Stock order following customer loss analysis method and system and computer readable storage medium
Through big data and intelligent analysis technology, machine learning models are used to predict the probability of customer churn in the financial market, solving the problem of time-consuming and labor-consuming traditional analysis methods, achieving efficient and accurate customer churn identification and personalized intervention, and enhancing customers Management effect.
Patent Information
- Application Number
- CN202411020625.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2024-11-15
AI Technical Summary
The existing technology in the financial markets customer churn analysis relies on traditional data statistics and manual analysis, which is time-consuming and labor-intensive, and it is difficult to respond to market changes and customer needs in real time, and it is impossible to fully and accurately identify potential churn customers.
Using big data and intelligent analysis methods, we use machine learning models to predict customer churn probability by monitoring customer behavior characteristics data, including decision tree models and other machine learning models, analyze customer transaction behavior in real time, generate churn curves, and automatically identify churn customers. And formulate personalized intervention strategies.
It greatly reduces manual intervention, improves the efficiency and accuracy of customer churn identification, can quickly identify potential lost customers, provide personalized retention strategies, and improves the accuracy and satisfaction of customer management.
Smart Images

Figure CN118967192A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the field of financial technology, and specifically to a method for analyzing the churn of stock copycat customers, a system for analyzing the churn of stock copycat customers, a computer-readable storage medium, and a computer program product. Background Art
[0002] With the development of financial markets and the diversification of investor behavior, copy trading has become an important investment method. Current customer churn analysis mostly relies on traditional data statistics and manual analysis methods, which is not only time-consuming and labor-intensive, but also difficult to respond to market changes and diversified customer needs in real time. Summary of the invention
[0003] Embodiments of the present application provide a stock copycat customer churn analysis method, a stock copycat customer churn analysis system, an electronic device, a chip, a computer-readable storage medium, and a computer program product.
[0004] The method for analyzing stock copycat customer churn provided in the embodiment of the present application includes:
[0005] Monitor customer behavior data;
[0006] Based on the behavioral characteristic data, the customer churn probability is predicted; the behavioral characteristic data includes one or more of the following: changes in transaction amount, changes in assets, frequency and amount of bank-to-securities transfers, changes in stock prices, changes in earnings, changes in investment advisory performance, and changes in transaction characteristics.
[0007] The stock copy trading customer churn analysis system provided in the embodiment of the present application includes:
[0008] Customer behavior analysis module, used to monitor customer behavior characteristic data,
[0009] The customer behavior analysis module is used to predict the probability of customer churn based on the behavior characteristic data; the behavior characteristic data includes one or more of the following: changes in transaction amount, changes in assets, frequency and amount of bank-to-securities transfers, changes in stock prices, changes in returns, changes in investment advisory performance, and changes in transaction characteristics.
[0010] The electronic device provided in an embodiment of the present application includes: a processor and a memory, the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory to execute the stock copy trading customer churn analysis method provided in any embodiment of the present application.
[0011] The chip provided in the embodiment of the present application includes: a processor, which is used to call and run a computer program from a memory, so that a device equipped with the chip executes the stock copy trading customer churn analysis method provided in any embodiment of the present application.
[0012] The computer-readable storage medium provided in the embodiments of the present application is used to store a computer program, and the computer program enables a computer to execute the stock copycat customer churn analysis method provided in any embodiment of the present application.
[0013] The computer program product provided in the embodiments of the present application includes a computer program, and when the computer program is executed by a processor, the computer program implements the stock copy trading customer churn analysis method provided in any embodiment of the present application.
[0014] Through the stock copy trading customer churn analysis method provided in the embodiment of the present application, the behavioral characteristic data of the lost customers are processed and analyzed. The behavioral characteristic data include one or more of the following: changes in transaction amount, changes in assets, frequency and amount of bank-to-securities transfers, changes in stock prices, changes in returns, changes in investment advisory performance and changes in transaction characteristics. This can quickly identify potential lost customers, greatly reduce manual intervention, and make the lost customer identification process more efficient. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0016] Figure 1 Schematic diagram of the implementation process of the stock copy trading customer churn analysis method provided in the embodiment of the present application Figure 1
[0017] Figure 2 A schematic diagram of a loss curve provided in an embodiment of the present application;
[0018] Figure 3 A schematic diagram of a decision tree monitoring system provided in an embodiment of the present application;
[0019] Figure 4 Schematic diagram of the implementation process of the stock copy trading customer churn analysis method provided in the embodiment of the present application Figure 2
[0020] Figure 5 A schematic diagram of the structure of the stock copy trading customer churn analysis system provided in an embodiment of the present application;
[0021] Figure 6 A schematic structural diagram of an electronic device provided in an embodiment of the present application;
[0022] Figure 7 A schematic structural diagram of a chip provided in an embodiment of the present application. DETAILED DESCRIPTION
[0023] The following will describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0024] It should be noted that in the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the embodiments of the present application, the character " / " generally indicates that the associated objects before and after are in an "or" relationship.
