SMS information optimization processing method and system based on user operation feedback
By establishing a multi-channel collaborative feedback management system and a multi-game management module, the problems of single user feedback channels and unscientific strategies are solved, in-depth analysis and personalized strategies are realized, and the scientificity of SMS management and user satisfaction are improved.
Patent Information
- Application Number
- CN202411244620.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-06
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-09-06
AI Technical Summary
In the prior art, the user feedback channel is single, the feedback characteristics are not thorough, the SMS management strategy is unscientific, and the SMS personalization is insufficient, resulting in poor SMS results and users are disgusted.
By establishing a feedback management system based on user segmentation and multi-channel collaboration, digging feedback feature matrix with user and communication dimensions, building a multi-objective game equilibrium analysis, determining SMS management strategies, and performing Nash equilibrium analysis for SMS customization Optimize SMS management with personalized strategies.
Comprehensive feedback management, in-depth feature analysis, scientific game management and personalized customization strategies have been realized, which improves user satisfaction and feedback effects, avoids user disgust, and makes SMS information management more scientific and effective.
Smart Images

Figure CN118764892B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of communication management technology, and in particular to a method and system for optimizing and processing short message information based on user operation feedback. Background Art
[0002] In the SMS information management scenario, with the explosive growth of information and the ever-changing user needs, the problem of SMS information optimization management is particularly prominent. Efficiently processing user feedback and optimizing SMS information to better meet user needs has become a crucial link in improving the quality of SMS management. Traditional SMS information management is often relatively single and lacks pertinence. It only focuses on users' active feedback or adopts relatively simple management strategies. The channels for collecting user feedback are relatively scattered and single, lacking multi-channel collaborative integration, and the analysis and processing of user feedback is not in-depth enough. Feedback features are not mined from multiple dimensions, making it difficult to fully and accurately understand user needs and behaviors. There are unreasonable situations in the formulation of SMS management strategies, resulting in poor SMS effects and easy to cause user disgust. SMS customization is relatively simple and fixed, and cannot cope well with the ever-changing user needs and communication environment.
[0003] At present, relevant technologies have technical problems such as single user feedback channel, insufficient analysis of feedback characteristics, unscientific SMS management strategy and insufficient SMS personalization. Summary of the invention
[0004] This application provides a method and system for optimizing the processing of SMS messages based on user operation feedback, and adopts a feedback management system based on user segmentation and multi-channel collaboration, including active and passive feedback. Combined with this system, the feedback feature matrix is mined, including user and communication dimensions. The game management module is constructed through sample-driven training, which has multiple game branches and can be coordinated. The feedback feature matrix is passed into the module for multi-objective game equilibrium analysis to determine the SMS management strategy, and then the Nash equilibrium analysis is performed for SMS customization to determine the personalized strategy. The sender optimizes the SMS information management based on the strategy, and achieves a win-win situation between the sender and the user through a comprehensive feedback management system, in-depth feedback feature analysis, scientific game management and personalized customization strategy, so as to improve user satisfaction and feedback effect, avoid user disgust, and make SMS information management more scientific and effective. Technical effect.
[0005] This application provides a method for optimizing SMS information processing based on user operation feedback, including:
[0006] A feedback management system is established based on user segmentation and multi-channel collaboration, wherein feedback types include active feedback and passive feedback from the user end; in combination with the feedback management system, data flow analysis is performed to mine a feedback feature matrix, wherein feedback features include user dimensions and communication dimensions; sample-driven training is performed to construct a game management module, wherein the game management module includes multiple game branches based on cooperative status, dynamic and static status, and information status, and the multiple game branches can collaborate; the feedback feature matrix is transmitted to the game management module, a multi-objective game equilibrium analysis is performed, and a short message management strategy is determined; the short message management strategy is identified, and a Nash equilibrium analysis based on the user end and the sending end is performed for short message customization to determine a personalized strategy; the sending end optimizes the management of short message information based on the short message management strategy and the personalized strategy.
