Customer classification method and apparatus

Abstract

The embodiment of the application provides a customer classification method and device, relates to the field of data analysis, and can also be applied to the financial field. The method comprises the following steps: dividing a customer set into an online consumption customer set and an offline consumption customer set, and respectively determining consumption intervals, consumption frequencies and consumption amounts; respectively performing clustering, and determining the dividing points of the consumption intervals, the consumption frequencies and the consumption amounts of the online consumption customer set and the offline consumption customer set according to the clustering results; and respectively determining the value classification results of each customer in the online consumption customer set and the offline consumption customer set according to the dividing points and a value classification rule. The embodiment of the application considers the differences in customer consumption behaviors in multiple scenarios, adopts a differentiated RFM value calculation mode, constructs different models and uses a segmentation mode, and finally obtains individualized classification results.

Description

Technical Field

[0001] This application relates to the field of data analysis, and can also be applied to the financial field, specifically to a customer classification method and apparatus. Background Technology

[0002] With the rapid development of internet technology, differentiated customer segmentation research has become a key focus for researchers. In refined credit card marketing, it's crucial not only to clearly define customers but also to establish the business usage scenarios after segmentation. Traditional customer segmentation methods, such as the RFM (Real-Time Metrics) method, utilize a combination of customer loyalty, customer size, and customer credit to segment customers, thereby improving customer acquisition capabilities and achieving precise marketing. Existing RFM customer value segmentation methods often employ mean, median, or interval scoring followed by mean or KMeans clustering, but there are few specific solutions explored for the banking credit card sector.

[0003] However, credit card spending in the banking industry generally includes both online and offline transactions. Online transactions are more frequent but smaller in value, while offline transactions are less frequent but larger in value. Current customer segmentation methods cannot address the need for different channels and marketing strategies for online and offline customers.

[0004] Furthermore, the current rate of new credit card customers is relatively high. The current customer segmentation method uses the same standard to classify the spending of newly opened customers and long-term customers who have had their cards for a long time, which raises questions about the accuracy of the new and old customer segmentation.

[0005] Furthermore, the customer segmentation results of clustering are usually largely dependent on the data distribution. If the conclusions of K-Means clustering are used directly for strategy recommendations, the uncertainty of the clustering results will render the strategy recommendations ineffective, and it will also pose a significant professional challenge for technical personnel without business experience. Summary of the Invention

[0006] To address the problems in the prior art, this application provides a customer classification method and apparatus that can consider the differences in customer consumption behavior in multiple scenarios, use differentiated calculation of RFM values, construct different models and use segmentation methods to finally obtain personalized classification results.

[0007] To solve at least one of the above problems, this application provides the following technical solution:

[0008] According to a first aspect of the embodiments of this application, this application provides a customer classification method, including:

[0009] According to the consumption scene and consumption details of the obtained customer set, the customer set is divided into an online consumption customer set and an offline consumption customer set, and consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set in a preset time period are respectively determined;

[0010] The consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are respectively clustered to obtain corresponding clustering results, and the consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are determined according to the clustering results.

[0011] According to the division points and the value classification rule, the value classification results of each customer in the online consumption customer set and the offline consumption customer set are respectively determined.

[0012] According to any embodiment of the present application, after the consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set in a preset time period are respectively determined according to the consumption details, the following steps are further included:

[0013] According to the consumption details, the time interval of each customer opening a consumption account is determined.

[0014] In the case that the time interval is less than the preset time period, the weight coefficients of the consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are respectively determined according to the time interval, the consumption frequencies and the preset time period, wherein the weight coefficients are negatively correlated with the time interval.

[0015] The consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are updated according to the weight coefficients.

[0016] According to any embodiment of the present application, after the consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set in a preset time period are respectively determined according to the consumption details, the following steps are further included:

[0017] Based on a preset grid space, the consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are divided into a plurality of grid regions, and the sample density of each grid region is respectively determined.

[0018] According to the sample density, the outlier factor of each grid region is determined, and the outlier factor is removed to obtain the consumption intervals, consumption frequencies and consumption amounts after the outlier factor is removed.

[0019] According to any of the embodiments of the present application, the clustering of the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set respectively obtains corresponding clustering results, which comprises:

[0020] The mean normalization of the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set respectively obtains corresponding mean normalization results;

[0021] The mean clustering of the mean normalization results obtains corresponding clustering results.

[0022] According to any of the embodiments of the present application, the determination of the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set according to the clustering results comprises:

[0023] According to the sample quantity of multiple different quantile points in the clustering results, the clustering results are divided by cluster to determine corresponding multiple clusters and a box plot;

[0024] According to the upper limit value and the lower limit value of the multiple clusters in the box plot, the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are determined.

[0025] According to any of the embodiments of the present application, the determination of the value classification result of each customer in the online consumption customer set and the offline consumption customer set according to the division point and the value classification rule comprises:

[0026] According to the proportion result of the division point of the consumption interval, the consumption frequency and the consumption amount in the high position, the value classification result of each customer is determined.