[0025] In the description of the embodiments of the present application, the term "corresponding" may indicate a direct or indirect correspondence between two items, or an association relationship between the two items, or a relationship between indication and being indicated, configuration and being configured, and the like.
[0026] To facilitate understanding of the technical solutions of the embodiments of the present application, the relevant technologies of the embodiments of the present application are described below. The following related technologies can be arbitrarily combined with the technical solutions of the embodiments of the present application as optional solutions, and they all belong to the protection scope of the embodiments of the present application.
[0027] The goal of this application is to use advanced data analysis technology, especially big data and intelligent analysis methods, to automate and optimize the customer churn analysis process, aiming to improve the efficiency and accuracy of customer management. With the development of financial markets and the diversification of investor behavior, copy trading has become an important investment method. However, in the fierce market competition, the problem of customer churn has become the focus of major financial institutions. Effective analysis and prediction of customer churn can not only help institutions formulate effective customer retention strategies, but also optimize their service quality and customer experience.
[0028] Current customer churn analysis relies on traditional data statistics and manual analysis methods, which are not only time-consuming and labor-intensive, but also difficult to respond to market changes and the diversification of customer needs in real time. In addition, due to the limitations of the analysis methods, it is often impossible to fully and accurately identify potential churn customers. Therefore, how to use big data technology to automate the analysis and prediction of customer churn can not only improve the analysis efficiency, but also improve the accuracy of the prediction, which is an urgent problem that the financial industry needs to solve.
[0029] The following is an explanation of the terms related to this application:
[0030] 1) BDA-Big Data Analytics: refers to the processing of massive data through various analytical techniques and tools to extract valuable information and knowledge. Big data analysis can reveal patterns and trends in data and support decision-making and strategy formulation.
[0031] 2) CRP-Customer Retention Prediction: A data analysis-based method used to predict which customers are likely to churn and take measures to retain them. Customer retention prediction analyzes customer behavior data, identifies potential churn customers, and provides corresponding prevention strategies.
[0032] 3) ETL-Extract, Transform, Load: refers to the three steps in the data processing process. The ETL process extracts data from multiple sources, cleans and transforms the data, and then loads it into a data warehouse or database for subsequent analysis.
[0033] 4) ML-Machine Learning: An artificial intelligence method that allows a system to learn and improve its performance through data. Machine learning is widely used in customer churn analysis to identify and predict customer behavior patterns.
[0034] Among the related technologies, customer churn analysis technology mainly relies on traditional statistical analysis and basic behavioral analysis methods, which are inefficient in identifying and predicting customer churn, lack personalization and in-depth insights, and are difficult to cope with the complex and changing demands in the modern financial environment.
[0035] There are several major limitations to the techniques used in customer churn analysis, as shown below:
[0036] 1) Single analysis method: Most existing customer churn analysis tools rely on traditional statistical analysis and basic behavioral analysis methods. Although they can provide some basic analysis, they lack flexibility and depth and are difficult to adapt to complex and changing customer behavior patterns.
[0037] 2) Insufficient data processing capabilities: Due to their reliance on simple data processing technologies and limited data sources, these systems perform poorly when faced with massive, multi-dimensional data and are unable to fully and accurately capture the subtle changes and complex associations in customer behavior, resulting in insufficient accuracy and reliability of analysis results.
[0038] 3) Lack of personalized analysis: Existing technologies are rarely able to provide personalized churn analysis and prediction services based on the characteristics of different customer groups and individuals, which limits the pertinence and practicality of the analysis results and cannot meet the needs of financial institutions for precise customer management.
[0039] 4) Insufficient real-time and dynamic adjustment capabilities: Most tools lack real-time monitoring and dynamic adjustment mechanisms, and are unable to update analysis models and strategies in a timely manner according to the latest market trends and changes in customer behavior, making it difficult to maintain the effectiveness and foresight of analysis in a rapidly changing financial environment.
[0040] refer to Figure 1 , Figure 1 Schematic diagram of the implementation process of the stock copy trading customer churn analysis method provided in the embodiment of the present application Figure 1 ,like Figure 1 As shown, the method comprises the following steps:
[0041] Step 101: Monitor the customer's behavioral characteristic data.