[0007] This application also provides a SMS information optimization processing system based on user operation feedback, including:
[0008] A feedback management system establishment module, the feedback management system establishment module is used to establish a feedback management system based on user segmentation and multi-channel collaboration, wherein the feedback type includes active feedback and passive feedback from the user end; a feedback feature matrix mining module, the feedback feature matrix mining module is used to combine the feedback management system to perform data flow analysis and mine the feedback feature matrix, wherein the feedback features include user dimensions and communication dimensions; a game management module construction module, the game management module construction module is used to perform sample-driven training and construct a game management module, the game management module includes multiple game branches based on cooperation status, dynamic and static status, and information status, and the multiple game branches can be coordinated; a game equilibrium analysis module, the game equilibrium analysis module is used to transmit the feedback feature matrix to the game management module, perform multi-objective game equilibrium analysis, and determine the SMS management strategy; a Nash equilibrium analysis module, the Nash equilibrium analysis module is used to identify the SMS management strategy, perform Nash equilibrium analysis based on the user end and the sending end for SMS customization, and determine the personalized strategy; an optimization management module, the optimization management module is used for the sending end to optimize the management of SMS information based on the SMS management strategy and the personalized strategy.
[0009] The SMS information optimization processing method and system based on user operation feedback proposed in this application first establishes a feedback management system based on user segmentation and multi-channel collaboration, including active and passive feedback. Combined with this system, the feedback feature matrix is mined, including user and communication dimensions. The game management module is constructed through sample-driven training, which has multiple game branches and can be coordinated. The feedback feature matrix is passed into the module for multi-objective game equilibrium analysis to determine the SMS management strategy, and then the Nash equilibrium analysis is performed for SMS customization to determine the personalized strategy. The sender optimizes SMS information management based on the strategy. Through a comprehensive feedback management system, in-depth feedback feature analysis, scientific game management and personalized customization strategies, a win-win situation between the sender and the user is achieved, user satisfaction and feedback effects are improved, user disgust is avoided, and SMS information management is made more scientific and effective. Technical effects. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the following is a brief introduction to the drawings of the embodiment of the present invention. The present application uses a flow chart to illustrate the operations performed by the system according to the embodiment of the present application. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.
[0011] Figure 1 A flowchart of a method for optimizing SMS information processing based on user operation feedback provided in an embodiment of the present application;
[0012] Figure 2 A schematic diagram of the structure of a text message optimization processing system based on user operation feedback provided in an embodiment of the present application.
[0013] Explanation of the accompanying drawings: feedback management system establishment module 10, feedback feature matrix mining module 20, game management module construction module 30, game equilibrium analysis module 40, Nash equilibrium analysis module 50, optimization management module 60. DETAILED DESCRIPTION
[0014] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it is implemented in accordance with the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.
[0015] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.
[0016] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the term "first or second" involved is merely to distinguish similar objects and does not represent a specific ordering of the objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.
[0017] The present application embodiment provides a method for optimizing SMS information processing based on user operation feedback, such as Figure 1 As shown, the method includes:
[0018] Step S100, based on user segmentation and multi-channel collaboration, establish a feedback management system, where the feedback types include active feedback and passive feedback from the user side. Specifically, determine the user segmentation criteria, and consider factors such as age, gender, region, and consumption habits, select the segmentation method that best suits the business scenario, achieve multi-channel collaboration, determine the feedback channels to be integrated such as SMS feedback, online customer service platform, etc., establish a channel connection and data flow mechanism and perform standardized processing, collect active feedback, provide users with active feedback channels, promptly record and classify the opinions, suggestions and complaints actively raised by users, and monitor feedback based on user behavior records through data analysis tools, while collecting passive feedback, inferring passive feedback through system monitoring of user behavior and reaction, and quantifying and evaluating it using data analysis technology, establishing a feedback management system, designing feedback management processes and mechanisms, establishing a feedback database, formulating feedback processing standards and specifications, and regularly evaluating and optimizing the feedback management system.
