Data processing method, device, electronic device and storage medium

Abstract

The embodiments of the present disclosure provide a data processing method, device, electronic device and storage medium, which relate to the field of computer technology. The method includes: obtaining a customer data set, which includes index data of existing customers; clustering existing customers according to the index data of existing customers based on a self-organizing map network to obtain multiple customer groups; classifying multiple customer groups according to the index data of existing customers and the index data of customers included in each customer group to obtain customer classification results. The method can fully mine the information of the customer data set, cluster and further classify through the self-organizing map network, obtain accurate customer classification results, and improve the accuracy of customer segmentation.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular, to a data processing method, a data processing device, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] Customer segmentation is the division of a broad customer base into smaller subsets or groups that exhibit similarities in variables such as demographics, preferences, and consumption patterns. It is intended to help businesses and service providers allocate resources more effectively to meet the needs of different customer groups.

[0003] In related technologies, enterprises and service providers rely on market research to gain insights into customer behaviors and preferences and classify customers accordingly. However, this method has limitations, resulting in inaccurate classification results and difficulty in accurately matching the needs of different customer groups.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] Embodiments of the present disclosure provide a data processing method, a data processing device, an electronic device, a computer-readable storage medium, and a computer program product.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to one aspect of the present disclosure, a data processing method is provided, the method comprising: obtaining a customer data set; the customer data set comprising indicator data of existing customers; clustering existing customers according to the indicator data of existing customers based on a self-organizing mapping network to obtain a plurality of customer groups; classifying the plurality of customer groups according to the indicator data of the existing customers and the indicator data of customers included in each of the customer groups to obtain a customer classification result.

[0008] In some embodiments of the present disclosure, obtaining a customer data set includes: obtaining historical data of existing customers; preprocessing the historical data of existing customers to obtain preprocessed data; and extracting indicator data of existing customers from the preprocessed data based on a predefined customer relationship management model; wherein the indicator data of existing customers includes the length of time from the customer's most recent order to an analysis time point, the customer's customer level, the number of orders placed by the customer during the statistical period, and the total amount of the customer's orders during the statistical period.

[0009] In some embodiments of the present disclosure, the existing customers are clustered according to the indicator data of the existing customers based on the self-organizing map network to obtain a plurality of customer groups, including: inputting the indicator data of the existing customers into the self-organizing map network for training to obtain a trained self-organizing map network; based on the trained self-organizing map network, the existing customers are clustered to obtain the plurality of customer groups.

[0010] In some embodiments of the present disclosure, the multiple customer groups are divided according to the indicator data of the existing customers and the indicator data of the customers included in each of the customer groups to obtain customer classification results, including: determining the mean of the indicator data of the existing customers according to the indicator data of the existing customers; the mean of the indicator data of the existing customers includes the mean of the duration, the mean of the customer level, the mean of the number of orders, and the mean of the total amount of orders corresponding to the existing customers; for each of the customer groups, determining the mean of the indicator data of the customer group according to the indicator data of the customers in the customer group; the mean of the indicator data of the customer group includes the mean of the duration, the mean of the customer level, the mean of the number of orders, and the mean of the total amount of orders corresponding to the customer group; comparing the mean of the indicator data of the existing customers with the mean of the indicator data of the customer groups to obtain comparison results corresponding to the customer groups; dividing the multiple customer groups according to the comparison results corresponding to each of the customer groups to obtain the customer classification results.

[0011] In some embodiments of the present disclosure, determining the mean of the indicator data of the existing customers based on the indicator data of the existing customers includes: calculating the mean of the duration corresponding to the existing customers based on the duration from the most recent order of each customer among the existing customers to the analysis time point; calculating the mean of the customer level corresponding to the existing customers based on the customer level of each customer among the existing customers; calculating the mean of the number of orders corresponding to the existing customers based on the number of orders of each customer among the existing customers within the statistical period; calculating the mean of the total amount of orders corresponding to the existing customers based on the total amount of orders of each customer among the existing customers within the statistical period.

[0012] In some embodiments of the present disclosure, for each of the customer groups, the mean of the indicator data of the customer group is determined based on the indicator data of the customers in the customer group, including: for each of the customer groups, according to the duration from the most recent order of each customer in the customer group to the analysis time point, calculating the mean of the duration corresponding to the customer group; according to the customer level of each customer in the customer group, calculating the mean of the customer level corresponding to the customer group; according to the number of orders of each customer in the customer group during the statistical period, calculating the mean of the number of orders corresponding to the customer group; according to the total amount of orders of each customer in the customer group during the statistical period, calculating the mean of the total amount of orders corresponding to the customer group.

[0013] In some embodiments of the present disclosure, the comparison of the mean value of the indicator data of the existing customers with the mean value of the indicator data of the customer group to obtain the comparison result corresponding to the customer group includes: if the mean value of the duration corresponding to the customer group is greater than the mean value of the duration corresponding to the existing customers, marking the customer group as a first label on the time dimension indicator, otherwise, marking the customer group as a second label on the time dimension indicator; if the mean value of the customer level corresponding to the customer group is greater than the mean value of the customer level corresponding to the existing customers, marking the customer group as a first label on the customer level dimension indicator, otherwise Then, the customer group is marked as the second label on the customer level dimension indicator; if the average number of orders corresponding to the customer group is greater than the average number of orders corresponding to the existing customers, the customer group is marked as the first label on the order number dimension indicator, otherwise, the customer group is marked as the second label on the order number dimension indicator; if the average total order amount corresponding to the customer group is greater than the average total order amount corresponding to the existing customers, the customer group is marked as the first label on the order total amount dimension indicator, otherwise, the customer group is marked as the second label on the order total amount dimension indicator.

[0014] In some embodiments of the present disclosure, the multiple customer groups are classified according to the comparison results corresponding to each of the customer groups to obtain the customer classification results, including: performing comparative analysis on the labels of each of the multiple customer groups on different dimensional indicators; if the labels of two or more customer groups on different dimensional indicators are the same, merging the two or more customer groups to obtain a merged result; and obtaining the customer classification result based on the merged result.

[0015] According to another aspect of the present disclosure, a data processing device is provided, the device comprising: a data acquisition module, configured to acquire a customer data set; the customer data set comprises indicator data of existing customers; a clustering module, configured to cluster existing customers according to the indicator data of existing customers based on a self-organizing mapping network, to obtain a plurality of customer groups; a classification module, configured to classify the plurality of customer groups according to the indicator data of the existing customers and the indicator data of the customers included in each of the customer groups, to obtain a customer classification result.