[0042] In an embodiment of the present application, the behavioral characteristic data includes one or more of the following: changes in transaction amount, changes in assets, frequency and amount of bank-to-securities transfers, changes in stock prices, changes in returns, changes in investment advisory performance, and changes in transaction characteristics.
[0043] Step 102: Predict the customer churn probability based on the behavior characteristic data.
[0044] In an embodiment of the present application, a machine learning model can be used to analyze behavioral feature data and calculate the probability of customer churn.
[0045] The machine learning model is trained through customer historical behavior feature data; the customer churn probability is predicted through the trained machine learning model.
[0046] In the embodiment of the present application, the method further includes: analyzing the customer churn situation through a churn curve, and defining customers whose continuous churn exceeds a preset number of months as churned customers.
[0047] refer to Figure 2 , Figure 2 This is a schematic diagram of the loss curve provided in the embodiment of the present application. Figure 2 As shown, the churn rate in the first month is 65.15%, the churn rate in two consecutive months is 57.58%, the churn rate in three consecutive months is 53.27%, the churn rate in four consecutive months is 50.60%, and the churn rate in five consecutive months is 49.05%. By way of example, the preset months are preferably 4 months.
[0048] In an embodiment of the present application, a method for defining lost customers is provided, in which a machine learning model is trained using historical behavioral characteristic data of lost customers and non-lost customers. The trained machine learning model is used to predict the probability of customer loss. By inputting one or more data of the customer's transaction amount changes, asset changes, bank-to-securities transfer frequency and amount, stock price changes, income changes, investment advisor performance changes, and transaction characteristic changes, the model can accurately predict the customer's loss risk.
[0049] In an embodiment of the present application, the machine learning model includes one or more of the following models: a logistic regression model, a decision tree model, a random forest model, and a gradient boosting tree model.
[0050] For example, taking the decision tree model as an example, the decision tree model uses information entropy and information gain to select the best split attribute. The information entropy calculation formula is as follows: Ent(D) represents the information entropy of the data set D, which measures the purity or uncertainty of the data set. k It represents the proportion of samples belonging to the kth category in the data set, that is, the proportion of lost customers and non-lost customers, and |y| represents a total of y samples; the information gain calculation formula is as follows: Gain(D,a) represents the information gain of attribute a on data set D, which represents the improvement of attribute a on classification purity; D v Represents the subset divided by the vth value of attribute a.
[0051] The decision tree model determines the root node features and split node features based on the historical behavioral feature data of the input lost customers and non-lost customers through the information gain of the behavioral features. For each child node, the node splitting continues, and the features of each child node are determined based on the information gain of the behavioral features until the stopping condition is met, thereby obtaining the trained decision tree model; wherein the stopping condition includes: the number of samples in the node is lower than the preset threshold and / or the maximum tree depth is reached.
[0052] In the embodiment of the present application, the parameters of the trained decision tree model can also be tuned, such as adjusting the depth of the tree, to obtain an optimized decision tree model.
[0053] In an embodiment of the present application, the method further includes: tracking customer transaction behaviors and account activities to generate a churn curve; automatically identifying and marking customers who meet the churn definition by analyzing the churn curve to generate a list of churned customers.
[0054] Here, by analyzing the churn curve, customers whose continuous churn exceeds a preset number of months are defined as churned customers, helping managers to perform customer maintenance in a timely manner.
[0055] In an embodiment of the present application, the method also includes: collecting multi-source data; collecting customer transaction data, asset changes, bank-to-securities transfer records, stock prices, earnings, and investment advisory performance from multiple data sources as customer historical behavior feature data to train and / or update the machine learning model.
[0056] By collecting data from multiple data sources, we can enrich the training set of the machine learning model and improve the stability and prediction accuracy of the machine learning model.
[0057] In an embodiment of the present application, the method further includes: collecting data through an API interface and data stream processing technology to ensure that the data in the system can be updated in real time.
[0058] In an embodiment of the present application, before training the machine learning model using customer historical behavior characteristic data, the method further includes: performing data cleaning operations on the historical behavior characteristic data; illustratively, the data cleaning operations include: data integrity check: processing missing values, outliers, and duplicate data to ensure data integrity and accuracy; and data format unification: standardizing data formats from different sources and performing necessary conversions and processing.
[0059] In an embodiment of the present application, the method also includes: extracting key features; illustratively, extracting key features such as transaction amount, asset change rate, transfer frequency, stock price volatility, return change rate and investment advisor performance from the original data; and generating new derivative features through feature engineering to enhance the expressiveness of the model.
[0060] In the embodiments of the present application, the method for training the machine learning model may adopt multi-task learning, transfer learning, or other methods, and the present application does not limit this.