[0027] According to the second aspect of the embodiments of the present application, the present application provides a customer classification device, which comprises:

[0028] The sample acquisition module is used for dividing the customer set into an online consumption customer set and an offline consumption customer set according to the consumption scene and the consumption details of the acquired customer set, and determining the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set in a preset time period respectively;

[0029] The sample clustering module is used for clustering the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set respectively to obtain corresponding clustering results, and determining the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set according to the clustering results;

[0030] a sample classification module, configured to determine a value classification result of each customer in the online consumption customer set and the offline consumption customer set respectively according to the demarcation point and the value classification rule.

[0031] According to any of the embodiments of the present application, after the sample acquisition module determines the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set respectively within the preset time period according to the consumption details, the method further comprises a time weighting module, configured to:

[0032] determine the time interval of each customer opening a consumption account according to the consumption details;

[0033] in the case that the time interval is less than the preset time period, determine the weight coefficient of the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set respectively according to the time interval, the consumption frequency and the preset time period, wherein the weight coefficient is negatively correlated with the time interval;

[0034] update the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set according to the weight coefficient.

[0035] According to any of the embodiments of the present application, after the sample acquisition module determines the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set respectively within the preset time period according to the consumption details, the method further comprises an outlier factor elimination module, configured to:

[0036] divide the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set into a plurality of grid areas based on a preset grid space, and determine the sample density of each grid area respectively;

[0037] determine the outlier factor of each grid area according to the sample density, and eliminate the outlier factor to obtain the consumption interval, the consumption frequency and the consumption amount after elimination of the outlier factor.

[0038] According to any of the embodiments of the present application, when the sample clustering module clusters the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set respectively to obtain the corresponding clustering result, the sample clustering module is specifically configured to:

[0039] perform mean normalization on the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set respectively to obtain the corresponding mean normalization result;

[0040] perform mean clustering on the mean normalization result to obtain the corresponding clustering result.

[0041] According to any of the embodiments of the present application, the sample clustering module is specifically used for:

[0042] According to the number of samples at different quantile points in the clustering result, the clustering result is divided cluster by cluster to determine a plurality of corresponding cluster categories and a box plot;

[0043] According to the upper limit value and the lower limit value of the plurality of cluster categories in the box plot, the boundary points of the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are determined respectively.

[0044] According to any of the embodiments of the present application, the sample classification module is specifically used for:

[0045] According to the proportion result of the boundary points of the consumption interval, the consumption frequency and the consumption amount being in a high position, the value classification result of each customer is determined.

[0046] According to a third aspect of the embodiments of the present application, the present application provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to realize the steps of the customer classification method.

[0047] According to a fourth aspect of the embodiments of the present application, the present application provides a computer readable storage medium, which stores a computer program, wherein the computer program is executed by a processor to realize the steps of the customer classification method.

[0048] According to a fifth aspect of the embodiments of the present application, the present application provides a computer program product, comprising computer programs / instructions, wherein the computer programs / instructions are executed by a processor to realize the steps of the customer classification method.

[0049] According to the technical solution, the application provides a customer classification method and device. The customer set is divided into an online consumption customer set and an offline consumption customer set according to the consumption scene and consumption details of the obtained customer set, and the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set in a preset time period are determined respectively. The consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set are clustered respectively to obtain corresponding clustering results, and the dividing points of the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set are determined according to the clustering results. According to the dividing points and the value classification rule, the value classification results of each customer in the online consumption customer set and the offline consumption customer set are determined respectively. The differences in customer consumption behaviors in multiple scenes are considered, the RFM value is calculated in a differentiated manner, different models are constructed and the segmentation method is used, and finally the personalized classification results are obtained. BRIEF DESCRIPTION OF DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0051] Figure 1 One of the flowcharts of the customer classification method in the embodiments of the present application;

[0052] Figure 2 The second flowchart of the customer classification method in the embodiments of the present application;

[0053] Figure 3 The third flowchart of the customer classification method in the embodiments of the present application;

[0054] Figure 4 One of the box plots of the customer classification method in the embodiments of the present application;

[0055] Figure 5 The second box plot of the customer classification method in the embodiments of the present application;

[0056] Figure 6 The fourth flowchart of the customer classification method in the embodiments of the present application;

[0057] Figure 7 The interval time diagram of the customer classification method in the embodiments of the present application;

[0058] Figure 8Figure 5 is a flowchart of a customer classification method according to an embodiment of the present application;

[0059] Figure 9 Figure 6 is a structural diagram of a customer classification device according to an embodiment of the present application;

[0060] Figure 10 Figure 7 is a structural diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0061] In order to make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will be combined with the accompanying drawings for the embodiments of the present application to make a clear and complete description of the technical solutions of the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0062] The acquisition, storage, use, and processing of data in the technical solutions of the present application comply with relevant provisions of national laws and regulations.

[0063] In view of the problem that current customer stratification methods cannot adopt different channels and strategies for marketing online and offline customers, the present application provides a customer classification method and device.

[0064] In order to consider the differences in customer consumption behaviors in multiple scenarios, a differentiated RFM value calculation method is used to construct different models and use segmentation methods to finally obtain personalized classification results. The present application provides an embodiment of a customer classification method, as shown in Figure 1 , which specifically includes the following content:

[0065] Step S101: According to the consumption scenarios and consumption details of the obtained customer set, the customer set is divided into an online consumption customer set and an offline consumption customer set, and the consumption interval, consumption frequency, and consumption amount of the online consumption customer set and the offline consumption customer set in a preset time period are determined respectively.