[0016] In some embodiments of the present disclosure, the data acquisition module is further configured to: acquire the historical data of the existing customers; preprocess the historical data of the existing customers to obtain preprocessed data; and extract the indicator data of the existing customers from the preprocessed data based on a predefined customer relationship management model; wherein the indicator data of the existing customers include the length of time from the customer's most recent order to the analysis time point, the customer's customer level, the number of orders placed by the customer during the statistical period, and the total amount of the customer's orders during the statistical period.

[0017] In some embodiments of the present disclosure, the clustering module is further configured to: input the indicator data of the existing customers into the self-organizing map network for training to obtain a trained self-organizing map network; and cluster the existing customers based on the trained self-organizing map network to obtain the multiple customer groups.

[0018] In some embodiments of the present disclosure, the classification module is further configured to: determine the mean of the indicator data of the existing customers based on the indicator data of the existing customers; the mean of the indicator data of the existing customers includes the mean of the duration, the mean of the customer level, the mean of the number of orders and the mean of the total order amount corresponding to the existing customers; for each of the customer groups, determine the mean of the indicator data of the customer group based on the indicator data of the customers in the customer group; the mean of the indicator data of the customer group includes the mean of the duration, the mean of the customer level, the mean of the number of orders and the mean of the total order amount corresponding to the customer group; compare the mean of the indicator data of the existing customers with the mean of the indicator data of the customer groups to obtain the comparison results corresponding to the customer groups; divide the multiple customer groups according to the comparison results corresponding to each of the customer groups to obtain the customer classification results.

[0019] In some embodiments of the present disclosure, the classification module is further configured to: calculate the mean duration corresponding to the existing customers based on the duration from the most recent order of each customer among the existing customers to the analysis time point; calculate the mean customer level corresponding to the existing customers based on the customer level of each customer among the existing customers; calculate the mean number of orders corresponding to the existing customers based on the number of orders placed by each customer among the existing customers during the statistical period; calculate the mean total order amounts corresponding to the existing customers based on the total order amounts of each customer among the existing customers during the statistical period.

[0020] In some embodiments of the present disclosure, the classification module is further configured to: for each of the customer groups, calculate the mean duration corresponding to the customer group based on the duration from the most recent order of each customer in the customer group to the analysis time point; calculate the mean customer level corresponding to the customer group based on the customer level of each customer in the customer group; calculate the mean number of orders corresponding to the customer group based on the number of orders placed by each customer in the customer group during the statistical period; calculate the mean total order amount corresponding to the customer group based on the total order amount of each customer in the customer group during the statistical period.

[0021] In some embodiments of the present disclosure, the classification module is further configured as follows: if the average duration corresponding to the customer group is greater than the average duration corresponding to the existing customers, the customer group is marked as a first label on the time dimension indicator, otherwise, the customer group is marked as a second label on the time dimension indicator; if the average customer level corresponding to the customer group is greater than the average customer level corresponding to the existing customers, the customer group is marked as a first label on the customer level dimension indicator, otherwise, the customer group is marked as a second label on the customer level dimension indicator; if the average number of orders corresponding to the customer group is greater than the average number of orders corresponding to the existing customers, the customer group is marked as a first label on the order number dimension indicator, otherwise, the customer group is marked as a second label on the order number dimension indicator; if the average total order amount corresponding to the customer group is greater than the average total order amount corresponding to the existing customers, the customer group is marked as a first label on the order total amount dimension indicator, otherwise, the customer group is marked as a second label on the order total amount dimension indicator.

[0022] In some embodiments of the present disclosure, the classification module is further configured to: perform comparative analysis on the labels of each of the multiple customer groups on different dimensional indicators; if the labels of two or more customer groups on different dimensional indicators are the same, merge the two or more customer groups to obtain a merged result; and obtain the customer classification result based on the merged result.

[0023] According to another aspect of the present disclosure, there is provided an electronic device, comprising: one or more processors; and a storage device configured to store one or more programs, so that when the one or more programs are executed by the one or more processors, the one or more processors implement the data processing method as described above.

[0024] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned data processing method is implemented.

[0025] According to yet another aspect of the present disclosure, a computer program product is provided, including: a computer program or instructions, wherein the computer program or instructions implement the above data processing method when executed by a processor.

[0026] The data processing method provided by the embodiment of the present disclosure first obtains a customer data set, which covers the indicator data of existing customers, providing a solid data foundation for subsequent analysis; next, based on the self-organizing mapping network, cluster analysis is performed based on the collected indicator data of existing customers, which can mine the inherent structure and pattern in the data, classify similar customers into one category, and form multiple customer groups with similar characteristics. This process not only considers the direct similarity between customers, but also captures the potential complex patterns in the data to ensure the rationality of the clustering results; then, the indicator data of customers in each customer group is further used to classify these groups in more detail, which can deeply analyze the characteristics of each group and obtain more accurate customer classification results. It can be seen that this method can fully mine the information of the customer data set, obtain accurate customer classification results, improve the accuracy of customer segmentation, and enable enterprises and service providers to accurately market and optimize resource allocation based on customer classification results, thereby matching the needs of different customer groups.

[0027] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The drawings herein are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the description are used to explain the principles of the present disclosure, and do not constitute improper limitations on the present disclosure.

[0029] Figure 1 A schematic diagram of an exemplary application system architecture to which the data processing method in the embodiment of the present disclosure can be applied is shown;

[0030] Figure 2 A flow chart of a data processing method in an embodiment of the present disclosure is shown;

[0031] Figure 3 A flowchart of obtaining a customer data set in a data processing method according to an embodiment of the present disclosure is shown;

[0032] Figure 4 A flow chart showing clustering based on a SOM network to obtain multiple customer groups in a data processing method according to an embodiment of the present disclosure is shown;

[0033] Figure 5 A network structure diagram of a SOM network is shown;

[0034] Figure 6 A flow chart showing classification of multiple customer groups to obtain customer classification results in a data processing method according to an embodiment of the present disclosure;

[0035] Figure 7 A schematic diagram of the structure of a data processing device in an embodiment of the present disclosure is shown;

[0036] Figure 8 A structural block diagram of an electronic device in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0037] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0038] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0039] It should be noted that the ordinal numbers such as "first" and "second" mentioned in the embodiments of the present disclosure are used to distinguish multiple objects, and are not used to limit the order, timing, priority or importance of multiple objects, and the descriptions of "first" and "second" do not limit the objects to be different.