[0061] For example, multi-task learning: by training the model on multiple related tasks simultaneously, the generalization ability of the model and the understanding of specific tasks can be improved. For example, while predicting customer churn, customer classification or behavior prediction can also be trained. Transfer learning: for specific customer groups, the transfer learning method can be used to first pre-train on a general data set, and then fine-tune on data in a specific field to improve the adaptability of the model to specific groups.
[0062] In the embodiment of the present application, the method further includes: if the behavior characteristic data of the customer meets the set conditions, determining that the customer has a risk of churn;
[0063] The set condition includes: one or more behavior characteristic data among the behavior characteristic data reaches a set threshold; wherein the set thresholds of different behavior characteristic data are different and / or the set thresholds of different behavior characteristic data are the same.
[0064] Here, the setting conditions can be:
[0065]
[0066] refer to Figure 3 , Figure 3 The decision tree monitoring diagram provided in the embodiment of the present application is as follows: Figure 3 As shown, when the transaction amount changes by less than or equal to -20% and the total asset value changes by less than or equal to -10%, it is determined that the customer is at risk of loss and a warning is issued; when the transfer frequency changes by greater than or equal to 30% and the transfer amount is greater than or equal to RMB 50,000, it is determined that the customer is at risk of loss and a warning is issued; when the income changes by less than or equal to -15%, it is determined that the customer is at risk of loss and a warning is issued; other situations indicate that the customer is not at risk of loss and no warning is issued.
[0067] Based on this, in an embodiment of the present application, the method further includes: sending warning information to management personnel; the warning information indicates that there is a risk of loss of the customer.
[0068] This allows managers to receive early warning information in a timely manner, take action based on the early warning information, connect with customers in a timely manner, and prevent customer loss.
[0069] In an embodiment of the present application, the method further includes: generating a personalized intervention strategy based on the customer's behavioral characteristic data.
[0070] According to the customer's behavioral characteristics, we develop personalized intervention strategies for them to improve the success rate of intervention and prevent customer churn. For example, if the customer's profit change is less than or equal to -15%, we provide them with strategies to increase their profits in order to increase their profits and prevent customer churn.
[0071] In an embodiment of the present application, the method further includes: visually displaying the customer's churn probability; and / or generating a customer churn analysis report based on the customer's behavioral characteristic data and the churn probability.
[0072] refer to Figure 4 , Figure 4 Schematic diagram of the implementation process of the stock copy trading customer churn analysis method provided in the embodiment of the present application Figure 2 ,like Figure 4 As shown, the method comprises the following steps:
[0073] Step 401: Data collection and preprocessing.
[0074] Multi-source data collection: collect customer transaction data, asset changes, bank-to-securities transfer records, stock prices, earnings, investment advisory performance and other information from multiple data sources.
[0075] Real-time data update: Through API interface and data stream processing technology, ensure that the data in the system can be updated in real time.
[0076] Data integrity check: handle missing values, outliers, and duplicate data to ensure data integrity and accuracy.
[0077] Data format unification: standardize data formats from different sources and perform necessary conversion and processing.
[0078] Key feature extraction: Extract key features such as transaction amount, asset change rate, transfer frequency, stock price volatility, return change rate and investment advisor performance from the original data.
[0079] Feature engineering: Generate new derived features to enhance the expressiveness of the model.
[0080] Step 402: Define and identify lost customers.
[0081] Continuously track customer transaction behavior and account activities to generate roll-rate curves.
[0082] A threshold is set based on historical data, and customers who have lost customers for four consecutive months are defined as lost customers.
[0083] Automatically identify and mark customers who meet the churn definition and generate a list of churned customers.
[0084] Step 403: Customer behavior analysis and churn probability prediction.
[0085] Conduct in-depth analysis of lost customers to identify behavioral changes and possible reasons for loss:
[0086] 1) Analysis of changes in stock trading amount: Analyze the changes in the customer's trading amount before and after following the order and the trend during the following period.
[0087] 2) Asset change analysis: Analyze the changes in the total value and allocation of customers' assets.
[0088] 3) Bank-to-securities transfer analysis: Analyze the frequency and amount changes of customers’ bank-to-securities transfers.
[0089] 4) Stock price change analysis: Analyze the stock price changes of the stocks copied by the client.
[0090] 5) Analysis of profit changes: Analyze the changes in profit during the customer's order follow-up period.
[0091] 6) Analysis of changes in investment advisory performance: Tracking changes in the performance of the investment advisors followed by the clients.
[0092] 7) Analysis of changes in trading characteristics: Analyze changes in customer trading characteristics, such as changing from long-term to short-term.