[0066] First, the consumption scenarios and consumption details of the customer set are obtained, and then according to these data, the customer set is divided into two different parts, namely the online consumption customer set and the offline consumption customer set. For example, the customer set can be divided according to the different scenarios selected by the customer during consumption, i.e., online consumption or offline consumption.

[0067] For the online consumption customer set and the offline consumption customer set in a preset time period, three key indicators can be determined: consumption interval, consumption frequency, and consumption amount.

[0068] The consumption interval (Recency) refers to the consumption time interval of each customer from the last consumption within a preset time period. The smaller the R value, the greater the value, and such users are also the groups that can react to products and activities the most;

[0069] The consumption frequency (Frequency) represents the frequency of consumption of each customer within a preset time period, which can be determined by calculating the number of consumptions of the customer within the preset time period. The greater the F value, the more transactions the user has within a fixed time period;

[0070] The consumption amount (Monetary) refers to the total consumption amount of each customer within a preset time period. The greater the M value, the greater the consumption ability of the user.

[0071] For example, the consumption and personal card information of credit card customers can be obtained from a bank credit card customer database, including: credit card customer consumption places, consumption time, consumption amount, and card issuance date. Based on the influence of the scene on customer consumption behavior, the online and offline consumption of customers is divided into online consumption customer set X={x1, x2,..., x n} and offline consumption customer set Y={y1, y2,..., y m}.

[0072] Where n and m are the number of online consumption customers and offline consumption customers, respectively. For each customer set, set the observation period as P, and calculate the consumption time interval R, consumption frequency F, and consumption amount M within the card issuance observation period through consumption details.

[0073] Step S102: clustering the consumption interval, consumption frequency, and consumption amount of the online consumption customer set and the offline consumption customer set, respectively, to obtain the corresponding clustering results, and determining the dividing points of the consumption interval, consumption frequency, and consumption amount of the online consumption customer set and the offline consumption customer set according to the clustering results.

[0074] First, the online consumption customer set and the offline consumption customer set are clustered and analyzed, and data points with similar characteristics are grouped together to form different clusters or categories. The consumption interval, consumption frequency, and consumption amount of the online consumption customer set and the offline consumption customer set can be clustered using a clustering algorithm.

[0075] In an embodiment of the customer classification method of the present application, the consumption interval, consumption frequency, and consumption amount of the online consumption customer set and the offline consumption customer set are clustered respectively to obtain the corresponding clustering results, referring to Figure 2 may specifically include the following contents:

[0076] S102A: respectively, the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are mean normalized to obtain corresponding mean normalized results.

[0077] S102B: the mean normalized results are mean clustered to obtain corresponding clustering results.

[0078] Exemplarily, a plurality of k-Means clustering model constructions can be performed on the consumption time interval R, the consumption frequency F and the consumption amount M respectively, the sum of squared errors (SSE) of each clustering result is calculated, and the K-Meams optimal clustering number s of each customer group is found according to the elbow rule, and the clustering result under the optimal clustering number is taken as the final clustering.

[0079] Specifically, for the consumption time interval R, the consumption frequency F and the consumption amount M, a plurality of k-Means clustering models can be constructed respectively. For each clustering model, the sum of squared errors SSE of its clustering result is calculated, which is used to represent the sum of squares of distances between each data point and the center point of its cluster. A smaller SSE value represents a better clustering effect, and the data points are closer to the cluster center point.

[0080] The elbow method is used to determine the optimal clustering number of each customer group. The elbow method is suitable for the case where the K value is relatively small. The best K value is determined by drawing a curve between the K value and the corresponding SSE. When the selected K value is less than the true K, increasing the K value will significantly reduce the SSE value. When the selected K value is greater than the true K, increasing the K value has no obvious effect on the SSE value. Therefore, there will be a turning point similar to the elbow on the curve, and the K value corresponding to the point is the optimal clustering number.

[0081] After obtaining the optimal clustering number of each customer group according to the elbow method, the k-Means algorithm is run using the optimal clustering number to obtain the final clustering result.

[0082] Step 103: according to the demarcation point and the value classification rule, the value classification result of each customer in the online consumption customer set and the offline consumption customer set is determined.

[0083] Based on the obtained demarcation point and value classification rule, each customer can be classified according to his online or offline consumption behavior.

[0084] For the online consumption customer set, according to the comparison of the consumption interval, consumption frequency and consumption amount of each customer with the online consumption demarcation point, the value classification result of each customer can be determined. For example, for a customer with a short consumption interval, a high consumption frequency and a large consumption amount, the customer can be classified as an important value customer. For a customer with a long consumption interval, a low consumption frequency and a small consumption amount, the customer can be classified as a general value customer, and so on. Similarly, for the offline consumption customer set, according to the comparison of the consumption interval, consumption frequency and consumption amount of each customer with the offline consumption demarcation point, the value classification result of each customer can be determined. According to the consumption behavior characteristics of the customers, the customers are divided into different value classifications, such as important value customers, general value customers and the like.

