positive (FP) Prediction value: 0 False Negative (FN) True Negative (TN)

[0133] Table 4

[0134]

[0135]

[0136] Table 5

[0137]

[0138]

[0139] Table 3 is a definition table of true values ​​and predicted values. Table 4 is a definition table of positive samples and negative samples. Table 5 is a schematic table of evaluation indicators. As shown in Tables 3 to 5, after creating a credit card customer classification model, the credit card customer classification model can also be evaluated. When the predicted value is 1, it means that the predicted credit card customer is a high-willing customer, and when the predicted value is 0, it means that the predicted credit card customer is a low-willing customer.

[0140] In summary, the credit card customer classification method provided by the embodiment of the present invention has the following beneficial effects:

[0141] 1. Focus on the behavior of the target customers in the life cycle of a credit card, and use human behavior throughout the stages of card opening, activation, binding and consumption in the life cycle of a credit card. This will allow us to more clearly see the business pain points encountered in the key links of the credit card life cycle, and provide data-level support for subsequent modeling.

[0142] 2. Innovatively introduced the knowledge graph commonly used in the financial field, and used the knowledge graph to screen out and eliminate high-risk customers from the overall customer base, making the model constructed in this project and the subsequent marketing strategy more targeted, reducing the workload and optimizing the customer base.

[0143] 3. In the model construction part, a model construction method combining the machine learning clustering model and the machine learning classification prediction model is proposed. The machine learning clustering model can be used to cluster new credit card customers who have applied for cards, customers who have not activated cards, customers who have activated cards but have not consumed, customers who have activated cards but have consumed little, and customers with low contribution to credit cards into related groups, and analyze the specific reasons; the machine learning classification prediction model can be used to make corresponding classification predictions for high-willing customers and low-willing customers in the group; combining the two models can make it possible to formulate subsequent marketing strategies more comprehensively and improve the success rate of model implementation.

[0144] 4. In the model training stage, an improved Stacking model is used, that is, the original data set is randomly sampled with replacement to obtain different data sets. Using differentiated training data can further increase the difference between model output values ​​(low correlation), thereby enhancing the stability of model prediction. At the same time, a subset of the original data set obtained by feature importance screening is used to retain the important part of the original data set and prevent the model from overfitting.

[0145] Based on the same inventive concept, an embodiment of the present invention also provides a credit card customer classification device. Since the principle of solving the problem by the device is similar to that of the credit card customer classification method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0146] Fig.12 : is a structural block diagram of a credit card customer classification device in an embodiment of the present invention. Fig.12 As shown, the credit card customer classification device includes:

[0147] A customer screening module, used to obtain target credit card customer data and perform customer screening on the target credit card customer data;

[0148] The credit card customer classification module is used to input the screened target credit card customer data into the credit card customer classification model to obtain the credit card customer classification result; wherein, the credit card customer classification model is trained based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification models.

[0149] In one embodiment, the credit card customer classification device further comprises:

[0150] A training data set division module, used for dividing the historical credit card training data set into the plurality of historical credit card sub-training data sets;

[0151] A first credit card data prediction module, used for inputting the plurality of historical credit card sub-training data sets into a corresponding first credit card classification model to obtain a corresponding first credit card data prediction result;

[0152] A second credit card data prediction module, used for inputting the plurality of first credit card data prediction results and features corresponding to the historical credit card training data set into a second credit card classification model to obtain a credit card data prediction result;

[0153] A loss function module, used to determine a loss function according to the credit card classification prediction result and the corresponding credit card actual result, and update the first credit card classification model and the second credit card classification model according to the loss function until the loss function converges;

[0154] A credit card customer classification model creation module is used to create the credit card customer classification model based on the first credit card classification model and the second credit card classification model after the loss function converges.

[0155] In one embodiment, it also includes:

[0156] A historical credit card feature module, used to input the historical credit card training data set into a credit card feature model to obtain historical credit card features;

[0157] The screening module is used to screen the historical credit card training data set according to the comparison result between the historical credit card feature and the feature threshold, so as to obtain the feature corresponding to the historical credit card training data set.

