Data processing method, device, equipment, storage medium and program product

Abstract

The present application provides a data processing method, device, equipment, storage medium and program product, which relates to the field of data analysis, wherein the method includes: classifying the original feature set of the customer group to be classified to obtain several group features under each category; detecting the correlation between several group features under each category, and determining the number of dimensionality reduction factors and corresponding group feature factors of each category based on the correlation detection results; obtaining the correlation between several group features under each category and each group feature factor, and attributing the corresponding group features in each category to the group feature factor with the greatest correlation, and performing cluster analysis on the customer group to be classified based on the factor analysis results, and obtaining various subgroups of the customer group to be classified based on the cluster analysis results. Through the above method, the present application can effectively avoid the dimensional disaster caused by too many features of the customer group, and improve the efficiency of customer group classification.

Description

Technical Field

[0001] The present application relates to the field of data analysis, and in particular to a data processing method, device, equipment, storage medium and program product. Background Art

[0002] With the continuous increase in banking business, banks are paying more and more attention to the efficiency of business delivery and the business experience of customer groups, among which how to classify customer groups has become a key means.

[0003] Especially for the retail business of banks, which caters to a huge customer base, if the customer base can be segmented, the efficiency of business delivery will be greatly improved, while improving the customer's business experience.

[0004] In the related technology, a classification scheme for clustering analysis of customer groups is proposed. Customers with the same characteristics are grouped into one category through clustering to achieve customer group segmentation. However, the problem of characteristic dimension is not considered. Especially for the huge customer group of retail business, the customer group has many characteristics, which will cause dimensional disaster in the clustering process and ultimately make it impossible to group customers. Summary of the invention

[0005] In order to solve the above problem, that is, to solve the problem that customer groups cannot be grouped due to too many clustering features, the present application provides a data processing method, device, equipment, storage medium and program product.

[0006] In order to achieve the above objectives, this application provides the following technical solutions:

[0007] According to one aspect of the present application, a data processing method is provided, comprising:

[0008] Classify the original feature set of the customer group to be classified, and obtain several group features under each category;

[0009] Detect the correlation between several group characteristics under each category, and determine the number of dimensionality reduction factors and corresponding group characteristic factors of each category based on the correlation detection results;

[0010] Obtain the correlation between several group characteristics under each category and each group characteristic factor, and attribute the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation, and obtain the factor analysis results;

[0011] Based on the factor analysis result, a cluster analysis is performed on the customer group to be classified, and based on the cluster analysis result, various subgroups of the customer group to be classified are obtained.

[0012] In one embodiment, before classifying the original feature set of the customer group to be classified, the method further includes:

[0013] Determining a feature category based on business information corresponding to the customer group to be classified;

[0014] The classifying of the original feature set of the customer group to be classified includes: classifying each group feature in the original data set into a corresponding feature category based on basic information of group features, wherein the basic information of group features includes a feature name or feature description information.

[0015] In one embodiment, detecting the correlation between a plurality of population characteristics under each category includes:

[0016] The Bartlett sphericity test was performed on several group characteristics under each category;

[0017] Based on the results of Bartlett's sphericity test, group characteristics and their correlations are obtained from several group characteristics under each category for group characteristic factor analysis.

[0018] In one embodiment, the detecting the correlation between the plurality of population characteristics under each category further includes:

[0019] Conduct statistical KMO tests on several group characteristics under each category;

[0020] The method of obtaining group characteristics and their correlations for performing group characteristic factor analysis from among several group characteristics under each category based on the Bartlett sphericity test results includes: obtaining group characteristics and their correlations for performing group characteristic factor analysis from among several group characteristics under each category based on the statistical quantity test results and the Bartlett sphericity test results.

[0021] In one embodiment, the method of determining the number of dimensionality reduction factors and corresponding population characteristic factors of each category based on the correlation detection result includes:

[0022] Obtaining a correlation matrix of several group characteristics of each category based on the correlation detection result, and obtaining a group characteristic value of the correlation matrix of each category;

[0023] The number of dimensionality reduction factors and the corresponding group characteristic factors of each category are determined based on the group characteristic values ​​of the correlation matrix of each category and the total number of corresponding group characteristics.