[0085] In an embodiment of the customer classification method of the present application, the determination of the consumption interval, consumption frequency and consumption amount demarcation points of the online consumption customer set and the offline consumption customer set according to the clustering result, see Figure 3 may further specifically include the following contents:

[0086] Step S103A: according to the sample quantity of a plurality of different quantile points in the clustering result, the clustering result is divided by cluster, a plurality of corresponding clusters and a box plot are determined;

[0087] Step S103B: according to the upper limit value and the lower limit value of the plurality of clusters in the box plot, the consumption interval, consumption frequency and consumption amount demarcation points of the online consumption customer set and the offline consumption customer set are respectively determined.

[0088] First, according to the sample quantity of each cluster in the clustering result, the sample quantity of a plurality of different quantile points is calculated. These quantile points can represent different clusters.

[0089] As shown in Figure 4 Next, according to the sample quantity of different quantile points, the clustering result is divided by cluster, and the division of the upper limit value and the lower limit value of the consumption interval, consumption frequency and consumption amount in each cluster is performed in combination with the box plot constructed for each cluster according to the value classification rule shown in Figure 5 to reflect the distribution of the data in the cluster.

[0090] The present application adopts the method of taking the mean value of the upper limit and the lower limit between clusters to obtain the RFM demarcation point, so as to further refine the K-MEANS clustering result, thereby obtaining a customer stratification result which is more easy to use and more instructive for marketing strategy.

[0091] Table 1 shows an example of the distribution of the data in the cluster according to the present disclosure:

[0092] Table 1

[0093]

[0094] From Table 1, it can be seen that the cyan cluster: the range of variable R (consumption interval) from the lower limit to the upper limit is wide, indicating that there is a large difference in the consumption interval of the customers in this cluster. The variable F (consumption frequency) also has a large range, indicating that there is a difference in the consumption frequency of the customers in this cluster. The range of variable M (consumption amount) from the lower limit to the upper limit is very wide, indicating that there is a great difference in the consumption amount of the customers in this cluster.

[0095] The green cluster: the range of variable R from the lower limit to the upper limit is small, indicating that the consumption interval of the customers in this cluster is relatively stable. The consumption interval of most customers is concentrated between 0 and 2 days. Variable F shows an increasing trend in the consumption frequency of customers. The range of variable M from the lower limit to the upper limit is also large, indicating that there is a large difference in the consumption amount of the customers in this cluster.

[0096] The pink cluster: compared with the cyan cluster, the range of variable R (consumption interval) is small, and the consumption interval of most customers is concentrated between 2 and 22 days. Variable F shows an increasing trend in the consumption frequency of customers. The range of variable M (consumption amount) is large, with a 75% quantile of 206,722.54, indicating that a part of the customers have a high consumption amount.

[0097] The orange cluster: variable R: the range of consumption interval is small, and the consumption interval of most customers is concentrated between 141 and 237 days. Variable F: the consumption frequency is low, and most customers have only a small amount of consumption in a period of time. Variable M: the range of consumption amount is narrow, indicating that the consumption amount of most customers is low.

[0098] Through the analysis of Table 1, the differences in the consumption interval, consumption frequency and consumption amount of the customers in different clusters can be understood.

[0099] Finally, according to the upper limit value and the lower limit value of the multiple clusters in the box plot, the dividing points of the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set can be determined respectively, and the dividing points will be used to divide the customers into different consumption groups, so as to further understand the consumption behavior characteristics of the customers.

[0100] In an embodiment of the customer classification method of the present application, the determination of the value classification result of each customer in the online consumption customer set and the offline consumption customer set according to the dividing points and the value classification rule comprises:

[0101] According to the proportion result of the dividing points of the consumption interval, the consumption frequency and the consumption amount being in a high position, the value classification result of each customer is determined.

[0102] Among them, for each customer, according to the comparison of its consumption interval, consumption frequency and consumption amount with the dividing point, determine its belonging category, and obtain the proportion of the customer in the high position in these categories, according to the result of the high position proportion, determine the value classification of each customer. Customers can be classified as high-value customers, medium-value customers or low-value customers, and the specific classification depends on their high position proportion in different variables.

[0103] According to the result of each value classification, the corresponding marketing strategy is generated. High-value customers may receive more attention and rewards to improve customer loyalty and encourage more consumption, medium-value customers may receive regular marketing activities to stimulate their consumption potential, and low-value customers may need to be re-attracted and retained, and personalized promotion activities can be used to re-stimulate their interest.

[0104] It can be understood that the classification result will be different from the above example due to different scenarios, and can be guided according to the actual situation.

[0105] Table 2 and Table 3 respectively show the marketing strategy table of online credit card consumption customers and a kind of offline credit card consumption customers according to the classification structure shown in the table. Figure 5 The marketing strategy table shown in the table

[0106] Table 2

[0107]

[0108] Table 3

[0109]

[0110]

[0111] The scheme described in the present application considers that different consumption scenarios will lead to differences in customer consumption behavior, for example, online consumption may be more convenient and frequent, while offline consumption may be more money-biased. Therefore, based on the difference calculation of RFM value, a model for different consumption scenarios is constructed, which can consider the differences in customer consumption behavior in multiple scenarios, adopt the way of difference calculation of RFM value, construct different models and use subdivision method, and finally obtain personalized classification result, to better meet the needs of customers in different consumption scenarios, and provide more targeted marketing and service strategies for financial institutions.