[0158] In one embodiment, the second credit card data prediction module includes:

[0159] A splicing unit, used for splicing the plurality of first credit card data prediction results and features corresponding to the historical credit card training data set to obtain a credit card feature splicing result;

[0160] The credit card data prediction result unit is used to input the credit card feature splicing result into the second credit card classification model to obtain the credit card data prediction result.

[0161] In one embodiment, the training data set partitioning module includes:

[0162] A random unit, used for randomly assigning data positions in the historical credit card training data set;

[0163] The partitioning unit is used to partition the randomly allocated data into the plurality of historical credit card sub-training data sets.

[0164] In one embodiment, the customer screening module includes:

[0165] A blacklist customer unit, used to determine blacklist customer data based on the customer credit card knowledge graph and the target credit card customer data;

[0166] A customer screening unit is used to screen the target credit card customer data according to the blacklist customer data.

[0167] In one embodiment, it also includes:

[0168] An entity relationship determination module, used to determine credit card entity data and credit card relationship data based on original customer credit card data;

[0169] A customer credit card knowledge graph construction module is used to construct the customer credit card knowledge graph based on the credit card entity data and the credit card relationship data.

[0170] In summary, the credit card customer classification device of the embodiment of the present invention performs customer screening on target credit card customer data, obtains a credit card customer classification model based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification model training, and inputs the screened target credit card customer data into the credit card customer classification model to obtain a credit card customer classification result, which can optimize the credit card customer group and improve the quality of credit card customers.

[0171] Fig.13 FIG. 9 is a schematic block diagram of the system structure of the electronic device 9600 according to an embodiment of the present application. Fig.13As shown, the electronic device 9600 may include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that Fig.13 is exemplary; other types of structures may also be used to supplement or replace this structure to implement telecommunication functions or other functions.

[0172] In one embodiment, the credit card customer classification method function may be integrated into the central processor 9100. The central processor 9100 may be configured to perform the following control:

[0173] Acquire target credit card customer data, and perform customer screening on the target credit card customer data;

[0174] The screened target credit card customer data is input into a credit card customer classification model to obtain a credit card customer classification result; wherein the credit card customer classification model is trained based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification models.

[0175] From the above description, it can be seen that the credit card customer classification method provided in the present application performs customer screening on target credit card customer data, obtains a credit card customer classification model based on a historical credit card training data set, multiple historical credit card sub-training data sets and a corresponding credit card classification model training, and inputs the screened target credit card customer data into the credit card customer classification model to obtain a credit card customer classification result, which can optimize the credit card customer group and improve the quality of credit card customers.

[0176] In another embodiment, the credit card customer classification device can be configured separately from the central processor 9100. For example, the credit card customer classification device can be configured as a chip connected to the central processor 9100, and the functions of the credit card customer classification method can be implemented through the control of the central processor.

[0177] like Fig.13 As shown, the electronic device 9600 may also include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily have to include Fig.13 In addition, the electronic device 9600 may also include Fig.13 For components not shown, reference may be made to the prior art.

[0178] like Fig.13 As shown, the central processing unit 9100 is sometimes also referred to as a controller or an operation control, and may include a microprocessor or other processor device and / or logic device. The central processing unit 9100 receives input and controls the operation of various components of the electronic device 9600.

[0179] The memory 9140 may be, for example, one or more of a cache, a flash memory, a hard drive, a removable medium, a volatile memory, a non-volatile memory or other suitable devices. The above-mentioned information related to the failure may be stored, and a program for executing the relevant information may also be stored. The CPU 9100 may execute the program stored in the memory 9140 to implement information storage or processing, etc.

[0180] The input unit 9120 provides input to the central processing unit 9100. The input unit 9120 is, for example, a key or a touch input device. The power supply 9170 is used to provide power to the electronic device 9600. The display 9160 is used to display display objects such as images and texts. The display may be, for example, an LCD display, but is not limited thereto.