[0024] In one embodiment, after obtaining the correlation between a plurality of group characteristics in each category and each group characteristic factor, and before attributing the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation, the method further includes:

[0025] The group characteristic factor loading matrix of each category is established based on the correlation between several group characteristics under each category and each group characteristic factor;

[0026] The method of attributing the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation includes:

[0027] Based on the group characteristic factor loading matrix, the corresponding group characteristics in each category are respectively attributed to the group characteristic factor with the greatest correlation.

[0028] In one embodiment, after obtaining various subgroups of the customer group to be classified based on the cluster analysis result, the method further includes:

[0029] Based on the distribution of group characteristic factors of each sub-group, group analysis is conducted on the business characteristics of each sub-group.

[0030] According to another aspect of the present application, there is provided a data processing device, comprising:

[0031] A feature classification module, which is configured to classify the original feature set of the customer group to be classified, and obtain a number of group features under each category;

[0032] A correlation detection module configured to detect correlations between a number of group characteristics under each category;

[0033] A factor determination module, which is configured to determine the number of dimensionality reduction factors and corresponding group characteristic factors for each category based on the correlation detection results;

[0034] A factor analysis module is configured to obtain the correlation between a number of group characteristics under each category and each group characteristic factor, and to attribute the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation, thereby obtaining the factor analysis result;

[0035] The customer group classification module is configured to perform cluster analysis on the customer group to be classified based on the factor analysis result, and obtain subgroups of the customer group to be classified based on the cluster analysis result.

[0036] According to another aspect of the present application, there is provided an electronic device, comprising: a memory and a processor;

[0037] The memory stores computer-executable instructions;

[0038] The processor executes the computer-executable instructions stored in the memory, so that the electronic device executes the data processing method.

[0039] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to implement the data processing method when executed by a processor.

[0040] According to another aspect of the present application, a computer program product is provided. The computer program product includes a computer program code. When the computer program code is executed on a computer, the computer executes the data processing method.

[0041] It can be understood that the data processing method, device, equipment, storage medium and program product provided by the embodiment of the present application obtains several group characteristics under each category by classifying the original feature set of the customer group to be classified; detects the correlation between several group characteristics under each category, and determines the number of dimensionality reduction factors and corresponding group characteristic factors of each category based on the correlation detection results; obtains the correlation between several group characteristics under each category and each group characteristic factor, and attributes the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation, and performs cluster analysis on the customer group to be classified based on the factor analysis results, and obtains various subgroups of the customer group to be classified based on the cluster analysis results. Through the above method, the present application can effectively avoid the dimensional disaster caused by too many characteristics of the customer group, and improve the efficiency of customer group classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0043] Figure 1 A possible scenario diagram provided for an embodiment of the present application;

[0044] Figure 2 A flowchart of a data processing method provided in an embodiment of the present application;

[0045] Figure 3 An example diagram of the number of dimensionality reduction factors for a certain category determined in this application;

[0046] Figure 4 This is an example diagram of clustering using K-means in this application;

[0047] Figure 5 A schematic diagram of the structure of a data processing device provided in an embodiment of the present application;

[0048] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0049] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0050] To facilitate understanding of the embodiments of the present application, the following terms are first explained before introducing the embodiments of the present application:

[0051] Customer segmentation: Split the target customer group into multiple sub-groups based on different criteria.

[0052] Clustering: The process of dividing a collection of physical or abstract objects into multiple classes consisting of similar objects is called clustering. The clusters generated by clustering are a set of data objects that are similar to the objects in the same cluster and different from the objects in other clusters.

[0053] K-means: K-means algorithm is a typical distance-based clustering algorithm. It uses distance as the evaluation index of similarity and uses Euclidean distance as the default similarity measure. That is, the closer the distance between two objects, the greater their similarity. The principle is to initialize k cluster centers first, summarize the samples belonging to each cluster based on the distance between the sample and the center point, and iterate to achieve the goal of minimizing the distance between the sample and the cluster center to which it belongs.