[0112] From the above description, it can be known that the customer classification method provided by the embodiments of the present application can consider the differences in customer consumption behavior in multiple scenarios, adopt the way of difference calculation of RFM value, construct different models and use subdivision method, and finally obtain its personalized classification result.

[0113] In order to effectively solve the problem of different observation time intervals of new card customers and old customers in the observation period, make the customer stratification result more referential, and enhance the accuracy of the clustering result, in an embodiment of the customer classification method, after the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set in the preset time period are respectively determined according to the consumption details, see Figure 6 The application can also specifically include the following contents:

[0114] Step S201: According to the consumption details, the time interval of each customer opening a consumption account is determined.

[0115] According to the consumption details of the customers, the time interval of each customer opening a consumption account is determined. This time interval reflects the length of time when the customer becomes a consumer of the bank, and can be used to distinguish new card customers and old customers.

[0116] Step S202: In the case where the time interval is less than the preset time period, the weight coefficients of the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are respectively determined according to the time interval, the consumption frequency and the preset time period, wherein the weight coefficients are negatively correlated with the time interval.

[0117] For customers whose time interval is less than the preset time period, the weight coefficients of the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are determined according to their time interval, consumption frequency and preset time period. The purpose of the weight coefficient is to differentiate customers with different time intervals, so as to better reflect their consumption behavior characteristics. Here, the weight coefficient is negatively correlated with the time interval, that is, the shorter the time interval, the greater the weight coefficient, and vice versa.

[0118] Step S203: According to the weight coefficients, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are updated.

[0119] According to the weight coefficients determined before, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are updated. By applying the weight coefficients, the consumption frequency and the consumption amount of the customers can be adjusted to make them more referential and accurate.

[0120] For example, the time interval I = {I1, I1,...} since the card was issued can be based on the time interval. The time interval I = {I1, I1,...} calculates the time distance weight W for each customer. As shown in Figure 7 For example, the online consumption frequency, assuming that the customer's consumption behavior in the observation period P-I has a probability of a that is the same as the consumption distribution of the time interval I, then I:(P-I) = F:F xa , F x a represent the same distribution of consumption frequency under a probability a.

[0121] Combining the customer's online consumption frequency F to eliminate the impact of different life cycle consumption behaviors, the time distance weight of the ith customer is

[0122]

[0123] The updated customer RFM value is the consumption time interval R w , consumption weighted frequency F w and consumption weighted amount M w In addition, the characteristics of high frequency and low amount online, high amount and low frequency offline, different values of a and b online and offline, the detailed calculation method is as follows:

[0124] R i w = R i , F i w = W i a F i , M i w = W i b M i

[0125] Online: Offline:

[0126] It can be understood that the probability value a of the above time distance weight calculation can be adjusted according to the actual scene characteristics in the application.

[0127] In the subsequent steps, the standard deviation standardization method can be used to normalize the above three variables to zero-mean, to obtain the corresponding indicators R s w , F s w and M s w to eliminate the dimension problem of distance calculation in customer clustering process, so that different indicators are comparable, and finally improve the algorithm iteration solving accuracy.

[0128] Through the above steps, the application can effectively solve the problem of different observation time intervals between new card customers and old customers during the observation period, make the customer stratification results more referential, and enhance the accuracy of the clustering results. By determining the time interval for opening a consumption account according to the consumption details, determining the weight coefficient according to the time interval, consumption frequency and preset time period, and updating the consumption frequency and consumption amount according to the weight coefficient, the consumption behavior characteristics of the customers can be better reflected, the accuracy of customer classification can be further improved, and the influence of the life cycle of the customers on the final stratification results can be weakened.

[0129] In order to effectively solve the problem of clustering the outliers as initial clustering centers, after determining the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set in the preset time period respectively according to the consumption details, referring to Figure 8 , the following contents can also be specifically included:

[0130] Step S301: Based on the preset grid space, the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set are divided into a plurality of grid regions, and the sample density of each grid region is determined.

[0131] Based on the preset grid space, the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set are divided into a plurality of grid regions. Each grid region represents a group of customer samples with similar consumption behavior characteristics. Then, for each grid region, the sample density in the region, i.e. the number of customer samples contained, is determined.

[0132] Step S302: According to the sample density, determine the outlier factor of each grid region, and remove the outlier factor to obtain the consumption interval, consumption frequency and consumption amount after the outlier factor is removed.

[0133] According to the sample density of each grid region, the outlier factor in the region is calculated. The outlier factor represents the customer sample with significantly different consumption behavior characteristics compared with other samples. By identifying the outlier factor, these abnormal data points can be removed to avoid their influence on the clustering results. Therefore, after the outlier factor is removed, more accurate and stable consumption interval, consumption frequency and consumption amount data are obtained for subsequent clustering analysis.

[0134] For example, the above two types of customers are monitored for outliers based on variable grid local outlier points. This method divides the data into a plurality of grid regions in the grid space, and then searches for the k-nearest neighbors of the samples in the grid space, and calculates the local outlier factor (LOF) of each data sample:

[0135]

[0136] For the i-th customer, its local density is:

[0137]

[0138] wherein is the k-th reachable distance of the customer relative to its near neighbor customers, is the Euclidean distance between the customer and the k-th nearest neighbor sample.