[0093] The churn probability prediction model comprehensively analyzes various behavioral characteristics and changes, and uses machine learning techniques (such as logistic regression, decision trees, random forests, etc.) to train the model to calculate the customer's churn probability. By inputting data such as changes in customer transaction amounts, asset changes, bank-to-securities transfer frequency and amount, stock price changes, earnings changes, investment advisory performance changes, and transaction characteristics changes, the model can accurately predict the customer's churn risk. Based on the prediction results.
[0094] Step 404: Generate a personalized intervention strategy.
[0095] Financial institutions can promptly intervene in customers with high churn risk, develop personalized retention strategies, and improve customer management effectiveness and customer satisfaction.
[0096] Step 405: Real-time monitoring and early warning of customer behavior data.
[0097] Decision tree model: Analyze customer behavior data in real time through the decision tree model. When each feature or combination of features reaches the set threshold, an early warning is triggered. The decision tree model uses information entropy and information gain to select the best split attribute.
[0098] Step 406: Data visualization and report generation.
[0099] The analysis results are presented to users in an intuitive and easy-to-understand form, and a detailed analysis report is generated.
[0100] Data visualization: Display customer churn analysis results through dashboards and charts.
[0101] Report generation: The system automatically generates detailed customer churn analysis reports and supports exporting to multiple formats such as PDF and Excel.
[0102] Figure 5 A schematic diagram of the structure of the stock copy trading customer churn analysis system provided in the embodiment of the present application, such as Figure 5 As shown, the embodiment of the present application provides a stock copy trading customer churn analysis system, including:
[0103] Customer behavior analysis module, used to monitor customer behavior characteristic data;
[0104] The customer behavior analysis module is used to predict the probability of customer churn based on the behavior characteristic data; the behavior characteristic data includes one or more of the following: changes in transaction amount, changes in assets, frequency and amount of bank-to-securities transfers, changes in stock prices, changes in returns, changes in investment advisory performance, and changes in transaction characteristics.
[0105] In the embodiment of the present application, the customer behavior analysis module is specifically used to input the behavior feature data into a machine learning model, and calculate the customer churn probability through the machine learning model.
[0106] In the embodiment of the present application, the system further includes:
[0107] The churn customer definition and identification module is used to analyze customer churn through the churn curve and define customers who have lost customers continuously for more than a preset number of months as churned customers.
[0108] refer to Figure 2 , Figure 2 This is a schematic diagram of the loss curve provided in the embodiment of the present application. Figure 2 As shown, the churn rate in the first month is 65.15%, the churn rate in two consecutive months is 57.58%, the churn rate in three consecutive months is 53.27%, the churn rate in four consecutive months is 50.60%, and the churn rate in five consecutive months is 49.05%. By way of example, the preset months are preferably 4 months.
[0109] By defining lost customers, the customer behavior analysis module trains the machine learning model through the historical behavior feature data of lost customers and non-lost customers to obtain a trained machine learning model. The trained machine learning model is used to predict the probability of customer loss. By inputting one or more data of the customer's transaction amount change, asset change, bank-to-securities transfer frequency and amount, stock price change, income change, investment advisor performance change and transaction feature change, the model can accurately predict the customer's loss risk.
[0110] Here, the change in stock trading amount is to analyze the change in the client's trading amount before and after following the orders and the trend during the following period; the asset change analysis is to analyze the changes in the client's total asset value and configuration; the bank-to-securities transfer is to analyze the client's bank-to-securities transfer frequency and amount changes; the stock price change is to analyze the stock price change of the client's followed stocks; the income change is to analyze the change in the client's income during the following period; the investment advisor performance change is to track the performance changes of the investment advisor followed by the client; the transaction feature change is to analyze the changes in the client's transaction features, such as from long-term to short-term.
[0111] In an embodiment of the present application, the machine learning model includes one or more of the following models: a logistic regression model, a decision tree model, a random forest model, and a gradient boosting tree model.
[0112] In an embodiment of the present application, the customer churn customer definition and identification module can continuously track customer transaction behavior and account activity to generate a churn curve; by analyzing the churn curve, it can automatically identify and mark customers who meet the churn definition to generate a list of churn customers.
[0113] In the embodiment of the present application, the system further includes:
[0114] The data collection and preprocessing module is used to collect multi-source data; collect customer transaction data, asset changes, bank-to-securities transfer records, stock prices, earnings, and investment advisory performance from multiple data sources as customer historical behavior feature data to train and / or update machine learning models.
[0115] In the embodiment of the present application, the data collection and preprocessing module is also used to collect data through an API interface and data stream processing technology to ensure that the data in the system can be updated in real time.