[0054] Bartlett's sphericity test: used to test the correlation of various variables. This test method is based on the correlation coefficient matrix. Its null hypothesis is that the correlation coefficient matrix is ​​a unit matrix, that is, all elements on the diagonal of the correlation coefficient matrix are 1, and all elements on the non-diagonal are zero. The statistics of the Bartlett's sphericity test method are obtained based on the determinant of the correlation coefficient matrix. If the value is large and the corresponding accompanying probability value is less than the specified significance level, the null hypothesis is rejected, indicating that the correlation coefficient matrix is ​​not a unit matrix, there is a correlation between the original variables, and it is suitable for principal component analysis; otherwise, the null hypothesis is established, there is no correlation between the original variables, and the data is not suitable for principal component analysis.

[0055] Correlation matrix: Correlation matrix is ​​also called correlation coefficient matrix, which is composed of correlation coefficients between columns of the matrix. The correlation coefficient is a measure of the degree of linear relationship between research variables.

[0056] KMO test: KMO (Kaiser-Meyer-Olkin) test statistics are indicators used to compare the simple correlation coefficient and partial correlation coefficient between variables. This test is to test the relative size of the simple correlation coefficient and partial correlation coefficient between the original variables. Its calculation formula is:

[0057]

[0058] where r ij is the simple correlation coefficient, It is the partial correlation coefficient. When the correlation coefficient between variables is very small, the KMO value is close to 1, indicating that the variable is suitable for factor analysis. Generally, factor analysis is performed when the KMO value is greater than 0.6.

[0059] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions in the embodiments of the present application will be described in more detail below in conjunction with the drawings in the embodiments of the present application. In the drawings, the same or similar reference numerals throughout represent the same or similar parts or parts with the same or similar functions. The described embodiments are part of the embodiments of the present application, not all of the embodiments. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limitations on the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0060] Bank retail business often requires refined management of customer groups, such as business promotion and customer maintenance. The premise of refined management is to classify customer groups, that is, to refine and segment customer groups, and carry out targeted refined management based on the typical characteristics of segmented customer groups, such as business promotion. In recent years, customer segmentation has received more and more attention and has been widely promoted to various retail business scenarios. However, the target scenarios faced by bank retail business are complex and diverse, and the target customer characteristics are multi-dimensional and have long-tail distribution. As a result, bank business personnel lack a unified idea when segmenting a certain customer group, making it difficult to start. Currently, business personnel often select a small number of features based on experience for manual segmentation, lacking unified segmentation standards and processes.

[0061] In the related art, clustering methods such as K-means algorithm are directly used to classify customer groups. The time complexity of the Kmeans clustering model is O(m*n*k*l), and the space complexity is O((m+k)*l), where m is the number of points (i.e., the number of customers), n is the number of features used for clustering, k is the number of clusters, and l is the number of model iterations. It can be seen that when the number of clustering features is too large, it will cause dimensionality disaster. The number of customer features in retail scenarios is huge, which makes it impossible for the model to perform clustering, and the original features need to be reduced in dimension. At the same time, the clustering results lack business interpretability, and it is difficult to form customer group analysis for specific scenarios. Therefore, it is difficult for business tasks to directly apply the clustering results to the refined management of banks. In addition, there is a correlation between features, and it is meaningless to cluster multiple features with strong correlation. There is no significant difference in features between the clustered customer groups, that is, it is impossible to obtain the typical features of the segmented customer groups, and it is even more impossible to correspond. Therefore, it is necessary to integrate the highly correlated features and fuse them into a single feature.

[0062] In view of this, the embodiments of the present application provide a data processing method, apparatus, device, storage medium and program product, which can specifically solve the customer grouping problem caused by the characteristics of the customer group of the retail banking business, such as multiple characteristics and long-tail distribution. The original characteristics of the customer group to be classified are first classified, and then factor analysis is performed on each type of feature to achieve feature dimensionality reduction. The features with strong correlation are integrated into a single factor, and then clustering is performed based on the factor analysis results. The target customer group is divided into multiple sub-groups with common characteristics, thereby effectively improving the quality of customer group classification.