[0139] In summary, through the above steps, the application can effectively solve the problem of clustering outliers as initial clustering centers. By dividing the grid space and determining the sample density, the outlier factor with abnormal consumption behavior characteristics is identified and excluded, ensuring the accuracy and stability of the clustering analysis. In this way, the consumption interval, consumption frequency and consumption amount data obtained are more reliable, which can provide more accurate basis for subsequent customer classification and personalized marketing strategies.

[0140] In order to consider the differences in customer consumption behavior in multiple scenarios, a differentiated RFM value calculation method is used to build different models and use segmentation methods to ultimately obtain personalized classification results. The application provides an embodiment of a customer classification device for implementing all or part of the contents of the customer classification method, as shown in Figure 9 , which specifically includes the following contents:

[0141] The sample acquisition module 1101 is configured to divide the customer set into online consumption customer set and offline consumption customer set according to the consumption scene and consumption details of the acquired customer set, and determine the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set in a preset time period respectively.

[0142] The sample clustering module 1102 is configured to cluster the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set respectively to obtain the corresponding clustering results, and determine the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set according to the clustering results.

[0143] The sample classification module 1103 is configured to determine the value classification result of each customer in the online consumption customer set and the offline consumption customer set according to the demarcation point and the value classification rule.

[0144] According to any embodiment of the application, after the sample acquisition module determines the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set in a preset time period according to the consumption details, a time weighting module is further included, which is configured to:

[0145] According to the consumption details, a time interval for each customer to open a consumption account is determined respectively;

[0146] In a case where the time interval is less than the preset time period, a consumption frequency and a weight coefficient of a consumption amount of the online consumption customer set and the offline consumption customer set are determined according to the time interval, the consumption frequency and the preset time period, wherein the weight coefficient is negatively correlated with the time interval;

[0147] The consumption frequency and the weight coefficient of the consumption amount of the online consumption customer set and the offline consumption customer set are updated according to the weight coefficient.

[0148] According to any one of the embodiments of the present application, after the sample acquisition module determines the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set in a preset time period according to the consumption details, the outlier factor elimination module is further included, which is configured to:

[0149] The consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are divided into a plurality of grid areas based on a preset grid space, and a sample density of each grid area is determined respectively;

[0150] The outlier factor of each grid area is determined according to the sample density, and the outlier factor is eliminated to obtain the consumption interval, the consumption frequency and the consumption amount after the outlier factor elimination.

[0151] According to any one of the embodiments of the present application, the sample clustering module is specifically configured to:

[0152] The consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are subjected to mean normalization respectively to obtain corresponding mean normalized results;

[0153] The mean normalized results are subjected to mean clustering to obtain corresponding clustering results.

[0154] According to any one of the embodiments of the present application, the sample clustering module is specifically configured to:

[0155] According to the sample quantity of a plurality of different quantile points in the clustering results, the clustering results are subjected to cluster division to determine a plurality of corresponding clusters and a box plot;

[0156] According to the upper limit value and the lower limit value of each cluster in the box plot, the dividing points of the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are determined respectively.

[0157] According to any embodiment of the present application, when determining the value classification result of each customer in the online consumption customer set and the offline consumption customer set according to the dividing points and the value classification rule, the sample classification module is specifically used for:

[0158] According to the proportion result of the dividing points of the consumption interval, the consumption frequency and the consumption amount being in a high position, the value classification result of each customer is determined.

[0159] From the above description, it can be known that the customer classification device provided by the embodiments of the present application can consider the differences in customer consumption behaviors in multiple scenarios, adopt a differentiated calculation of RFM values, construct different models and use a segmentation method, and finally obtain a personalized classification result.

[0160] From the hardware level, in order to consider the differences in customer consumption behaviors in multiple scenarios, adopt a differentiated calculation of RFM values, construct different models and use a segmentation method, and finally obtain a personalized classification result, the present application provides an embodiment of an electronic device for implementing all or part of the contents of the customer classification method, which specifically includes the following contents:

[0161] A processor, a memory, a communications interface and a bus; wherein the processor, the memory and the communications interface complete mutual communication through the bus; the communications interface is used for realizing information transmission between the customer classification device and a core business system, a customer terminal and a related database and other related devices; the logic controller can be a desktop computer, a tablet computer and a mobile terminal, and the present embodiment is not limited thereto. In the present embodiment, the logic controller can be implemented with reference to the embodiments of the customer classification method and the embodiments of the customer classification device, the contents of which are incorporated herein, and repeated descriptions are not repeated.

[0162] It can be understood that the customer terminal can include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device and the like. Among them, the smart wearable device can include smart glasses, a smart watch, a smart bracelet and the like.

[0163] In actual application, part of the customer classification method can be executed on the electronic device as described above, or all operations can be completed in the client device. Specifically, the selection can be made according to the processing capability of the client device and the restriction of the customer use scenario, etc. The present application does not limit this. If all operations are completed in the client device, the client device can further include a processor.

[0164] The client device described above can have a communication module (i.e., a communication unit) and can be communicatively connected with a remote server to realize data transmission with the server. The server can include a server of a task scheduling center side, and can also include a server of an intermediate platform in other implementation scenarios, such as a server of a third-party server platform communicatively connected with the server of the task scheduling center. The server can include a single computer device, or can include a server cluster composed of multiple servers, or a server structure of a distributed device.