[0116] In an embodiment of the present application, after the data collection and preprocessing module collects multi-source data, it can also perform data cleaning operations on the data; illustratively, the data cleaning operations include: data integrity check: processing missing values, outliers and duplicate data to ensure data integrity and accuracy; and data format unification: standardizing data formats from different sources and performing necessary conversions and processing.
[0117] In an embodiment of the present application, after the data collection and preprocessing module collects multi-source data, it can also extract key features; illustratively, key features such as transaction amount, asset change rate, transfer frequency, stock price volatility, return change rate and investment advisory performance are extracted from the original data; and new derivative features are generated through feature engineering to enhance the expressiveness of the model.
[0118] In an embodiment of the present application, the data collection and preprocessing module can also use traditional statistical analysis methods, such as regression analysis and time series analysis, to perform preliminary screening and analysis of customer behavior data, so that the machine learning model can perform more in-depth predictions.
[0119] In the embodiments of the present application, the method for training the machine learning model may adopt multi-task learning, transfer learning, or other methods, and the present application does not limit this.
[0120] For example, multi-task learning: by training the model on multiple related tasks simultaneously, the generalization ability of the model and the understanding of specific tasks can be improved. For example, while predicting customer churn, customer classification or behavior prediction can also be trained. Transfer learning: for specific customer groups, the transfer learning method can be used to first pre-train on a general data set, and then fine-tune on data in a specific field to improve the adaptability of the model to specific groups.
[0121] In the embodiment of the present application, the system further includes: an early warning and intervention module, which is used to determine that the customer has a risk of churn if the behavior characteristic data of the customer meets the set conditions;
[0122] The set condition includes: one or more behavior characteristic data among the behavior characteristic data reaches a set threshold; wherein the set thresholds of different behavior characteristic data are different and / or the set thresholds of different behavior characteristic data are the same.
[0123] For example, a decision tree model can be used to analyze customer behavior data in real time, and when each feature or a combination of features reaches a set threshold, it is determined that the customer is at risk of churn.
[0124] The decision tree model uses information entropy and information gain to select the best split attribute. The information entropy calculation formula is as follows: Ent(D) represents the information entropy of the data set D, which measures the purity or uncertainty of the data set. k It represents the proportion of samples belonging to the kth category in the data set, that is, the proportion of lost customers and non-lost customers, and |y| represents a total of y samples; the information gain calculation formula is as follows: Gain(D,a) represents the information gain of attribute a on data set D, which represents the improvement of attribute a on classification purity; D v represents the subset divided by the vth value of attribute a, D is the total number of samples in data set D, |D v | is a subset D v The number of samples.
[0125] Exemplarily, the setting conditions may be:
[0126]
[0127] refer to Figure 3 , Figure 3 The decision tree monitoring diagram provided in the embodiment of the present application is as follows: Figure 3 As shown, when the transaction amount changes by less than or equal to -20% and the total asset value changes by less than or equal to -10%, it is determined that the customer is at risk of loss and a warning is issued; when the transfer frequency changes by greater than or equal to 30% and the transfer amount is greater than or equal to RMB 50,000, it is determined that the customer is at risk of loss and a warning is issued; when the income changes by less than or equal to -15%, it is determined that the customer is at risk of loss and a warning is issued; other situations indicate that the customer is not at risk of loss and no warning is issued.
[0128] In the embodiment of the present application, the system further includes:
[0129] The early warning and intervention module is also used to generate personalized intervention strategies based on the customer's behavioral characteristic data.
[0130] The early warning and intervention module specifies personalized retention strategies for customers, which can improve the effectiveness of customer management and customer satisfaction.
[0131] In the embodiment of the present application, the early warning and intervention module is also used to send early warning information to management personnel; the early warning information indicates that there is a risk of loss of the customer.
[0132] The early warning and intervention module sends early warning information to managers, reminding them that there is a risk of customer loss, so that managers can retain users in time.
[0133] In the embodiment of the present application, the warning information includes a personalized intervention strategy specified for the customer, and the manager can retain the customer according to the personalized intervention strategy. For example, the manager can be an investment consultant and / or a customer manager.
[0134] In the embodiment of the present application, the system further includes:
[0135] The data visualization and report generation module is used to visualize the customer's churn probability; and / or to generate a customer churn analysis report based on the customer's behavioral feature data and the churn probability.
[0136] The data visualization and report generation module presents the analysis results to users in an intuitive and easy-to-understand form and generates a detailed analysis report. For example, it displays the customer churn analysis results through dashboards and charts. It automatically generates a detailed customer churn analysis report and supports exporting to multiple formats such as PDF and Excel.