[0019] In a possible implementation, a feedback management system is established based on user segmentation and multi-channel collaboration, wherein the feedback type includes active feedback and passive feedback from the user end, and step S100 further includes step S110, obtaining SMS elements, dividing multiple user classes based on individual user characteristics, and weighting the multiple user classes to determine a first weight distribution. Specifically, the SMS elements, i.e., the purpose and field of the SMS, are clarified. The elements are very important for understanding the nature and objectives of the SMS. For example, if the SMS is about fashion shopping, then its purpose may be to promote goods or provide fashion information. According to the SMS elements, users are classified based on individual user characteristics, considering basic characteristics such as gender and age of users, as well as historical data streams based on SMS targets. For example, if the SMS is about fashion shopping, users who are interested in fashion can be divided into a specific user class based on the user's historical purchase records, browsing behavior and other data, and the divided multiple user classes are weighted. The weighting process can be determined based on different factors, such as the size, importance, and activity of the user class. Through weighting, the relative importance of each user class in the entire user group can be determined, and finally the first weight distribution can be determined. The distribution reflects the importance of different user classes in SMS feedback management.
[0020] Step S120, traverse multiple SMS channels, perform channel weighting, and determine the second weight distribution. Specifically, multiple SMS channels are determined, such as mobile phone SMS, email SMS, push, etc. Each channel has its own characteristics and may cover different user groups, and the effect and efficiency of information transmission are also different. Each SMS channel is weighted, and the weighting can consider factors such as channel coverage, user participation, and feedback rate. For example, mobile phone SMS may have a higher coverage, but users may be more likely to ignore it; while email SMS may have a lower coverage, but users may read it more carefully. Through weighting, the relative importance of each channel in SMS feedback management can be determined, and finally the second weight distribution is determined. The distribution reflects the importance of different SMS channels in SMS feedback management.
[0021] Step S130, combining the multiple user classes with the multiple SMS channels, weighting and averaging the first weight distribution and the second weight distribution, and determining the information flow framework. Specifically, multiple user classes are combined with multiple SMS channels, and the feedback of different user classes on different SMS channels is considered. For example, a certain user class may be more inclined to give feedback via mobile phone SMS, while another user class may be more inclined to give feedback via email SMS. The first weight distribution (user class weight distribution) and the second weight distribution (SMS channel weight distribution) are weighted and averaged. The weighting process can be adjusted according to actual conditions. For example, the weights can be determined according to business needs, user behavior and other factors. By averaging, a comprehensive weight distribution can be obtained. This distribution reflects the relative importance of different user classes on different SMS channels. Finally, the information flow framework is determined. The framework describes the flow of SMS between different user classes and SMS channels, as well as the importance of each link, and provides a basis for establishing a feedback management system.
[0022] Step S140, based on the information flow framework, establish the feedback management system. Specifically, based on the information flow framework, establish the feedback management system, which includes the collection, classification, processing, analysis and other links of feedback. In the feedback collection link, determine from which user types and SMS channels to collect feedback according to the information flow framework. For example, a special feedback channel can be set up to allow users to provide feedback through mobile phone text messages, email text messages or push notifications. In the feedback classification link, classify the feedback according to the characteristics of user types and SMS channels. For example, the feedback can be divided into different types such as product problems, service quality, and functional requirements. In the feedback processing link, determine the corresponding processing method and priority according to the type and importance of the feedback. For example, urgent problems and complaints should be handled first; general suggestions can be handled at an appropriate time. In the feedback analysis link, use data analysis tools to analyze the feedback data to understand the needs and opinions of users, and provide a basis for improving SMS services. By establishing a feedback management system, SMS feedback can be better managed, user satisfaction can be improved, and strong support can be provided for the optimization of SMS services.