[0040] It should be noted that the collection, collection, update, analysis, processing, use, transmission, storage and other aspects of the customer information involved in this disclosure are in compliance with the relevant laws and regulations, are used for legitimate purposes, and do not violate public order and good morals. Necessary measures are taken for customer information to prevent illegal access to customer information data and maintain customer information security, network security and national security.

[0041] Figure 1 FIG. 1 shows an exemplary application system architecture diagram to which the data processing method in the embodiment of the present disclosure can be applied. Figure 1 As shown, the system architecture may include a terminal device 101 , a network 102 and a server 103 .

[0042] The network 102 is a medium for providing a communication link between the terminal device 101 and the server 103, and can be a wired network or a wireless network.

[0043] Optionally, the wireless network or wired network described above uses standard communication technology and / or protocol. The network is typically the Internet, but may also be any network, including but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or any combination of a virtual private network). In some embodiments, technologies and / or formats including Hyper Text Mark-up Language (HTML), Extensible Markup Language (XML), etc. are used to represent data exchanged over the network. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPSec) and the like may be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies may also be used to replace or supplement the above data communication technologies.

[0044] The terminal device 101 can be various electronic devices, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, smart speakers, smart watches, wearable devices, augmented reality devices, virtual reality devices, etc.

[0045] Optionally, the client of the application installed in different terminal devices 101 is the same, or the client of the same type of application based on different operating systems. Based on the different terminal platforms, the specific form of the client of the application can also be different, for example, the application client can be a mobile client, a PC client, etc.

[0046] The server 103 may be a server that provides various services, such as a background management server that provides support for the device operated by the user using the terminal device 101. The background management server may analyze and process the received request and other data, and feed back the processing results to the terminal device.

[0047] Optionally, the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or 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, CDN (Content Delivery Network), as well as big data and artificial intelligence platforms.

[0048] Those skilled in the art will know that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration, and any number of terminal devices, networks and servers may be provided according to actual needs, and the embodiments of the present disclosure do not limit this.

[0049] Under the above system architecture, a data processing method is provided in an embodiment of the present disclosure, and the method can be executed by any electronic device with computing and processing capabilities.

[0050] In some embodiments, the data processing method provided in the embodiments of the present disclosure may be executed by a terminal device of the above-mentioned system architecture; in other embodiments, the data processing method provided in the embodiments of the present disclosure may be executed by a server in the above-mentioned system architecture; in other embodiments, the data processing method provided in the embodiments of the present disclosure may be implemented by a terminal device and a server in the above-mentioned system architecture through interaction.

[0051] Figure 2 A flow chart of a data processing method in an embodiment of the present disclosure is shown as follows: Figure 2 As shown, the data processing method provided in the embodiment of the present disclosure includes the following steps.

[0052] Step S210, obtaining a customer data set; the customer data set includes indicator data of existing customers.

[0053] In the disclosed embodiments, existing customers can be understood as existing customers, that is, customer groups that have established business relationships with an enterprise or service provider and have continued to trade or interact for a certain period of time. In other words, existing customers can be customers who have purchased products or services, or customers who have registered and used a service of an enterprise or service provider.

[0054] In this step, the historical data of existing customers is analyzed to extract the customer's index data. For example, for the telecommunications industry, the historical data of telecommunications customers is collected, and the collected historical data is analyzed to extract the telecommunications customer's index data.

[0055] Step S220 , based on the self-organizing map network, clustering the existing customers according to the indicator data of the existing customers to obtain multiple customer groups.

[0056] Among them, the Self-Organizing Maps (SOM) neural network algorithm is an unsupervised learning algorithm that aims to map high-dimensional input data to low-dimensional space while maintaining the topological structure of the data. The SOM algorithm is an artificial neural network formed by simulating the characteristics of signal processing in the human brain. It is based on a competitive learning strategy, which enables neurons in the network to respond to input patterns through competition. The winning neurons and neurons in their neighborhood adjust their weights according to the input pattern to better fit the input data.

[0057] In the disclosed embodiment, after obtaining the indicator data of existing customers, a cluster analysis is performed on the existing customers based on a self-organizing neural network and in combination with the indicator data of the existing customers to discover the similarities and differences between these customers, and similar customers are assigned to a group to form multiple customer groups.

[0058] For example, in the telecommunications industry, after extracting the indicator data of telecommunications customers, cluster analysis is performed on telecommunications customers based on a self-organizing neural network and combined with the indicator data of telecommunications customers to discover the similarities and differences between telecommunications customers, and similar customers are assigned to a group to form multiple telecommunications customer groups.

[0059] Step S230 , classifying multiple customer groups according to the indicator data of the existing customers and the indicator data of the customers included in each customer group, to obtain a customer classification result.

[0060] In the disclosed embodiment, based on the self-organizing neural network clustering, the indicator data of the existing customers and the indicator data of the customers in each customer group are further utilized to classify these customer groups in more detail to obtain more accurate and specific customer classification results.

[0061] For example, in the telecommunications industry, the indicator data of telecommunications customers and the indicator data of customers in each telecommunications customer group are further used to classify these customer groups in more detail to obtain more accurate and specific telecommunications customer classification results. Based on this, different telecommunications products or services are provided to different telecommunications customers according to the telecommunications customer classification results.

[0062] The data processing method provided by the embodiment of the present disclosure first obtains a customer data set, which covers the indicator data of existing customers, providing a solid data foundation for subsequent analysis; next, based on the SOM network, cluster analysis is performed according to the collected indicator data of existing customers, which can mine the inherent structure and pattern in the data, classify similar customers into one category, and form multiple customer groups with similar characteristics. This process not only considers the direct similarity between customers, but also captures the potential complex patterns in the data to ensure the rationality of the clustering results; then, the indicator data of customers in each customer group is further used to classify these groups in more detail, which can deeply analyze the characteristics of each group and obtain more accurate customer classification results. It can be seen that this method can fully mine the information of the customer data set, obtain accurate customer classification results, and improve the accuracy of customer segmentation, so that enterprises and service providers can accurately market and optimize resource allocation based on customer classification results, thereby matching the needs of different customer groups.

[0063] Figure 3 FIG. 1 is a flowchart of obtaining a customer data set in a data processing method according to an embodiment of the present disclosure. Figure 3 As shown, follow the steps below to obtain the customer dataset.

[0064] Step S310, obtaining historical data of existing customers.