[0063] Figure 1 This is a possible scenario diagram provided for the embodiment of the present application. It should be noted that the data processing method disclosed in the present application can be used in the field of financial technology. It can also be used in any field other than the field of financial technology, and the application field of the data processing method disclosed in the present application is not limited.

[0064] Take the retail business push scenario of a bank as an example. Figure 1As shown, the terminal device 110 and the server 120 are connected to each other via a wired or wireless network. Optionally, the terminal 110 is used to determine the target customer group, that is, the customer group to be classified, and collect relevant data information of the customer group to be classified, such as basic user information, service subscription information, etc., and transmit the relevant data information of the target customer group to the server 120; the server 120 is used to classify the target customer group based on the data provided by the terminal 110. Optionally, in the process of customer group classification, the server 120 undertakes the main computing work and the terminal device 110 undertakes the secondary computing work; or, the server 120 undertakes the secondary computing work and the terminal device 110 undertakes the main computing work; or, the server 120 or the terminal device 110 can undertake the computing work separately.

[0065] Among them, the terminal device 110 may include but is not limited to computers, smart phones, tablet computers, e-book readers, Moving Picture experts group audio layer III (MP3) players, Moving Picture experts group audiolayer IV (MP4) players, portable computers, car computers, wearable devices, desktop computers, set-top boxes, smart TVs, etc.

[0066] Server 120 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), as well as big data and artificial intelligence platforms.

[0067] The above is a brief description of the scenario diagram of this application. Figure 1 The data processing method provided in the embodiment of the present application is described in detail by taking the server 120 in the example.

[0068] Please refer to Figure 2 , Figure 2 A flowchart of a data processing method provided in an embodiment of the present application includes steps S201-S204.

[0069] Step S201: classify the original feature set of the customer group to be classified to obtain a number of group features under each category.

[0070] It can be understood that the original feature set can include all features of the customer group to be classified, including similar features, etc., wherein the original feature set can be acquired or generated using existing technologies. For example, the original feature set is generated based on the active data of the customer group to be classified, wherein the active data can be active data generated by transactions or other operations between the customer group and an operating institution such as a bank, etc.

[0071] It should be noted that in the technical solution of this embodiment, the collection, storage, use, processing, transmission, provision and disclosure of active data involved, such as financial data or user data, etc., all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0072] In one implementation, before classifying the original feature set of the customer group to be classified in step S201, the following steps may also be included:

[0073] Determining a feature category based on business information corresponding to the customer group to be classified;

[0074] The classifying of the original feature set of the customer group to be classified includes: classifying each group feature in the original data set into a corresponding feature category based on basic information of group features, wherein the basic information of group features includes a feature name or feature description information.

[0075] Taking retail banking as an example, feature categories can be divided into three categories: asset holding, capital flow, and transaction behavior. In one implementation, according to the characteristics of retail banking, the original features are divided into two levels. The first level is the feature type to which the feature belongs, and the second level is the specific feature type corresponding to the type, namely, asset holding, capital flow, and transaction behavior. Then, each feature in the original feature set is classified into one of the three categories of asset holding, capital flow, and transaction behavior. For details, please refer to Table 1 below, where t1+t2+t3=n, and t1, t2, and t3 represent the total number of features in the three types, respectively, and n represents the total number of features in the original feature set.

[0076] Table 1:

[0077]

[0078] It can be understood that, as shown in Table 1, not only are multiple feature types divided, but the multiple feature types have their own type names, retaining a certain degree of business interpretability, which is convenient for improving the efficiency of subsequent interpretation of customer groups.

[0079] Step S202: Detect the correlation between several group characteristics under each category, and determine the number of dimensionality reduction factors and corresponding group characteristic factors of each category based on the correlation detection result.