[0023] Step S200, in combination with the feedback management system, data flow analysis is performed to mine the feedback feature matrix, wherein the feedback feature includes the user dimension and the communication dimension. Specifically, the analysis scope is determined based on the feedback management system, covering the user categories divided by individual user characteristics and the management and weight allocation of multiple SMS channels. The feedback features include the user dimension and the communication dimension, wherein the user dimension includes collecting basic user information, analyzing user behavior data, and considering user historical feedback records; the communication dimension covers analyzing the time, frequency, and timing of SMS sending, studying the type and format of SMS content, and paying attention to the characteristics of SMS channels, performing data flow analysis, collecting relevant data flows from the feedback management system and preprocessing, extracting features from the two dimensions to construct a feedback feature matrix, mining the feedback feature matrix, finding out the relationship and rules between different features through data analysis, selecting important features and interpreting the results, and converting the analysis results into actual business recommendations and decision support.
[0024] Step S300, sample-driven training is performed to construct a game management module, wherein the game management module includes multiple game branches with cooperation status, dynamic and static status, and information status as baselines, and the multiple game branches can work together. Specifically, a large amount of sample data related to SMS feedback and management is collected and sorted, and the game management module is determined to be based on cooperation status, dynamic and static status, and information status as baselines, wherein the cooperation status is measured by analyzing the degree of cooperation between the user and the sender, the dynamic and static status evaluates whether the user's reaction to the SMS is active or passive, and the information status considers the amount of information and accuracy of the SMS content, and multiple game branches are constructed based on the baseline, and corresponding strategies are determined for different states. At the same time, a coordination mechanism is established to enable different game branches to cooperate and influence each other, and the relationship and degree of influence are determined through data analysis and model prediction. The coordination effect is continuously monitored and evaluated and adjusted and optimized to improve the performance and adaptability of the game management module, and provide effective support for the optimized management of SMS information.
[0025] In one possible implementation, sample-driven training is performed to construct a game management module, wherein the game management module includes multiple game branches based on cooperative state, dynamic and static state, and information state, and the multiple game branches can work together. Step S300 further includes step S310, wherein the univariate game branch includes a cooperative game block and a non-cooperative game block, the binary game branch includes a static game block and a dynamic game block, and the ternary game branch includes a complete information game block and an incomplete information game block. Specifically, it is clear that the branches of univariate games are divided into cooperative games and non-cooperative games. Cooperative games emphasize cooperation and coordination among participants to maximize common interests; non-cooperative games focus on the participants' pursuit of maximization of their own interests. The branches of binary games are divided into static games and dynamic games. Static games refer to the participants making decisions at the same time, and the decision-making process does not change over time; dynamic games refer to the participants' decisions having a sequence, and the later decision-makers can respond according to the actions of the earlier decision-makers. The branches of ternary games are divided into complete information games and incomplete information games. Complete information games refer to the participants having a complete understanding of each other's strategies, benefits and other information; incomplete information games refer to the participants having incomplete understanding of each other's information.
[0026] Step S320, traverse the feedback feature matrix to determine the game characteristics based on the feature decision criteria. Specifically, the feedback feature matrix is comprehensively traversed, and the matrix contains various feedback characteristics of the user dimension and the communication dimension, such as the user's age, gender, behavior data, and the sending time, frequency, content type of the text message, etc. The game characteristics are determined according to the feature decision criteria. The feature decision criteria can be formulated according to specific business needs and analysis goals. For example, if the feedback feature matrix shows that there is obvious feature correlation synergy between users, and the analysis is a static target analysis based on all information, then it may be determined as a completely static cooperative game characteristic. If the user's behavior changes significantly over time and the information is not fully understood, it may be determined as a dynamic incomplete information game characteristic, etc. Through in-depth analysis of the feedback characteristics, the game characteristics suitable for the current situation are determined to provide a basis for subsequent branch fitting processing.
[0027] Step S330, based on the game characteristics, branch fitting processing is performed to initialize the game management module. Specifically, the game characteristics are determined, and branch fitting processing is performed according to the characteristics. If it is determined to be a cooperative game characteristic, the relevant data and analysis focus are adjusted to the cooperative game block; if it is a static game characteristic, the strategy and analysis method of the static game block are focused on; if it is a complete information game characteristic, the tools and models of the complete information game block are fully utilized to initialize the game management module. According to the determined game characteristics, the parameters, strategies and initial states of the module are adjusted to enable it to adapt to specific game scenarios. For example, if it is determined to be a completely static cooperative game, the initial strategy of the module is set to a mode of promoting cooperation and making decisions based on all information. Through branch fitting processing and initialization, the game management module can effectively analyze and manage specific feedback situations, providing strong support for the subsequent SMS management strategy formulation and personalized strategy determination.