[0137] In an embodiment of the present application, the system also includes a performance evaluation and iterative improvement module: used to iteratively optimize the system parameters based on the prediction results of machine learning, and / or the warning results and personalized intervention strategies of the warning and intervention strategies.
[0138] By evaluating the prediction results, early warning and intervention strategies of the machine learning model, the system's performance in different aspects is analyzed. Based on the evaluation results, the system parameters and training strategies are further optimized to improve the accuracy of customer churn analysis and the effectiveness of intervention strategies.
[0139] In the embodiment of the present application, the system further includes: a user interaction interface, through which feedback information on early warning and intervention strategies from investment advisors and account managers is collected, and the feedback information is used for continuous optimization and training of the model, forming a positive feedback loop. This ensures that the system can continuously adapt to the changing needs of users and the dynamic changes of the financial market.
[0140] The system architecture in the embodiment of the present application can adopt front-end and back-end separation and microservice architecture, use a distributed database and big data processing framework, and the system is deployed on a cloud platform to ensure high availability and security of data.
[0141] Those skilled in the art should understand that Figure 5 The implementation functions of each unit in the stock copy trading customer churn analysis system shown can be understood by referring to the relevant description of the aforementioned method. Figure 5 The functions of each unit in the stock copycat customer churn analysis system shown can be implemented by a program running on a processor, or by a specific logic circuit.
[0142] The stock copycat customer churn analysis method and stock copycat customer churn analysis system provided in the embodiment of the present application can quickly identify potential churn customers through automated data processing and analysis mechanisms. The decision tree model is used to analyze customer behavior data in real time, greatly reducing the need for manual intervention, making the churn customer identification process more efficient; the system can generate highly personalized intervention strategy recommendations based on the specific behavior and characteristics of the customer. By analyzing the customer's transaction amount, asset changes, bank-securities transfer and other data, targeted retention measures are provided to improve the accuracy of customer relationship management; the system has the ability to monitor customer behavior data in real time and can identify potential churn risks at the first time. Thresholds are set by the decision tree model. When the customer behavior data reaches the warning standard, the system automatically sends a warning notification to the investment advisor and the account manager to take timely measures; using a distributed database and a big data processing framework, the system can quickly process and analyze massive data and significantly improve data processing efficiency. Automated data collection, cleaning and feature extraction processes ensure high quality and consistency of data; the system displays customer churn analysis results through dashboards and charts to help users intuitively understand the analysis results. Automatically generate detailed customer churn analysis reports, support export to PDF, Excel and other formats, easy to archive and share; adopt front-end and back-end separation and microservice architecture, the system has good scalability and maintainability. Through cloud platform deployment, the elastic expansion capability of cloud resources is utilized to meet large-scale data processing needs while ensuring high availability and security of data; through performance evaluation and iterative improvement mechanism, the system can continuously optimize model parameters and training strategies, improve the accuracy of customer churn analysis and the effectiveness of intervention strategies. User feedback loops further ensure that the system can adapt to the changing needs of users and the dynamic changes of the financial market.
[0143] Figure 6 It is a schematic structural diagram of an electronic device 600 provided in an embodiment of the present application. Figure 6 The electronic device 600 shown includes a processor 610, which can call and run a computer program from a memory to implement the method in the embodiment of the present application.
[0144] Alternatively, if Figure 6 As shown, the electronic device 600 may further include a memory 620. The processor 610 may call and run a computer program from the memory 620 to implement the method in the embodiment of the present application.
[0145] The memory 620 may be a separate device independent of the processor 610 , or may be integrated into the processor 610 .
[0146] Alternatively, if Figure 6 As shown, the electronic device 600 may further include a transceiver 630, and the processor 610 may control the transceiver 630 to communicate with other devices, specifically, may send information or data to other devices, or receive information or data sent by other devices.
[0147] The transceiver 630 may include a transmitter and a receiver. The transceiver 630 may further include an antenna, and the number of the antennas may be one or more.
[0148] The electronic device 600 may specifically be the stock copycat customer churn analysis system of the embodiment of the present application, and the electronic device 600 may implement the corresponding processes implemented by the stock copycat customer churn analysis system in each method of the embodiment of the present application, which will not be described in detail here for the sake of brevity.
[0149] Illustratively, an embodiment of the present application further provides a computer program product, including a computer program, which can be executed by the processor 610 of the electronic device 600 to complete the steps of any of the aforementioned methods.
[0150] Figure 7 It is a schematic structural diagram of the chip of an embodiment of the present application. Figure 7 The chip 700 shown includes a processor 710, which can call and run a computer program from a memory to implement the method in the embodiment of the present application.