[0181] The memory 9140 may be a solid-state memory, such as a read-only memory (ROM), a random access memory (RAM), a SIM card, etc. It may also be a memory that saves information even when the power is off, can be selectively erased, and is provided with more data, examples of which are sometimes referred to as EPROMs, etc. The memory 9140 may also be some other type of device. The memory 9140 includes a buffer 9141 (sometimes referred to as a buffer memory). The memory 9140 may include an application / function storage unit 9142, which is used to store application programs and function programs or processes for executing the operation of the electronic device 9600 through the central processor 9100.

[0182] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for communication functions of the electronic device and / or for executing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0183] The communication module 9110 is a transmitter / receiver 9110 that sends and receives signals via an antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processor 9100 to provide input signals and receive output signals, which may be the same as the case of a conventional mobile communication terminal.

[0184] Based on different communication technologies, multiple communication modules 9110 may be provided in the same electronic device, such as a cellular network module, a Bluetooth module and / or a wireless LAN module, etc. The communication module (transmitter / receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide an audio output via the speaker 9131 and receive an audio input from the microphone 9132, thereby realizing a common telecommunication function. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. In addition, the audio processor 9130 is also coupled to the central processor 9100, so that recording can be performed on the local machine through the microphone 9132, and the sound stored on the local machine can be played through the speaker 9131.

[0185] The embodiment of the present invention also provides a computer-readable storage medium capable of implementing all the steps of the credit card customer classification method in the above embodiment, where the execution subject is a server or a client. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, all the steps of the credit card customer classification method in the above embodiment are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0186] Acquire target credit card customer data, and perform customer screening on the target credit card customer data;

[0187] The screened target credit card customer data is input into a credit card customer classification model to obtain a credit card customer classification result; wherein the credit card customer classification model is trained based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification models.

[0188] In summary, the computer-readable storage medium of the embodiment of the present invention performs customer screening on target credit card customer data, obtains a credit card customer classification model based on a historical credit card training data set, multiple historical credit card sub-training data sets and a corresponding credit card classification model training, and inputs the screened target credit card customer data into the credit card customer classification model to obtain a credit card customer classification result, which can optimize the credit card customer group and improve the quality of credit card customers.

[0189] The embodiment of the present invention also provides a computer program product capable of implementing all the steps of the credit card customer classification method in the above embodiment, where the execution subject is a server or a client. The computer program product includes a computer program / instruction. When the computer program / instruction is executed by a processor, all the steps of the credit card customer classification method in the above embodiment are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0190] Acquire target credit card customer data, and perform customer screening on the target credit card customer data;

[0191] The screened target credit card customer data is input into a credit card customer classification model to obtain a credit card customer classification result; wherein the credit card customer classification model is trained based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification models.

[0192] In summary, the computer program product of the embodiment of the present invention performs customer screening on target credit card customer data, obtains a credit card customer classification model based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification model training, and inputs the screened target credit card customer data into the credit card customer classification model to obtain a credit card customer classification result, which can optimize the credit card customer group and improve the quality of credit card customers.

[0193] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the hardware + program embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0194] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0195] Although the present application provides method operation steps as described in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-creative labor. The order of steps listed in the embodiments is only one way of executing the order of many steps and does not represent the only execution order. When the actual device or client product is executed, it can be executed in the order of the method shown in the embodiments or the drawings or in parallel (for example, in a parallel processor or multi-threaded processing environment).

[0196] Although the present specification embodiment provides the method operation steps as described in the embodiment or flow chart, more or less operation steps may be included based on conventional or non-creative means. The order of steps listed in the embodiment is only one way in the order of execution of many steps, and does not represent a unique execution order. When the device or terminal product in practice is executed, it can be executed in sequence or in parallel (such as a parallel processor or a multi-threaded processing environment, or even a distributed data processing environment) according to the method shown in the embodiment or the accompanying drawings. The term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, product or equipment including a series of elements not only includes those elements, but also includes other elements not clearly listed, or also includes elements inherent to such process, method, product or equipment. In the absence of more restrictions, it is not excluded that there are other identical or equivalent elements in the process, method, product or equipment including the elements.