[0028] Step S400, the feedback feature matrix is transmitted to the game management module, and a multi-objective game equilibrium analysis is performed to determine the SMS management strategy. Specifically, the accuracy and integrity of the feedback feature matrix are ensured, an effective data transmission channel is established to transmit it to the game management module, the objectives of the multi-objective game are clarified, and the game relationship between the user and the sender is analyzed based on different branches and blocks in the game management module, including cooperation and non-cooperation, static and dynamic, complete information and incomplete information, etc., and the equilibrium point of the multi-objective game is comprehensively analyzed and found. The SMS management strategy is determined according to the equilibrium state, such as optimizing the SMS content, formulating a reasonable sending plan, improving the title to attract user attention, etc., and a comprehensive SMS management strategy is formulated by comprehensively considering the balance of multiple objectives to improve the effect and efficiency of SMS management.
[0029] In a possible implementation, the feedback feature matrix is transmitted to the game management module, a multi-objective game equilibrium analysis is performed, and the SMS management strategy is determined. Step S400 further includes step S410, traversing the feedback feature matrix, dividing the feedback features based on the feature decision criteria, and determining multiple management objectives, wherein the multiple management objectives correspond to SMS management dimensions, and the feature decision criteria are determined based on the SMS management dimensions. Specifically, the feedback feature matrix is comprehensively traversed. The matrix contains rich feedback features in user dimensions and communication dimensions, and is an important basis for formulating SMS management strategies. Feedback features are divided based on feature decision criteria. Feature decision criteria are determined according to the dimensions of SMS management, such as the click-through rate, reply rate, user satisfaction, and sending frequency of SMS. According to the criteria, feedback features are classified and divided into different feature subsets, and multiple management goals are determined. The management goals correspond to the dimensions of SMS management, and each goal represents a specific management direction. For example, if the SMS management dimensions include improving click-through rate and reply rate, then the corresponding management goal can be to increase the attractiveness and interactivity of SMS; if the dimensions include improving user satisfaction, the management goal can be to optimize SMS content and sending methods to meet user needs.
[0030] Step S420, traverse the multiple management goals, determine the management policy set of each management goal, and generate multiple management policy sets. Specifically, analyze the multiple management goals one by one, each management goal needs a set of corresponding management policies to achieve, determine the management policy set of each management goal, and propose a series of possible management policies for each management goal through analysis and research. For example, for the management goal of improving the click-through rate of SMS, you can consider optimizing SMS titles, adding personalized elements, setting up reward mechanisms and other strategies; for the management goal of improving user satisfaction, you can adopt strategies such as improving the quality of SMS content, reasonably controlling the sending frequency, and responding to user feedback in a timely manner to generate multiple management policy sets. Since there are multiple management goals, each goal has its own policy set. By combining these policy sets, you can generate multiple different management policy sets, and each management policy set represents a possible SMS management solution.
[0031] Step S430, combining the multiple management strategy sets, performing management equilibrium game optimization based on the multiple management objectives, and determining the SMS management strategy. Specifically, performing analysis in combination with the multiple management strategy sets, comparing and evaluating different management strategy sets, considering the effect and feasibility of each strategy set in achieving multiple management objectives, performing management equilibrium game optimization based on multiple management objectives, weighing and gaming among multiple management objectives, and finding an optimal equilibrium state, considering the mutual relationship and conflict between different management objectives, as well as the balance of interests between the user and the sender, determining the SMS management strategy, and selecting the best set of strategies from the multiple management strategy sets through management equilibrium game optimization as the final SMS management strategy. The strategy should be able to achieve a good balance between multiple management objectives, both meeting the needs of users and achieving the management objectives of the sender, thereby improving the effect and efficiency of SMS management.