[0080] In this embodiment, the correlation between group characteristics is detected, and the correlation between group characteristics is obtained by using the Bartlett test. The detection of the correlation between a number of group characteristics under each category includes the following steps:

[0081] The Bartlett sphericity test was performed on several group characteristics under each category;

[0082] Based on the results of Bartlett's sphericity test, group characteristics and their correlations are obtained from several group characteristics under each category for group characteristic factor analysis.

[0083] Combined with Table 1, a certain feature category Y i All features below Perform Bartlett's sphericity test to determine The correlation between characteristics, among which the group characteristics used for group characteristic factor analysis are to exclude certain variables that may provide some information independently and are not suitable for factor analysis. These variables can be detected using the Bartlett sphericity test to further implement factor analysis of the characteristics under the qualified characteristic categories.

[0084] In one implementation, for group characteristics that may provide information independently, they can be used as feature factors alone or discarded. In practical applications, adaptive adjustments can be made based on the importance of the group characteristics.

[0085] Furthermore, in order to improve the efficiency of the correlation detection result, a KMO test method is also combined. The detection of the correlation between several group characteristics under each category also includes:

[0086] Conduct statistical KMO tests on several group characteristics under each category;

[0087] The method of obtaining group characteristics and their correlations for performing group characteristic factor analysis from among several group characteristics under each category based on the Bartlett sphericity test results includes: obtaining group characteristics and their correlations for performing group characteristic factor analysis from among several group characteristics under each category based on the statistical quantity test results and the Bartlett sphericity test results.

[0088] In one implementation, a KMO test is performed by calculating the KMO value between the features. If the KMO value is greater than 0.6, its correlation value is retained and used to calculate the eigenvalue ev and eigenvector v of the correlation matrix M.

[0089] Furthermore, the method of determining the number of dimensionality reduction factors and corresponding group characteristic factors of each category based on the correlation detection result includes the following steps:

[0090] Obtaining a correlation matrix of several group characteristics of each category based on the correlation detection result, and obtaining a group characteristic value of the correlation matrix of each category;

[0091] The number of dimensionality reduction factors and the corresponding group characteristic factors of each category are determined based on the group characteristic values ​​of the correlation matrix of each category and the total number of corresponding group characteristics.

[0092] In this embodiment, the optimal number of dimensionality reduction factors F for each type of feature is determined by calculating the group characteristic value ev and the number of factors (total number of group characteristics) of the correlation matrix and drawing a graph. i and the corresponding population characteristic factors. Figure 3 As shown in the figure, take the eight group characteristics of a customer group in the asset holding category as an example, among which Figure 3 To determine the dimensionality reduction factor F i The graph, where the ordinate is the eigenvalue ev, the abscissa is the number of common factors, select the abscissa corresponding to the point closest to 1 on the ordinate, the abscissa corresponding to the point closest to 1 on the ordinate is 3, which is the optimal dimensionality reduction factor number F i = 3. The characteristic factors determined are shown in Table 2 below.

[0093] Table 2:

[0094]

[0095] It can be understood that the horizontal coordinate corresponding to the point closest to 1 on the vertical coordinate is selected, where the closer the vertical coordinate is to 1, the corresponding factor number can express 90% of the original characteristics, that is, the original information loss after attribution is less than 10%.

[0096] Step S203: Obtain the correlation between several group characteristics in each category and each group characteristic factor, and attribute the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation, to obtain factor analysis results.

[0097] In this embodiment, by attributing the group characteristics under each category respectively, the dimensionality reduction of the group characteristics is achieved, and similar characteristics are classified into one category, which can effectively avoid the dimensional disaster of too many features and solve the problem that similar features cannot derive the characteristics of the segmented customer groups.

[0098] In one embodiment, after obtaining the correlation between a plurality of group characteristics under each category and each group characteristic factor, and before attributing the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation, the following steps may be further included:

[0099] Establish a group characteristic factor loading matrix for each category based on the correlations between several group characteristics under each category and each group characteristic factor respectively;

[0100] Specifically, attribute the corresponding group characteristics in each category to the group characteristic factor with the highest correlation, which is:

[0101] Based on the group characteristic factor loading matrix, attribute the corresponding group characteristics in each category to the group characteristic factor with the highest correlation.