[0065] In the disclosed embodiment, historical data of existing customers are collected. Such historical data may include personal information of customers (such as name, age, gender, contact information, etc.), transaction records (such as purchase history, purchase frequency, consumption amount, etc.), behavior data (such as browsing history, click behavior, search keywords, etc.) and other customer-related data. For example, for the telecommunications industry, historical data of telecommunications customers is collected from the telecommunications customer database.

[0066] Step S320, preprocessing the historical data of existing customers to obtain preprocessed data.

[0067] In the disclosed embodiment, after obtaining the historical data of the existing customers, the historical data are preprocessed to obtain the preprocessed data.

[0068] Preprocessing includes data cleaning operations, such as checking and deleting duplicate records in historical data to ensure that each customer is counted only once; or identifying and correcting obvious errors in historical data, such as incorrect date formats and unreasonable values.

[0069] In an exemplary embodiment of the present disclosure, the preprocessing includes normalization processing. After obtaining the historical data of the existing customers, the historical data are normalized to obtain the normalized data.

[0070] The normalization formula is as follows:

[0071]

[0072] In formula (1), X s is the normalized data, X i is the actual value in historical data, n is the number of existing customers, X min =min(X i ), X max =max(X i ).

[0073] Specifically, identify the features that need to be normalized from the historical data of existing customers, such as purchase frequency and consumption amount. For each feature, calculate its minimum and maximum values ​​in the historical data, and use the normalization formula to convert the actual value of each feature in the historical data into a normalized value. Combine the normalized values ​​of all features to form normalized data, that is, preprocessed data.

[0074] Step S330: extracting index data of existing customers from the preprocessed data based on a predefined customer relationship management model.

[0075] Among them, the customer relationship management model includes the RFM model, which determines the customer value through three indicators: R (Recency, the customer's most recent purchase time), F (Frequency, the customer's purchase frequency within a certain period of time), and M (Monetary, the total amount of consumption by the customer in a certain period of time).

[0076] In the disclosed embodiment, the predefined customer relationship management model is an improved RFM model, which can also be called a TLFM model, which is constructed based on four indicators: the time from the customer's last order to the analysis time point (T), the customer's membership level (L), the number of orders placed by the customer during the statistical period (F), and the total amount of orders placed by the customer during the statistical period (M). The statistical period can be set according to actual needs or experience, such as 1 month.

[0077] In the disclosed embodiment, based on the TLFM model, the index data of the existing customers are extracted from the preprocessed data. The index data of the existing customers include the time from the customer's most recent order to the analysis time point, the customer's customer level, the number of orders of the customer in the statistical period, and the total amount of the customer's orders in the statistical period.

[0078] For example, for the telecommunications industry, based on the TLFM model, the length of time from the last payment of the telecommunications customer to the analysis time point, the customer membership level of the telecommunications customer, the number of payments made by the telecommunications customer during the statistical period, and the total amount paid by the telecommunications customer during the statistical period are extracted from the pre-processed telecommunications customer data.

[0079] In the data processing method of the disclosed embodiment, after obtaining the historical data of existing customers, the obtained historical data is preprocessed, including data cleaning and normalization processing, which can remove duplicate records and correct erroneous data, and the normalization processing helps to eliminate the dimensional differences between different data features, and can be compared and analyzed on the same scale. And, after data preprocessing, the indicator data of existing customers is extracted based on a predefined customer relationship management model (i.e., TLFM model). Compared with the RFM model, not only the customer's purchase time (T), number of orders (F) and order amount (M) are considered, but also the dimension of customer membership level (L) is introduced, so that the extracted indicator data is more comprehensive and detailed, and the in-depth mining and analysis of the existing customer data is realized, providing strong support for the subsequent classification of existing customers.

[0080] Figure 4 The flowchart of clustering based on the SOM network to obtain multiple customer groups in the data processing method of the embodiment of the present disclosure is shown. Figure 4 As shown, the specific steps include the following steps.

[0081] Step S410: input the index data of existing customers into the SOM network for training to obtain a trained SOM network.

[0082] SOM network is an unsupervised learning algorithm suitable for clustering analysis of unlabeled or unknown clustering structure data sets. Figure 5 Figure 2 shows the network structure diagram of the SOM network. Figure 5 As shown in the figure, the SOM network consists of two parts: the input layer and the output layer. The number of neurons in the input layer corresponds to the feature dimension of the input vector. The output layer is also called the competition layer or mapping layer, which is responsible for the clustering and mapping of the data. The number of neurons in the output layer can be determined according to the needs of the actual application and the desired clustering precision.

[0083] In the SOM network, the neurons between the input layer and the output layer are connected by the weight vector W. jiThese weights are iteratively updated during training to reflect the intrinsic structure and clustering patterns of the input data. In addition, there may be local connections between neurons in the output layer, which ensure that neighboring neurons can work together when responding to similar inputs.

[0084] The principle of SOM network algorithm is: In SOM network, the input layer neurons and the competition layer (output layer) neurons are connected by weight vector W. ji These weight vectors are continuously updated during the training process to reflect the intrinsic structure and clustering pattern of the input data; there may be local connections between neurons in the competition layer, and these local connections can ensure that adjacent neurons can work together when responding to similar inputs, thereby enhancing the network's topology preservation ability; for each input pattern, the SOM network calculates the Euclidean distance between the input pattern and the weight vector of each neuron in the competition layer, and finds the neuron with the smallest distance as the winning neuron; according to the Kohonen learning rule, the weights of the winning neuron and the neurons in a small neighborhood around it are corrected.

[0085] Training results of the SOM network: Through continuous weight adjustment, the SOM network will form an internal representation of the input space on the competition layer. This representation not only reflects the density distribution of the original space samples, but also maintains the topological relationship between the original space samples. Similar samples in the input space will be mapped to adjacent neurons on the competition layer, thus forming clusters.

[0086] In the disclosed embodiment, the SOM network may be trained according to the following steps to obtain a trained SOM network.

[0087] (1) Network topology and initialization

[0088] Select a SOM network topology, such as a hexagonal SOM network, and assign small random initial values ​​to the initial weights to ensure that the network has diversity at the beginning of training. Set a relatively large initial neighborhood N to cover more neurons and promote global learning. Set the number of iterations to 500 as the total number of steps for network training.

[0089] (2) Input Mode and Network Learning

[0090] For each input pattern X s ={X s1 ,X s1 ,…,X sn}, where n represents the dimension of the input data and s is the serial number of the input pattern, which is input into the network to let the network learn the input pattern. Calculate the input pattern X sThe distance from all neurons can usually be calculated using the Euclidean distance, which is calculated as follows:

[0091]

[0092] In formula (2), d j For input mode X s The distance to the jth neuron, W ji is the weight of the i-th dimension of the j-th neuron. According to formula (2), we can select s The neuron i with the smallest distance * as the winning neuron.