[0032] In a possible implementation, combining the multiple groups of management strategy sets, performing management equilibrium game optimization based on the multiple management objectives, and determining the SMS management strategy, step S430 further includes step S431, obtaining an equilibrium game expression: ;in, For the The management policy set of the management target is relative to The cost function of the management strategy set for the remaining management objectives, The total amount of multiple management objectives. Indicates The set of strategies that management targets can choose from, Indicates Management target selection Strategy in The probability distribution of is a multiplication symbol, which is used to multiply the probability distribution of the management target to select a specific strategy. Indicates that except In addition to the management objectives, the combined strategy set of other management objectives, For the The cost of losses incurred by other management targets under the management strategy set of the management target. Specifically, Indicates The management policy set of the management target is relative to The cost function of the management strategy set for the remaining management objectives other than the management objective measures the cost of the first The effect of the set of strategies adopted for a management goal, is the total amount of multiple management objectives, representing the number of objectives in the entire management system. Indicates The set of strategies that can be selected for management objectives is Each management objective can select appropriate strategies from this strategy set to achieve its goals. Indicates Management target selection Strategy in The probability distribution of reflects the possibility of choosing different strategies in different situations. Indicates that except In addition to the management objectives, the combined strategy set of other management objectives represents the combination of strategies adopted by other management objectives. For the The cost of the loss of other management objectives under the management strategy set of the management objective, that is, when the first When a management target adopts a specific policy set, the impact on other management targets is , determine the set of strategies available to it By evaluating and analyzing different strategies, we can determine the feasibility and effectiveness of each strategy in achieving the management goal. The probability distribution of the combined strategy set considering other management objectives , analyze the interrelationships and impacts between different management objectives, and determine the impact of the strategy combination of other management objectives on the current management objective under specific circumstances, that is, , by calculating , it can measure the cost function of each management goal under different strategies relative to other management goals, help to make trade-offs and bargains among multiple management goals, find the optimal strategy combination, compare and evaluate different strategies for each management goal, determine the optimal strategy according to the size of the cost function, and select the strategy combination that can minimize the overall cost when considering multiple management goals. The final SMS management strategy should be a strategy that strikes a balance between multiple management goals, which can not only meet the requirements of each management goal, but also minimize the negative impact on other management goals, improve the effectiveness and efficiency of SMS management, and maximize the interests of users and senders.
[0033] In a possible implementation, the feedback feature matrix is transmitted to the game management module, a multi-objective game equilibrium analysis is performed, and a short message management strategy is determined. Step S400 further includes step S440, identifying the short message management strategy, determining short message customization elements, and initializing them. Specifically, the short message management strategy that has been determined is identified. The strategy is obtained through the previous multi-objective game equilibrium analysis and other steps, including measures such as optimizing short message content and controlling sending frequency, and determining short message customization elements. Short message customization elements refer to key factors that need to be considered in the process of short message customization. The elements may include the subject, style, language expression, and sending time of the short message. For example, if the short message management strategy emphasizes improving user satisfaction, then the short message customization elements may include using a friendly language style, determining the subject according to the user's interests, and initializing the short message customization elements. The initialization process may include determining the initial value or range of each element. For example, for the sending time element, an initial sending time period may be determined based on historical data and user behavior analysis.
[0034] Step S450, determining game constraints based on the SMS customized elements, wherein the game constraints include avoidance coefficient pairs, the avoidance coefficient pairs correspond to the SMS customized elements one by one, and each avoidance coefficient pair corresponds to a user end and a sending end respectively. Specifically, based on the SMS customization elements, the game constraints are determined. The constraints are the rules that need to be followed when the user end and the sender end are playing games. The game constraints include avoidance coefficient pairs. The avoidance coefficient pairs correspond to the SMS customization elements one by one, which means that each SMS customization element has a corresponding avoidance coefficient pair. The avoidance coefficient reflects the relative importance and degree of compromise of the user end and the sender end on the element. Each avoidance coefficient pair corresponds to the user end and the sender end respectively. If a certain SMS customization element is more important to the sender end but average to the receiver end, then the avoidance coefficient of the receiver end of the element is higher. For example, if the sender end hopes to send SMS at a specific time to improve the effect, but this time may cause interference to the user, then the avoidance coefficient of the user end on this time element will be higher, indicating that the user is more inclined to avoid this time. Conversely, the avoidance coefficient of the sender end on this element is lower, indicating that the sender end is more persistent in sending SMS at this time.