[0102] In this embodiment, by establishing the factor loading matrix M, the attribution efficiency of group characteristics is improved. Among them, the factor loading a ij is the correlation coefficient between the i-th variable and the j-th common factor, which represents the degree that X i depends on F j The loading represents the load of the i-th variable on the j-th common factor, which reflects the relative importance of the i-th variable on the j-th common factor.

[0103] It can be understood that in factor analysis, usually only m (m < p) main factors are selected. That is, according to the correlations of the variables, the first main factor f1 is selected to make its variance contribution in the common factor variances of each variable the largest. Then, eliminate the influence of this factor, and from the remaining correlations, select the factor that is uncorrelated with it to make its variance contribution in the remaining factor variances of each variable the largest. Repeat this process until the common factor variances of each variable are decomposed completely.

[0104] Step S204: Perform cluster analysis on the customer group to be classified based on the factor analysis result, and obtain various sub-groups of the customer group to be classified based on the cluster analysis result.

[0105] Among them, the cluster analysis can be K-means clustering. Compared with directly using K-means clustering to group customer groups, in this embodiment, by reducing the dimension and attributing features, the curse of dimensionality when there are too many features can be effectively avoided. As Figure 4 shown, Figure 4 shows using K-means to cluster 3 factors, resulting in 4 clusters: cluster 0, cluster 1, cluster 2, and cluster 4, corresponding to 4 sub-groups. Among them, the abscissa is the frequency, that is, the value of a specific factor after attribution, which can be obtained according to matrix decomposition, and the ordinate is the percentage value. The solid columns in the figure represent the positive example samples, and the hollow columns represent the overall samples.

[0106] In one implementation, after obtaining various sub-groups of the customer group to be classified based on the cluster analysis result, the following steps may further be included:

[0107] Based on the distribution of group characteristic factors of each sub-group, group analysis is conducted on the business characteristics of each sub-group.

[0108] In this embodiment, group analysis is performed on each subgroup. Figure 4 As shown, the typical feature of the subgroup corresponding to cluster 2 is factor 3 (consumption record). Factor 3 corresponds to three features (online consumption amount in the past six months, whether there is a large consumption amount in the current month, and average consumption). The business meaning of these three features measures the customer's consumption capacity, that is, the typical feature of this customer group is that the customer has a high consumption capacity.

[0109] Compared with the related art, the characteristic factors in this embodiment retain a certain degree of business explanatory power, and customer groups are grouped on the basis of feature dimensionality reduction, and the business characteristics of subgroups of customer group classification are quickly analyzed using the reduced dimensionality characteristic factors.

[0110] Please refer to Figure 5 , Figure 5 A schematic diagram of the structure of a data processing device provided in an embodiment of the present application includes:

[0111] A feature classification module 51, which is configured to classify the original feature set of the customer group to be classified, and obtain a number of group features under each category;

[0112] A correlation detection module 52, which is configured to detect the correlation between a number of group characteristics under each category;

[0113] A factor determination module 53, which is configured to determine the number of dimension reduction factors and corresponding group characteristic factors of each category based on the correlation detection result;

[0114] A factor analysis module 54 is configured to obtain the correlation between a plurality of group characteristics under each category and each group characteristic factor, and to attribute the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation, to obtain a factor analysis result;

[0115] The customer group classification module 55 is configured to perform cluster analysis on the customer group to be classified based on the factor analysis result, and obtain subgroups of the customer group to be classified based on the cluster analysis result.

[0116] In one embodiment, the device further comprises:

[0117] A category determination module, which is configured to determine a feature category based on the business information corresponding to the customer group to be classified;

[0118] The feature classification module 51 is specifically configured to classify each group feature in the original data set into a corresponding feature category based on the basic information of group features, wherein the basic information of group features includes a feature name or feature description information.

[0119] In one embodiment, the correlation detection module 52 is specifically configured to perform Bartlett sphericity test on several group characteristics under each category respectively; and obtain group characteristics and their correlations for performing group characteristic factor analysis from several group characteristics under each category based on the Bartlett sphericity test results.