[0165] Figure 10 A schematic block diagram of a system configuration of the electronic device 9600 of an embodiment of the present application is shown in FIG. 9. As shown in FIG. 9, the electronic device 9600 can include a central processor 9100 and a memory 9140; the memory 9140 is coupled to the central processor 9100. It is worth noting that the structure shown in FIG. 9 is exemplary; other types of structures can also be used to supplement or replace the structure to realize telecommunication functions or other functions. Figure 10 Figure 10 The structure shown in FIG. 9 is exemplary; other types of structures can also be used to supplement or replace the structure to realize telecommunication functions or other functions.

[0166] In an embodiment, the customer classification method function can be integrated into the central processor 9100. The central processor 9100 can be configured to perform the following control:

[0167] Step S101: According to the consumption scenario and consumption details of the obtained customer set, the customer set is divided into an online consumption customer set and an offline consumption customer set, and the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set in a preset time period are determined respectively.

[0168] Step S102: The consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set are respectively clustered to obtain corresponding clustering results, and the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set are determined according to the clustering results.

[0169] Step S103: According to the demarcation point and the value classification rule, the value classification result of each customer in the online consumption customer set and the offline consumption customer set is determined respectively.

[0170] ​From the above description, the electronic device provided by the embodiments of the present application considers the differences in customer consumption behaviors in multiple scenarios, adopts a differentiated RFM value calculation method, constructs different models and uses segmentation methods, and finally obtains personalized classification results.

[0171] In another embodiment, the customer classification device can be configured separately from the central processor 9100, for example, the customer classification device can be configured as a chip connected with the central processor 9100, and the customer classification method function is realized through the control of the central processor.

[0172] As shown in FIG. 9, the electronic device 9600 can further include a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It should be noted that the electronic device 9600 does not necessarily include all the components shown in FIG. 9; in addition, the electronic device 9600 can include components not shown in FIG. 9, which can be referred to the prior art. Figure 10 Figure 10 As shown in FIG. 9, the central processor 9100, also sometimes referred to as a controller or operating control, can include a microprocessor or other processor device and / or logic device, which receives input and controls the operation of the various components of the electronic device 9600. Figure 10 As shown in FIG. 9, the central processor 9100, also sometimes referred to as a controller or operating control, can include a microprocessor or other processor device and / or logic device, which receives input and controls the operation of the various components of the electronic device 9600.

[0173] Figure 10 The memory 9140, for example, can be one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, or other suitable device. Information related to failure can be stored, and in addition, programs for executing the information can be stored. The central processor 9100 can execute the programs stored in the memory 9140 to achieve information storage or processing, etc.

[0174] The input unit 9120 provides input to the central processor 9100. The input unit 9120 is, for example, a key or 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 text. The display can be, for example, an LCD display, but is not limited thereto.

[0175] The input unit 9120 provides input to the central processor 9100. The input unit 9120 is, for example, a key or 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 text. The display can be, for example, an LCD display, but is not limited thereto.

[0176] ​​The memory 9140 can be a solid state memory such as a read only memory (ROM), a random access memory (RAM), a SIM card, or the like. It can also be a memory that retains information even when power is off, can be selectively erased, and is provided with more data, an example of which is sometimes referred to as an EPROM or the like. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 can include an application / function storage 9142 for storing application programs and function programs or for storing a flow for executing an operation of the electronic device 9600 by the central processing unit 9100.

[0177] The memory 9140 can also include a data storage 9143 for storing data such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. A driver storage 9144 of the memory 9140 can include various drivers of the electronic device for a communication function and / or for performing other functions of the electronic device such as a messaging application, a phonebook application, and the like.

[0178] The communication module 9110 is a transmitter / receiver 9110 that transmits and receives signals via an antenna 9111. The communication module (transmitter / receiver) 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which can be the same as in the case of a conventional mobile communication terminal.

[0179] Based on different communication technologies, a plurality of communication modules 9110 such as a cellular network module, a Bluetooth module, and / or a wireless local area network module, and the like can be provided in the same electronic device. The communication module (transmitter / receiver) 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and receive audio input from the microphone 9132, thereby enabling a conventional telecommunication function. The audio processor 9130 can include any suitable buffer, decoder, amplifier, and the like. In addition, the audio processor 9130 is also coupled to the central processing unit 9100, thereby enabling recording on the local by the microphone 9132 and playing a sound stored on the local by the speaker 9131.

[0180] The embodiment of the present application further provides a computer readable storage medium capable of implementing all steps of the customer classification method in the above embodiment, wherein the computer program is stored on the computer readable storage medium, and the computer program is executed by a processor to implement all steps of the customer classification method in the above embodiment, for example, the computer program is executed by the processor to implement the following steps:

[0181] Step S101: According to the consumption scene and consumption details of the obtained customer set, the customer set is divided into an online consumption customer set and an offline consumption customer set, and consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set in a preset time period are determined respectively.

[0182] Step S102: The consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set are clustered respectively to obtain corresponding clustering results, and the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set are determined according to the clustering results.

[0183] Step S103: According to the demarcation point and the value classification rule, the value classification result of each customer in the online consumption customer set and the offline consumption customer set is determined respectively.