[0151] Alternatively, if Figure 7 As shown, the chip 700 may further include a memory 720. The processor 710 may call and run a computer program from the memory 720 to implement the method in the embodiment of the present application.
[0152] The memory 720 may be a separate device independent of the processor 710 , or may be integrated into the processor 710 .
[0153] Optionally, the chip 700 may further include an input interface 730. The processor 710 may control the input interface 730 to communicate with other devices or chips, and specifically, may obtain information or data sent by other devices or chips.
[0154] Optionally, the chip 700 may further include an output interface 740. The processor 710 may control the output interface 740 to communicate with other devices or chips, and specifically, may output information or data to other devices or chips.
[0155] The chip can be applied to the stock copycat customer churn analysis system in the embodiment of the present application, and the chip can implement the corresponding processes implemented by the stock copycat customer churn analysis system in each method of the embodiment of the present application. For the sake of brevity, it will not be repeated here.
[0156] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.
[0157] It should be understood that the processor of the embodiment of the present application may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by the hardware integrated logic circuit or software instructions in the processor. The above processor can be a general processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or the hardware and software modules in the decoding processor are combined and performed. The software module can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, and other mature storage media in the art. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.
[0158] It can be understood that the memory in the embodiments of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct RAM bus random access memory (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.
[0159] It should be understood that the above-mentioned memory is exemplary but not restrictive. For example, the memory in the embodiments of the present application may also be static random access memory (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (synch link DRAM, SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DR RAM), etc. That is to say, the memory in the embodiments of the present application is intended to include but not limited to these and any other suitable types of memory.
[0160] The embodiment of the present application also provides a computer-readable storage medium for storing a computer program. The computer-readable storage medium can be applied to the stock copy trading customer churn analysis system in the embodiment of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the stock copy trading customer churn analysis system in each method of the embodiment of the present application, which will not be described in detail here for the sake of brevity.
[0161] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0162] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0163] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0164] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0165] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a stock copycat customer churn analysis system, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.
[0166] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A method for analyzing stock copycat customer churn, characterized in that: The method comprises: Monitor customer behavior data; Based on the behavioral characteristic data, the customer churn probability is predicted; the behavioral characteristic data includes one or more of the following: changes in transaction amount, changes in assets, frequency and amount of bank-to-securities transfers, changes in stock prices, changes in earnings, changes in investment advisory performance, and changes in transaction characteristics.
2. The method according to claim 1, characterized in that: The predicting the probability of customer churn based on the behavioral feature data includes: The behavioral feature data is input into a machine learning model, and the customer churn probability is calculated by the machine learning model.
3. The method according to claim 1, characterized in that Also includes: If the customer's behavioral characteristic data meets the set conditions, it is determined that the customer is at risk of churn; the set conditions include: one or more of the behavioral characteristic data reaches a set threshold; wherein the set thresholds for different behavioral characteristic data are different and / or the set thresholds for different behavioral characteristic data are the same.
4. The method according to claim 2, characterized in that: Also includes: Generate personalized intervention strategies based on customer behavioral characteristic data; Send early warning information to management personnel; The early warning information indicates that there is a risk of loss of the customer.
5. The method according to any one of claims 2 to 4, characterized in that Also includes: Visualize the customer's churn probability; and / or generate a customer churn analysis report based on the customer's behavioral feature data and the churn probability.
6. A stock copy trading customer churn analysis system, characterized in that: The system comprises: Customer behavior analysis module, used to monitor customer behavior characteristic data; The customer behavior analysis module is used to predict the probability of customer churn based on the behavior characteristic data; the behavior characteristic data includes one or more of the following: changes in transaction amount, changes in assets, frequency and amount of bank-to-securities transfers, changes in stock prices, changes in returns, changes in investment advisory performance, and changes in transaction characteristics.
7. The system according to claim 6, characterized in that The customer behavior analysis module is specifically used to input the behavior feature data into a machine learning model and calculate the customer churn probability through the machine learning model.
8. The system according to claim 6 or 7, characterized in that: The system also includes an early warning and intervention module; The early warning and intervention module is used to determine that the customer is at risk of churn if the customer's behavioral characteristic data meets set conditions; the set conditions include: one or more of the behavioral characteristic data reaches a set threshold; wherein the set thresholds for different behavioral characteristic data are different and / or the set thresholds for different behavioral characteristic data are the same.
9. A computer-readable storage medium, characterized in that: Used to store a computer program, wherein the computer program enables a computer to execute the stock copycat customer churn analysis method as described in any one of claims 1 to 5.
10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the stock copycat customer churn analysis method according to any one of claims 1 to 5.
Citation Information
Cited By
Loss measuring and calculating method and device for two-fusion transaction type clients
CN119338512A