[0120] In one embodiment, the correlation detection module 52 is further configured to perform statistical KMO detection on several group characteristics under each category respectively; and obtain group characteristics and their correlations for performing group characteristic factor analysis among several group characteristics under each category based on the statistical test results and the Bartlett sphericity test results.

[0121] In one embodiment, the factor determination module 53 is specifically configured to obtain a correlation matrix of several group characteristics of each category based on the correlation detection results, and obtain the group characteristic value of the correlation matrix of each category; and determine the number of dimensionality reduction factors and the corresponding group characteristic factors of each category based on the group characteristic value of the correlation matrix of each category and the total number of corresponding group characteristics.

[0122] In one embodiment, the device further comprises:

[0123] A matrix building module, which is configured to build a group characteristic factor loading matrix of each category based on the correlation between a number of group characteristics under each category and each group characteristic factor;

[0124] The method of attributing the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation is specifically: attributing the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation based on the group characteristic factor loading matrix.

[0125] In one embodiment, the device further comprises:

[0126] The group analysis module is configured to perform group analysis on the business characteristics of each sub-group based on the distribution of group characteristic factors of each sub-group.

[0127] It should be noted here that the above-mentioned device provided in the present application can correspondingly implement all the method steps implemented by the server in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as the method embodiment will not be described in detail here.

[0128] Please refer to Figure 6 , Figure 6 An electronic device provided in an embodiment of the present application includes: a memory 61 and a processor 62;

[0129] The memory 61 stores computer-executable instructions;

[0130] The processor 62 executes the computer-executable instructions stored in the memory 61, so that the electronic device executes the data processing method.

[0131] It should be noted here that the above-mentioned electronic device provided in the present application can correspondingly implement all the method steps implemented by the server in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as the method embodiment will not be described in detail here.

[0132] The embodiment of the present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the data processing method.

[0133] It should be noted here that the above-mentioned storage medium provided in the present application can correspondingly implement all the method steps implemented by the server in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as the method embodiment will not be described in detail here.

[0134] Correspondingly, an embodiment of the present application further provides a computer program product, which includes a computer program code. When the computer program code is run on a computer, the computer executes the data processing method.

[0135] It should be noted here that the above-mentioned program product provided by the present application can correspondingly implement all the method steps implemented by the server in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as the method embodiment will not be specifically described here.

[0136] The embodiment of the present application also provides a chip, including a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program from the memory to execute the data processing method.

[0137] It should be noted here that the above-mentioned chip provided in the present application can correspondingly implement all the method steps implemented by the server in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as the method embodiment will not be described in detail here.

[0138] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or a non-transitory medium) and a communication medium (or a temporary medium).

[0139] As is well known to those skilled in the art, the term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information, such as computer-readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0140] Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0141] In the description of the embodiments of the present application, the term "and / or" merely represents an association relationship that describes associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" represents any combination of at least two of any one or more of a plurality of. For example, at least one of A, B, may represent any one or more elements selected from a set that communicates A, B, and C. In addition, the term "plurality" means two or more, unless otherwise precisely and specifically specified.

[0142] In the description of the embodiment of the present application, the terms "first", "second", "third", "fourth", etc. (if present) are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0143] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some or all of the technical group features therein. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A data processing method, characterized in that: include: Classify the original feature set of the customer group to be classified to obtain a number of group features under each category, wherein the categories are asset holding category, capital flow category, and transaction behavior category; Detect the correlation between several group characteristics under each category, and determine the number of dimensionality reduction factors and corresponding group characteristic factors of each category based on the correlation detection results; Obtain the correlation between several group characteristics under each category and each group characteristic factor; The group characteristic factor loading matrix of each category is established based on the correlation between several group characteristics under each category and each group characteristic factor; Based on the group characteristic factor loading matrix, the corresponding group characteristics in each category are respectively attributed to the group characteristic factor with the greatest correlation, so as to obtain the factor analysis result; Performing cluster analysis on the customer group to be classified based on the factor analysis result, and obtaining various subgroups of the customer group to be classified based on the cluster analysis result; The method of determining the number of dimensionality reduction factors and corresponding group characteristic factors of each category based on the correlation detection result includes: Obtaining a correlation matrix of several group characteristics of each category based on the correlation detection result, and obtaining a group characteristic value of the correlation matrix of each category; The number of dimensionality reduction factors and the corresponding group characteristic factors of each category are determined based on the group characteristic values ​​of the correlation matrix of each category and the total number of corresponding group characteristics. The corresponding group characteristic factors of the asset holding class are channel preferences, online channel contract signing and usage records, and consumption records.