[0093] (3) Weight adjustment

[0094] For the winning neuron i * and the nodes in its neighborhood, adjusting neuron i according to a certain learning rate * The connection weight calculation of its neighboring nodes is as follows:

[0095] W ji (t+1)=W ji (t)+η(t)(X s -W ji (t)) (3)

[0096] In formula (3), W ji (t+1) is the updated weight, W ji (t) is the weight before updating, is the learning rate, η0 is the initial learning rate, τ1 is the time constant, and k is the number of iterations.

[0097] (4) Neighborhood function calculation and update

[0098] The neighborhood function is used to determine the range of neurons whose weights need to be adjusted in each iteration. The neighborhood function usually decreases as the number of iterations increases to gradually focus on the winning neurons. The neighborhood function formula is:

[0099]

[0100]

[0101] in, is the winning neuron i * The area around with radius d shrinks with time, and q is Other neurons in the *| is the distance between the neurons in the neighborhood and the winning neuron, which decreases with time. σ0 is the initial value of σ, k is the number of iterations, and τ2 is the time constant. After each iteration, the neighborhood range of the current winning neuron is updated according to the neighborhood function.

[0102] (5) Iteration and training completed

[0103] Continue to input new patterns and return to step (2) until all patterns are input into the network. When the number of network training steps reaches the maximum number of iterations, the trained SOM network is obtained.

[0104] A set of test data sets is input into the trained SOM network to evaluate the clustering effect of the network. If the clustering effect of the network is very poor, it may be because the network is over-adapted to the training data set. At this time, you can return to step (1) and reset the training parameters, such as reducing the initial neighborhood size, increasing the number of iterations, or adjusting the learning rate. If the clustering effect of the network is good, it means that the network has completed the training and can be used for subsequent tasks or applications.

[0105] Step S420: clustering the existing customers based on the trained SOM network to obtain multiple customer groups.

[0106] After obtaining the trained SOM network, the indicator data of the existing customers are input into the trained SOM network, the existing customers are clustered, and multiple customer groups are obtained.

[0107] In the data processing method provided by the embodiment of the present disclosure, the indicator data of existing customers are input into the SOM network for training, and the existing customers are clustered based on the trained network. Compared with the existing clustering classification method, the SOM network can process large amounts of data more quickly and accurately, improve the automation and efficiency of data processing, and can quickly and accurately mine the characteristics of customer groups from massive data without human intervention, thereby realizing in-depth analysis of customer information and making the obtained customer groups more accurate.

[0108] Figure 6 FIG. 1 is a flow chart showing a method for classifying multiple customer groups to obtain customer classification results in a data processing method according to an embodiment of the present disclosure. Figure 6 As shown, the specific steps include the following steps.

[0109] Step S610, determining the mean of the indicator data of the existing customers according to the indicator data of the existing customers; wherein the mean of the indicator data of the existing customers includes the mean of the duration, the mean of the customer level, the mean of the number of orders and the mean of the total amount of the orders corresponding to the existing customers.

[0110] In the disclosed embodiment, the mean of various indicators (such as duration, customer level, number of orders, and total order amount) of existing customers is calculated, and the overall characteristics of the existing customers are represented by the mean.

[0111] In some embodiments of the present disclosure, the mean of the indicator data of existing customers is determined based on the indicator data of existing customers, including: calculating the mean of the duration corresponding to the existing customers based on the duration from the most recent order of each customer among the existing customers to the analysis time point; calculating the mean of the customer level corresponding to the existing customers based on the customer level of each customer among the existing customers; calculating the mean of the number of orders corresponding to the existing customers based on the number of orders placed by each customer among the existing customers during the statistical period; calculating the mean of the total amount of orders corresponding to the existing customers based on the total amount of orders placed by each customer among the existing customers during the statistical period.

[0112] For the indicator of customer purchase time (T), obtain the duration from the most recent order of each customer among the existing customers to the analysis time point, and then calculate the mean of all the durations to obtain the corresponding mean duration of the existing customers.

[0113] For the indicator of customer membership level (L), the customer level of each customer among the existing customers is obtained, and then the mean of all customer levels is calculated to obtain the mean customer level corresponding to the existing customers.

[0114] For the indicator of number of orders (F), the number of orders for each existing customer during the statistical period is obtained, and then the average number of orders for all customers is calculated to obtain the average number of orders corresponding to the existing customers.

[0115] For the indicator of order amount (M), the total order amount of each existing customer during the statistical period is obtained, and then the total order amount of all customers is averaged to obtain the average total order amount corresponding to the existing customers.

[0116] Step S620, for each customer group, determine the mean of the indicator data of the customer group according to the indicator data of the customers in the customer group; wherein the mean of the indicator data of the customer group includes the mean of the duration, the mean of the customer level, the mean of the number of orders and the mean of the total amount of orders corresponding to the customer group.

[0117] In the disclosed embodiment, for each customer group, the mean of each indicator (such as duration, customer level, number of orders, and total order amount) of the customer group is calculated, and the overall characteristics of the customer group are represented by the mean.

[0118] In some embodiments of the present disclosure, for each customer group, the mean of the indicator data of the customer group is determined based on the indicator data of the customers in the customer group, including: for each customer group, according to the duration of the distance from the most recent order of each customer in the customer group to the analysis time point, calculating the mean of the duration corresponding to the customer group; according to the customer level of each customer in the customer group, calculating the mean of the customer level corresponding to the customer group; according to the number of orders of each customer in the customer group during the statistical period, calculating the mean of the number of orders corresponding to the customer group; according to the total amount of orders of each customer in the customer group during the statistical period, calculating the mean of the total amount of orders corresponding to the customer group.

[0119] For each customer group, calculate the mean of the customer group's purchase time (T), customer membership level (L), number of orders (F), and order amount (M). The specific implementation is:

[0120] For the dimension of customer purchase time (T), the duration from the most recent order of each customer in the customer group to the analysis time point is obtained, and then the mean of all the durations in the customer group is calculated to obtain the mean duration corresponding to the customer group; for the dimension of customer membership level (L), the customer level of each customer in the customer group is obtained, and then the mean of all customer levels in the customer group is calculated to obtain the mean customer level corresponding to the customer group; for the dimension of number of orders (F), the number of orders of each customer in the customer group during the statistical period is obtained, and then the mean of the number of orders of all customers in the customer group is calculated to obtain the mean number of orders corresponding to the customer group; for the dimension of order amount (M), the total order amount of each customer in the customer group during the statistical period is obtained, and then the total order amount of all customers in the customer group is calculated to obtain the mean total order amount corresponding to the customer group.