[0035] Step S460, based on the initialized SMS customization elements and the game constraints, a multi-objective game equilibrium analysis is performed. Specifically, based on the initialized SMS customization elements and combined with the game constraints, a multi-objective game equilibrium analysis is performed. The purpose of the analysis is to find the equilibrium state between the user end and the sender end between multiple goals under the premise of satisfying the constraints. In the analysis process, the importance and avoidance coefficient of each SMS customization element are considered. For elements with higher avoidance coefficients, more consideration is given in the decision-making process to meet the needs of the user end; for elements with lower avoidance coefficients, the sender end can be more persistent in its position. Through the multi-objective game equilibrium analysis, the final SMS customization plan is determined. The plan should strike a balance between the interests of the user end and the sender end, satisfying the needs of the user and achieving the goals of the sender end. It should also meet the game constraints to ensure that both parties abide by the rules during the game.
[0036] Step S500, identifying the SMS management strategy, performing Nash equilibrium analysis based on the user end and the sender end for SMS customization, and determining a personalized strategy. Specifically, identifying the determined SMS management strategy, and then performing game optimization between features based on feedback information, determining the feedback adjustment target, such as improving the attractiveness and relevance of a certain type of SMS based on the user's feedback on the low opening rate, and then determining the adjustment direction and amplitude, the key factors can be found and the adjustment method and amplitude can be determined by analyzing the feedback feature matrix and user behavior data, considering the benefit status of the user end and the sender end, and analyzing the interests of both parties to find a balance point, such as providing the price information or preferential activities of commodity A to the sender end for users who have the demand to purchase commodity A, and performing personalized customization on this basis, providing personalized SMS content according to the user's personal characteristics and behavior data, optimizing the sending time and frequency, and adopting an interactive form, etc., and finally determining the personalized strategy, which should be obtained through game optimization and personalized customization based on the consideration of the benefit status of both parties, including specific measures such as SMS content, sending time, frequency, and form to improve the effect and user satisfaction.
[0037] Step S600, the sending end optimizes the management of SMS information based on the SMS management strategy and the personalization strategy. Specifically, the sending end first deeply understands the SMS management strategy and the personalization strategy, including macro guidelines and specific measures. In terms of optimizing SMS content, the sending end adjusts the theme and language style according to the strategy, determines the theme around the user's interest points, uses language expressions that meet the user's preferences, and adds valuable information, such as useful information and preferential activities. In terms of controlling the sending frequency, the sending end analyzes user feedback and behavior data to determine the appropriate frequency, formulates a reasonable sending plan, considers the user's life rhythm and habits to select the time point, and determines different frequency strategies according to business needs and target user groups. When implementing personalized sending, the sending end uses user data for precise sending, subdivides user groups to customize different content, and continuously monitors user feedback and behavior changes during the sending process, adjusts the personalization strategy in real time, establishes an evaluation index system to measure the management effect, and continuously optimizes according to the evaluation results. If user satisfaction decreases, the content and frequency are optimized, and if the conversion rate is low, the personalization strategy is adjusted to improve the pertinence and attractiveness.