2. The method according to claim 1, characterized in that Before classifying the original feature set of the customer group to be classified, it also includes: Determining a feature category based on business information corresponding to the customer group to be classified; The classifying of the original feature set of the customer group to be classified includes: classifying each group feature in the original data set into a corresponding feature category based on basic information of group features, wherein the basic information of group features includes a feature name or feature description information.

3. The method according to claim 1, characterized in that: The detection of correlations between several group characteristics under each category includes: The Bartlett sphericity test was performed on several group characteristics under each category; Based on the results of Bartlett's sphericity test, group characteristics and their correlations are obtained from several group characteristics under each category for group characteristic factor analysis.

4. The method according to claim 2, characterized in that: The detecting of the correlation between the plurality of group characteristics under each category also includes: Conduct statistical KMO tests on several group characteristics under each category; The group characteristics and their correlations are obtained from among several group characteristics under each category for performing group characteristic factor analysis based on the Bartlett sphericity test results, including: the group characteristics and their correlations are obtained from among several group characteristics under each category for performing group characteristic factor analysis based on the statistical quantity test results and the Bartlett sphericity test results.

5. The method according to claim 1, characterized in that After obtaining various subgroups of the customer group to be classified based on the cluster analysis results, the method further includes: Based on the distribution of group characteristic factors of each sub-group, group analysis is conducted on the business characteristics of each sub-group.

6. A data processing device, characterized in that: include: A feature classification module is configured to classify the original feature set of the customer group to be classified, and obtain a number of group features under each category, wherein the categories are asset holding category, capital flow category, and transaction behavior category; A correlation detection module configured to detect correlations between a number of group characteristics under each category; A factor determination module, which is configured to determine the number of dimensionality reduction factors and corresponding group characteristic factors for each category based on the correlation detection results; A factor analysis module is configured to obtain the correlation between a number of group characteristics under each category and each group characteristic factor, and to attribute the corresponding group characteristics in each category to the group characteristic factor with the greatest correlation, thereby obtaining the factor analysis result; A customer group classification module, which is configured to perform cluster analysis on the customer group to be classified based on the factor analysis result, and obtain subgroups of the customer group to be classified based on the cluster analysis result; The factor determination module is specifically configured to: obtain a correlation matrix of several group characteristics of each category based on the correlation detection result, and obtain a group characteristic value of the correlation matrix of each category; Determine the number of dimensionality reduction factors and corresponding group characteristic factors of each category based on the group characteristic values ​​of the correlation matrix of each category and the total number of corresponding group characteristics; The device also includes: A matrix building module, which is configured to build a group characteristic factor loading matrix of each category based on the correlation between a number of group characteristics under each category and each group characteristic factor; The corresponding group characteristics in each category are respectively attributed to the group characteristic factors with the greatest correlation. Specifically, based on the group characteristic factor loading matrix, the corresponding group characteristics in each category are respectively attributed to the group characteristic factors with the greatest correlation. The corresponding group characteristic factors of the asset holding class are channel preference, online channel contract signing and usage records, and consumption records.

7. An electronic device, characterized in that: include: Memory and processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the electronic device executes the data processing method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the data processing method according to any one of claims 1 to 5.

9. A computer program product, characterized in that The computer program product comprises a computer program code, and when the computer program code is run on a computer, the computer is enabled to execute the data processing method according to any one of claims 1 to 5.

Images