[0121] In an exemplary embodiment of the present disclosure, when calculating the mean of each indicator for the customer group, the weight values ​​of different indicators can be introduced. For example, for the telecommunications industry, the weight of the RFM indicator is obtained by the Analytic Hierarchy Process (AHP) method, and the weight value is [W1, W2, W3] = [0.221, 0.341, 0.439], and the weight value for the TLFM indicator of the present disclosure can be set according to the weight value.

[0122] Step S630, comparing the mean value of the indicator data of the existing customers with the mean value of the indicator data of the customer group to obtain a comparison result corresponding to the customer group.

[0123] In the disclosed embodiment, for each customer group, the mean value of the indicator data of the existing customers is compared with the mean value of the indicator data of the customer group to obtain a comparison result corresponding to the customer group. Based on this, the comparison result reflects which indicators of the customer group are higher or lower than the overall level of the existing customers.

[0124] In some embodiments of the present disclosure, the mean indicator data of existing customers and the mean indicator data of customer groups are compared to obtain comparison results corresponding to the customer groups, including: if the mean duration corresponding to the customer group is greater than the mean duration corresponding to the existing customers, the customer group is marked as a first label on the time dimension indicator, otherwise, the customer group is marked as a second label on the time dimension indicator; if the mean customer level corresponding to the customer group is greater than the mean customer level corresponding to the existing customers, the customer group is marked as a first label on the customer level dimension indicator, otherwise, the customer group is marked as a second label on the customer level dimension indicator; if the mean number of orders corresponding to the customer group is greater than the mean number of orders corresponding to the existing customers, the customer group is marked as a first label on the order number dimension indicator, otherwise, the customer group is marked as a second label on the order number dimension indicator; if the mean total order amount corresponding to the customer group is greater than the mean total order amount corresponding to the existing customers, the customer group is marked as a first label on the order total amount dimension indicator, otherwise, the customer group is marked as a second label on the order total amount dimension indicator.

[0125] For this customer group, if the mean duration of this customer group is greater than the mean duration of existing customers, it is marked as the first label on this dimension, such as with a + mark; otherwise, it is marked as the second label, such as with a - mark. If the mean customer level of the customer group is greater than the mean customer level of existing customers, it is marked as the first label on this dimension, such as with a + mark; otherwise, it is marked as the second label, such as with a - mark. If the mean number of orders of the customer group is greater than the mean number of orders of existing customers, it is marked as the first label on this dimension, such as with a + mark; otherwise, it is marked as the second label, such as with a - mark. If the mean total amount of orders of the customer group is greater than the mean total amount of orders of existing customers, it is marked as the first label on this dimension, such as with a + mark; otherwise, it is marked as the second label, such as with a - mark.

[0126] Table 1 is a schematic diagram of the comparison results corresponding to the customer groups. As shown in Table 1, based on the SOM network, cluster analysis is performed on the existing customers to obtain 9 customer groups. The mean value of the indicator data of each customer group is compared with the mean value of the indicator data of the customer group to obtain the comparison results corresponding to the 9 customer groups.

[0127] Table 1 Comparison results of customer groups

[0128] Group Number T L F M Comparison results 1 0.7543 1.2589 -0.5098 0.2532 ++-+ 2 0.5015 -0.7567 1.2531 -0.5078 +-+- 3 1.2598 -0.5063 0.2547 -0.7512 +-+- 4 -0.5089 1.2532 -0.7515 0.2598 -+-+ 5 0.2567 -0.5062 1.2543 -0.7531 +-+- 6 1.2515 -0.7598 0.2567 -0.5045 +-+- 7 -0.7531 -0.6515 -0.5098 1.2567 ---+ 8 0.5043 -0.7512 1.2531 -0.2598 +-++ 9 1.2567 -0.5035 0.2515 -0.7543 +-+- Overall mean 0.2567 -0.5031 0.2515 -0.5043 null

[0129] Step S640: divide the multiple customer groups according to the comparison result corresponding to each customer group to obtain a customer classification result.

[0130] In the disclosed embodiment, customer groups can be divided according to the comparison results. For example, customer groups that are higher than the average of existing customers in multiple indicators can be divided into one category, indicating that the average values ​​of these customer groups in each indicator are very high; similarly, customer groups that are lower than the average of existing customers in multiple indicators can also be divided into another category.

[0131] In some embodiments of the present disclosure, multiple customer groups are classified according to the comparison results corresponding to each customer group to obtain customer classification results, including: performing comparative analysis on the labels of each customer group in the multiple customer groups on different dimensional indicators; if the labels of two or more customer groups on different dimensional indicators are the same, merging the two or more customer groups to obtain a merged result; and obtaining a customer classification result based on the merged result.

[0132] After obtaining the labels of each customer group on different dimensional indicators, the customer groups with the same labels on each dimensional indicator are grouped together and merged to finally obtain the customer classification result, wherein the customer classification result includes one or more customer groups.

[0133] Taking Table 1 as an example, the comparison results of the 2nd, 3rd, 5th, 6th and 9th customer groups are the same. These 5 customer groups are classified into one category, and finally 5 customer groups are obtained.

[0134] In the disclosed embodiment, after obtaining the customer classification results, that is, determining the customer groups included in the existing customers, the weighted value of each customer group is calculated according to the weights of each indicator corresponding to TLFM, and the weighted value is used as the value of the customer group. Then, according to the value of the customer value, the value of each customer group is arranged.

[0135] Table 2 shows the information of various customer groups corresponding to existing customers.

[0136] Customer Classification T L F M Number of clustered customers Category A 0.4052 2.2999 2.9989 2.8827 263 Category B 0.7895 1.5583 0.6825 -0.6348 737 Category C -0.7175 0.1393 0.4253 0.6716 302 Category D 0.8696 -0.4207 -0.3927 -0.4035 1703 Category E -0.9631 -0.5396 -0.4990 -0.4934 1510

[0137] Table 2 shows the information of various customer groups corresponding to existing customers. As shown in Table 2, existing customers can be divided into five customer groups: A, B, C, D and E according to the mean values ​​of their various indicators. Among the various customer groups, the A customer group accounts for 5.8%, and the B, C, and D customer groups account for 16.3%, 6.4%, 37.6% and 33.4% respectively. In addition, according to the weights of each indicator corresponding to TLFM, the weighted values ​​of the five customer groups A, B, C, D and E are calculated. Finally, according to the calculated weighted values, the weighted value of the A customer group is the highest, and the A customer group is considered to be the core customer, and the B, C, D and E customer groups are considered to be high-value customers, secondary-value customers, important customers and general customers respectively.