[0197] For the convenience of description, the above devices are described in various modules according to their functions. Of course, when implementing the embodiments of this specification, the functions of each module can be implemented in the same or more software and / or hardware, or the module implementing the same function can be implemented by a combination of multiple sub-modules or sub-units. 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. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0198] Those skilled in the art also know that, in addition to implementing the controller in a purely computer-readable program code, the controller can be made to implement the same function in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, such a controller can be considered as a hardware component, and the devices for implementing various functions included therein can also be considered as structures within the hardware component. Or even, the devices for implementing various functions can be considered as both software modules for implementing the method and structures within the hardware component.

[0199] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0200] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0201] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0202] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0203] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0204] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0205] Those skilled in the art will appreciate that the embodiments of this specification may be provided as methods, systems or computer program products. Therefore, the embodiments of this specification may take the form of complete hardware embodiments, complete software embodiments or embodiments combining software and hardware. Moreover, the embodiments of this specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0206] The various embodiments in this specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The embodiments of this specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0207] Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. In the description of this specification, the description of the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of this specification. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples without contradiction.

[0208] The above is only an example of the embodiment of the present specification and is not intended to limit the embodiment of the present specification. For those skilled in the art, the embodiment of the present specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiment of the present specification shall be included in the scope of the claims of the embodiment of the present specification.

Claims

1. A credit card customer classification method, characterized in that: include: Acquire target credit card customer data, and perform customer screening on the target credit card customer data; The screened target credit card customer data is input into a credit card customer classification model to obtain a credit card customer classification result; wherein the credit card customer classification model is trained based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification models.

2. The credit card customer classification method according to claim 1, characterized in that: The steps to create a credit card customer classification model include: Dividing the historical credit card training data set into the plurality of historical credit card sub-training data sets; Inputting the plurality of historical credit card sub-training data sets into the corresponding first credit card classification model to obtain corresponding first credit card data prediction results; Inputting the features corresponding to the plurality of first credit card data prediction results and the historical credit card training data set into a second credit card classification model to obtain a credit card data prediction result; Determine a loss function according to the credit card classification prediction result and the corresponding credit card actual result, and update the first credit card classification model and the second credit card classification model according to the loss function until the loss function converges; The credit card customer classification model is created according to the first credit card classification model and the second credit card classification model after the loss function converges.

3. The credit card customer classification method according to claim 2, characterized in that: Also includes: Inputting the historical credit card training data set into a credit card feature model to obtain historical credit card features; The historical credit card training data set is screened according to the comparison result between the historical credit card feature and the feature threshold to obtain the feature corresponding to the historical credit card training data set.

4. The credit card customer classification method according to claim 2, characterized in that: Inputting the features corresponding to the plurality of first credit card data prediction results and the historical credit card training data set into a second credit card classification model to obtain the credit card data prediction results includes: splicing the plurality of first credit card data prediction results and features corresponding to the historical credit card training data set to obtain a credit card feature splicing result; The credit card feature concatenation result is input into the second credit card classification model to obtain the credit card data prediction result.

5. The credit card customer classification method according to claim 2, characterized in that: Dividing the historical credit card training data set into the plurality of historical credit card sub-training data sets comprises: Randomly assigning data positions in the historical credit card training data set; The randomly assigned data is divided into the multiple historical credit card sub-training data sets.

6. The credit card customer classification method according to claim 1, characterized in that: Performing customer screening on the target credit card customer data includes: Determine blacklist customer data based on the customer credit card knowledge graph and the target credit card customer data; The target credit card customer data is screened according to the blacklist customer data.

7. The credit card customer classification method according to claim 6, characterized in that: Also includes: Determine credit card entity data and credit card relationship data based on original customer credit card data; The customer credit card knowledge graph is constructed based on the credit card entity data and the credit card relationship data.

8. A credit card customer classification device, characterized in that: include: A customer screening module, used to obtain target credit card customer data and perform customer screening on the target credit card customer data; The credit card customer classification module is used to input the screened target credit card customer data into the credit card customer classification model to obtain the credit card customer classification result; wherein, the credit card customer classification model is trained based on a historical credit card training data set, multiple historical credit card sub-training data sets and corresponding credit card classification models.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the credit card customer classification method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the credit card customer classification method according to any one of claims 1 to 7 are implemented.

Images