[0184] From the above description, it can be seen that the computer readable storage medium provided by the embodiment of the present application considers the difference of customer consumption behavior in multiple scenes, adopts a differentiated calculation of RFM value, constructs different models and uses a segmentation method, and finally obtains a personalized classification result.

[0185] The embodiment of the present application further provides a computer program product capable of implementing all steps of the customer classification method in the above embodiment, wherein the computer program / instruction is executed by a processor to implement the steps of the customer classification method, for example, the computer program / instruction implements the following steps:

[0186] Step S101: According to the consumption scene and consumption details of the obtained customer set, the customer set is divided into an online consumption customer set and an offline consumption customer set, and consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set in a preset time period are determined respectively.

[0187] Step S102: The consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set are clustered respectively to obtain corresponding clustering results, and the consumption interval, consumption frequency and consumption amount of the online consumption customer set and the offline consumption customer set are determined according to the clustering results.

[0188] Step S103: determining the value classification result of each customer in the online consumption customer set and the offline consumption customer set respectively according to the demarcation point and the value classification rule.

[0189] From the above description, it can be seen that the computer program product provided by the embodiment of the present application considers the difference in customer consumption behavior in multiple scenarios, adopts a differentiated RFM value calculation manner, constructs different models and uses a segmentation manner, and finally obtains a personalized classification result.

[0190] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, device, or computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a computer program product implemented on one or more computer usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer usable program code.

[0191] The present application is described with reference to flowcharts and / or block diagrams of the method, device (apparatus), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams 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 apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The device that implements the functions specified in one or more flows and / or blocks.

[0192] These computer program instructions can also be stored in a computer readable memory capable of guiding a computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer readable memory produce a product including instruction devices that implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The device that implements the functions specified in one or more flows and / or blocks.

[0193] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1steps of the functions specified in the one or more blocks.

[0194] The principles and implementation manners of the present application are described in the embodiments. The above description of the embodiments is only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application scopes will be changed. In summary, the content of the description should not be understood as a limitation of the present application.

Claims

1. A method of classifying customers, characterized by, The method comprises: According to the consumption scene and consumption details of the obtained customer set, the customer set is divided into an online consumption customer set and an offline consumption customer set, and consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set in a preset time period are determined respectively; The consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are clustered respectively to obtain corresponding clustering results, and the consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are determined according to the clustering results; According to the clustering results, the consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are determined, including: according to the sample quantity of a plurality of different quantile points in the clustering results, the clustering results are divided by cluster to determine a plurality of corresponding clusters and a box plot; according to the upper limit value and the lower limit value of the plurality of clusters in the box plot, the consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are determined respectively. After the consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set in the preset time period are determined according to the consumption details, the following steps are further included:

2. The method of claim 1, wherein, According to the consumption details, the time interval for each customer to open a consumption account is determined respectively; In the case that the time interval is less than the preset time period, the weight coefficients of the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are determined according to the time interval, the consumption frequency and the preset time period, wherein the weight coefficients are negatively correlated with the time interval; The consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set are updated according to the weight coefficients. After the consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set in the preset time period are determined according to the consumption details, the following steps are further included:

3. The method of claim 1, wherein, Based on a preset grid space, the consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are divided into a plurality of grid regions, and the sample density of each grid region is determined respectively; According to the sample density, the outlier factor of each grid region is determined, and the outlier factor is removed to obtain the consumption intervals, consumption frequencies and consumption amounts after the outlier factor is removed. The consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are clustered respectively to obtain corresponding clustering results, including:

4. The method of claim 1, wherein, The consumption intervals, consumption frequencies and consumption amounts of the online consumption customer set and the offline consumption customer set are subjected to mean normalization respectively to obtain corresponding mean normalization results; ​ The mean normalization result is subjected to mean clustering to obtain a corresponding clustering result.

5. The method of claim 1, wherein, The value classification result of each customer in the online consumption customer set and the offline consumption customer set is determined according to the demarcation point and a value classification rule. The value classification result of each customer is determined according to the proportion of the demarcation point of the consumption interval, the consumption frequency and the consumption amount being in a high position.

6. A client classification apparatus characterized by comprising: The device comprises: The sample obtaining module is configured to divide a customer set into an online consumption customer set and an offline consumption customer set according to the consumption scene and the consumption details of the customer set, and determine the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set in a preset time period respectively; The sample clustering module is configured to cluster the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set respectively to obtain a corresponding clustering result, and determine the demarcation point of the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set according to the clustering result; The sample classification module is configured to determine the value classification result of each customer in the online consumption customer set and the offline consumption customer set according to the demarcation point and a value classification rule. When determining the demarcation point of the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set according to the clustering result, the sample clustering module is specifically configured to divide the clustering result by cluster according to the sample quantity of a plurality of different demarcation points in the clustering result, determine a plurality of corresponding clusters and a box plot, and determine the demarcation point of the consumption interval, the consumption frequency and the consumption amount of the online consumption customer set and the offline consumption customer set according to the upper limit value and the lower limit value of the plurality of clusters in the box plot.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the steps of the customer classification method in any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the customer classification method in any one of claims 1 to 5.

9. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instruction is executed by the processor to implement the steps of the customer classification method in any one of claims 1 to 5.

Images