[0138] In summary, the data processing method provided by the embodiment of the present disclosure first obtains a customer data set, which covers the indicator data of existing customers, providing a solid data foundation for subsequent analysis; next, based on the SOM network, cluster analysis is performed according to the collected indicator data of existing customers, which can mine the inherent structure and pattern in the data, classify similar customers into one category, and form multiple customer groups with similar characteristics. This process not only considers the direct similarity between customers, but also captures the potential complex patterns in the data to ensure the rationality of the clustering results; then, the indicator data of customers in each customer group is further used to classify these groups more finely, which can deeply analyze the characteristics of each group and obtain more accurate customer classification results. It can be seen that this method can fully mine the information of the customer data set, obtain accurate customer classification results, improve the accuracy of customer segmentation, and enable enterprises and service providers to accurately market and optimize resource allocation based on customer classification results, so as to match the needs of different customer groups.

[0139] Figure 7 FIG. 2 shows a schematic diagram of the structure of a data processing device in an embodiment of the present disclosure. Figure 7 As shown, the device 700 includes: a data acquisition module 710, a clustering module 720 and a classification module 730.

[0140] The data acquisition module 710 is configured to: acquire a customer data set; the customer data set includes index data of existing customers. The clustering module 720 is configured to: cluster existing customers according to the index data of existing customers based on the SOM network to obtain multiple customer groups. The classification module 730 is configured to: classify multiple customer groups according to the index data of existing customers and the index data of customers included in each customer group to obtain customer classification results.

[0141] In some embodiments of the present disclosure, the data acquisition module 710 is further configured to: acquire historical data of existing customers; preprocess the historical data of existing customers to obtain preprocessed data; and extract indicator data of existing customers from the preprocessed data based on a predefined customer relationship management model; wherein the indicator data of existing customers includes the length of time from the customer's most recent order to the analysis time point, the customer's customer level, the number of orders placed by the customer during the statistical period, and the total amount of the customer's orders during the statistical period.

[0142] In some embodiments of the present disclosure, the clustering module 720 is further configured to: input the indicator data of existing customers into the SOM network for training to obtain a trained SOM network; and cluster the existing customers based on the trained SOM network to obtain multiple customer groups.

[0143] In some embodiments of the present disclosure, the classification module 730 is further configured to: determine the mean of the indicator data of existing customers based on the indicator data of existing customers; the mean of the indicator data of existing customers includes the mean duration, mean customer level, mean number of orders and mean total order amount corresponding to the existing customers; for each customer group, determine the mean of the indicator data of the customer group based on the indicator data of the customers in the customer group; the mean of the indicator data of the customer group includes the mean duration, mean customer level, mean number of orders and mean total order amount corresponding to the customer group; compare the mean of the indicator data of the existing customers with the mean of the indicator data of the customer group to obtain a comparison result corresponding to the customer group; divide multiple customer groups according to the comparison result corresponding to each customer group to obtain a customer classification result.

[0144] In some embodiments of the present disclosure, the classification module 730 is further configured to: calculate the mean duration corresponding to the existing customers based on the duration from the most recent order of each customer among the existing customers to the analysis time point; calculate the mean customer level corresponding to the existing customers based on the customer level of each customer among the existing customers; calculate the mean number of orders corresponding to the existing customers based on the number of orders placed by each customer among the existing customers during the statistical period; calculate the mean total order amount corresponding to the existing customers based on the total order amount of each customer among the existing customers during the statistical period.

[0145] In some embodiments of the present disclosure, the classification module 730 is further configured to: for each customer group, calculate the mean duration corresponding to the customer group based on the duration from the most recent order of each customer in the customer group to the analysis time point; calculate the mean customer level corresponding to the customer group based on the customer level of each customer in the customer group; calculate the mean number of orders corresponding to the customer group based on the number of orders placed by each customer in the customer group during the statistical period; calculate the mean total order amount corresponding to the customer group based on the total order amount of each customer in the customer group during the statistical period.

[0146] In some embodiments of the present disclosure, the classification module 730 is further configured as follows: if the average duration corresponding to a customer group is greater than the average duration corresponding to existing customers, the customer group is marked as a first label on the time dimension indicator; otherwise, the customer group is marked as a second label on the time dimension indicator; if the average customer level corresponding to the customer group is greater than the average customer level corresponding to existing customers, the customer group is marked as a first label on the customer level dimension indicator; otherwise, the customer group is marked as a second label on the customer level dimension indicator; if the average number of orders corresponding to the customer group is greater than the average number of orders corresponding to existing customers, the customer group is marked as a first label on the order number dimension indicator; otherwise, the customer group is marked as a second label on the order number dimension indicator; if the average total order amount corresponding to the customer group is greater than the average total order amount corresponding to existing customers, the customer group is marked as a first label on the order total amount dimension indicator; otherwise, the customer group is marked as a second label on the order total amount dimension indicator.

[0147] In some embodiments of the present disclosure, the classification module 730 is further configured to: perform comparative analysis on the labels of each customer group in multiple customer groups on different dimensional indicators; if the labels of two or more customer groups on different dimensional indicators are the same, merge the two or more customer groups to obtain a merged result; and obtain a customer classification result based on the merged result.

[0148] Since the principle of solving the problem in the data processing device embodiment is similar to that in the above method embodiment, the implementation of the data processing device embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be repeated.

[0149] Figure 8 FIG. 1 shows a structural block diagram of an electronic device in an embodiment of the present disclosure. It should be noted that: Figure 8 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0150] like Figure 8As shown, the electronic device 800 includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage part 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the electronic device 800 are also stored. The CPU 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0151] The following components are connected to the I / O interface 805: an input section 806 including a keyboard, a mouse, etc.; an output section 807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN card, a modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I / O interface 805 as needed. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 810 as needed, so that a computer program read therefrom is installed into the storage section 808 as needed.

[0152] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part 809, and / or installed from a removable medium 811. When the computer program is executed by a central processing unit (CPU) 801, the above-mentioned functions defined in the system of the present disclosure are executed.

[0153] It should be noted that the computer-readable medium shown in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, terminal device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, a terminal device or a device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, a terminal device or a device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0154] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0155] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware. The units described may also be set in a processor, for example, it may be described as: a processor includes a data acquisition module, a clustering module and a classification module. The names of these modules do not constitute a limitation on the modules themselves in some cases, for example, the data acquisition module may also be described as a "module for acquiring a customer data set".

[0156] As another aspect, the present disclosure further provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments; or may exist independently without being assembled into the electronic device. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed by an electronic device, the electronic device implements the method described in the following embodiments. For example, the electronic device may implement the following Figure 2 The steps shown.

[0157] According to one aspect of the present disclosure, a computer program product or a computer program is provided, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in various optional implementations of the above-mentioned embodiments.

[0158] It should be understood that any number of elements in the drawings of the present disclosure is for illustration rather than limitation, and any naming is only for distinction rather than having any limiting meaning.

[0159] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0160] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A data processing method, characterized in that: The method comprises: Acquire a customer data set; the customer data set includes indicator data of existing customers; Based on a self-organizing map network, clustering the existing customers according to the indicator data of the existing customers to obtain multiple customer groups; The multiple customer groups are classified according to the indicator data of the existing customers and the indicator data of the customers included in each of the customer groups to obtain a customer classification result.

2. The method according to claim 1, characterized in that The obtaining of the customer data set includes: Obtaining historical data of the existing customers; Preprocessing the historical data of the existing customers to obtain preprocessed data; Based on a predefined customer relationship management model, extracting indicator data of the existing customers from the preprocessed data; Among them, the indicator data of the existing customers include the length of time from the customer's most recent order to the analysis time point, the customer's customer level, the number of orders placed by the customer during the statistical period, and the total amount of the customer's orders during the statistical period.

3. The method according to claim 1, characterized in that The method of clustering the existing customers based on the self-organizing map network according to the indicator data of the existing customers to obtain multiple customer groups includes: Inputting the indicator data of the existing customers into the self-organizing map network for training to obtain a trained self-organizing map network; Based on the trained self-organizing map network, the existing customers are clustered to obtain the multiple customer groups.

4. The method according to claim 1, characterized in that The step of dividing the plurality of customer groups according to the indicator data of the existing customers and the indicator data of the customers included in each of the customer groups to obtain customer classification results includes: Determine the mean of the indicator data of the existing customers according to the indicator data of the existing customers; the mean of the indicator data of the existing customers includes the mean of the duration, the mean of the customer level, the mean of the number of orders and the mean of the total amount of the orders corresponding to the existing customers; For each of the customer groups, determine the mean of the indicator data of the customer group according to the indicator data of the customers in the customer group; the mean of the indicator data of the customer group includes the mean of the duration, the mean of the customer level, the mean of the number of orders and the mean of the total amount of orders corresponding to the customer group; Compare the mean value of the indicator data of the existing customers with the mean value of the indicator data of the customer group to obtain a comparison result corresponding to the customer group; The multiple customer groups are divided according to the comparison result corresponding to each of the customer groups to obtain the customer classification result.

5. The method according to claim 4, characterized in that Determining the mean value of the indicator data of the existing customers according to the indicator data of the existing customers includes: Calculate the average duration corresponding to each of the existing customers based on the duration between the most recent order of each customer and the analysis time point; Calculating the average of the customer levels of each customer among the existing customers according to the customer level of each customer; Calculate the average number of orders for each of the existing customers according to the number of orders placed by each customer during the statistical period; According to the total order amount of each customer among the existing customers during the statistical period, the average total order amount corresponding to the existing customers is calculated.

6. The method according to claim 4, characterized in that For each of the customer groups, determining the mean value of the indicator data of the customer group according to the indicator data of the customers in the customer group includes: For each of the customer groups, according to the time between the most recent order of each customer in the customer group and the analysis time point, calculate the average time corresponding to the customer group; Calculating a customer level mean corresponding to the customer group according to the customer level of each customer in the customer group; Calculate the average number of orders for each customer in the customer group according to the number of orders placed during the statistical period; According to the total order amount of each customer in the customer group during the statistical period, the average total order amount corresponding to the customer group is calculated.

7. The method according to claim 4, characterized in that The comparing the mean value of the indicator data of the existing customers with the mean value of the indicator data of the customer group to obtain the comparison result corresponding to the customer group includes: If the average duration corresponding to the customer group is greater than the average duration corresponding to the existing customers, the customer group is marked as the first label on the time dimension indicator; otherwise, the customer group is marked as the second label on the time dimension indicator; If the customer level mean value corresponding to the customer group is greater than the customer level mean value corresponding to the existing customers, the customer group is marked as a first label on the customer level dimension indicator; otherwise, the customer group is marked as a second label on the customer level dimension indicator; If the average number of orders corresponding to the customer group is greater than the average number of orders corresponding to the existing customers, the customer group is marked as a first label on the order number dimension indicator; otherwise, the customer group is marked as a second label on the order number dimension indicator; If the average total order amount corresponding to the customer group is greater than the average total order amount corresponding to the existing customers, the customer group is marked as the first label on the total order amount dimension indicator; otherwise, the customer group is marked as the second label on the total order amount dimension indicator.

8. The method according to claim 7, characterized in that The step of classifying the plurality of customer groups according to the comparison result corresponding to each of the customer groups to obtain the customer classification result includes: Comparing and analyzing the labels of each of the multiple customer groups on different dimensional indicators; If the labels of two or more customer groups on different dimensional indicators are the same, the two or more customer groups are merged to obtain a merged result; Based on the merging result, the customer classification result is obtained.

9. A data processing device, characterized in that: The device comprises: A data acquisition module is configured to acquire a customer data set; the customer data set includes indicator data of existing customers; A clustering module is configured to cluster the existing customers according to the indicator data of the existing customers based on a self-organizing map network to obtain a plurality of customer groups; The classification module is configured to classify the multiple customer groups according to the indicator data of the existing customers and the indicator data of the customers included in each of the customer groups to obtain a customer classification result.

10. An electronic device, characterized in that: include: one or more processors; A storage device configured to store one or more programs, when the one or more programs are executed by the one or more processors, enables the one or more processors to implement the data processing method according to any one of claims 1 to 8.

11. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the data processing method according to any one of claims 1 to 8 is implemented.

12. A computer program product comprising: A computer program or instruction, characterized in that when the computer program or instruction is executed by a processor, it implements the data processing method according to any one of claims 1 to 8.

Images