[0038] In a possible implementation, the sending end optimizes the management of SMS information based on the SMS management strategy and the personalization strategy, and step S600 further includes step S610, determining feedback trend curves of upper feedback and lower feedback for the neighborhood feedback cycle. Specifically, the concept of neighborhood feedback cycle is clarified. The neighborhood feedback cycle can be understood as a cycle formed by the feedback of users on information such as text messages within a specific time period. The cycle may be affected by many factors, such as time, season, events, etc., and the upper feedback and lower feedback are distinguished. The upper feedback can be relatively macro and overall feedback, such as the user's overall satisfaction with the text message service within a period of time, the overall impression of a certain brand, etc.; the lower feedback can be more specific and detailed feedback, such as specific opinions on the content of a certain text message, the sending time, etc., by analyzing the upper feedback and lower feedback within the neighborhood feedback cycle, their feedback trend curves are determined, and this is achieved by collecting and organizing the feedback data of users at different time points and drawing a curve of feedback changes over time. For example, if the user's satisfaction with text messages gradually increases within a certain period of time, then the trend curve of the upper feedback will show an upward trend; if the negative feedback on a specific type of text message gradually increases, then the trend curve of the lower feedback may show a downward trend.
[0039] Step S620, based on the feedback trend curve, trace back and mine the strongly related features of the user end. Specifically, based on the feedback trend curve, trace back and mine. This means that it is necessary to deeply analyze the reasons for the changes in the feedback curve, find out the key factors that affect user feedback, and consider the possible decisive factors. As mentioned in the note, there may be a decisive factor that suddenly causes the user's preferences, interests, habits, etc. to be adjusted. For example, the user's demand and feedback for text messages may change due to the emergence of new competitors in the market or major changes in the user's living environment. Through the analysis of user behavior, feedback data, and external environment, the strongly related features of the user end are mined, including the user's new needs, changes in preferences, sensitivity to specific information, etc. For example, if it is found that the user has positive feedback on text messages on environmental protection topics within a certain period of time, then environmental awareness may be a strongly related feature.
[0040] Step S630, based on the strongly correlated features, policy compensation is performed. Specifically, once the strongly correlated features are determined, policy compensation is performed based on these features. The purpose of policy compensation is to adjust the existing SMS management strategy and personalization strategy to better meet the needs of users and adapt to changes. If the mined strongly correlated features indicate that the user's interests have changed, the sender can adjust the subject and content of the SMS accordingly to meet the user's new interests. For example, if the user's attention to science and technology news increases, the sending of science and technology-themed SMS can be increased. If the strongly correlated features show that the user has new requirements for a specific sending time or frequency, the sender can adjust the sending plan to improve user satisfaction. For example, if the user's feedback is more positive at night, the amount of SMS sent at night can be increased. Through policy compensation, the sender can better respond to changes in user feedback, improve the effect of SMS information optimization management, and enhance user participation and satisfaction.
[0041] The embodiment of the present application adopts a feedback management system based on user segmentation and multi-channel collaboration, including active and passive feedback. Combined with this system, the feedback feature matrix is mined, including user and communication dimensions. The game management module is constructed through sample-driven training, which has multiple game branches and can be coordinated. The feedback feature matrix is passed into the module for multi-objective game equilibrium analysis to determine the SMS management strategy, and then the Nash equilibrium analysis is performed on SMS customization to determine the personalized strategy. The sender optimizes SMS information management based on the strategy, achieving a win-win situation between the sender and the user, improving user satisfaction and feedback effects, avoiding user disgust, and making SMS information management more scientific and effective. Technical effect.
[0042] In the above, refer to Figure 1The text message optimization processing method based on user operation feedback according to an embodiment of the present invention is described in detail. Figure 2 A short message optimization processing system based on user operation feedback according to an embodiment of the present invention is described.
[0043] The SMS information optimization processing system based on user operation feedback according to the embodiment of the present invention is used to solve the technical problems of single user feedback channel, in-depth analysis of feedback characteristics, unscientific SMS management strategy and insufficient SMS personalization in the prior art. Through a comprehensive feedback management system, in-depth feedback characteristic analysis, scientific game management and personalized customization strategy, it achieves a win-win situation between the sender and the user, improves user satisfaction and feedback effect, avoids user disgust, and makes SMS information management more scientific and effective. The SMS information optimization processing system based on user operation feedback includes: a feedback management system establishment module 10, a feedback feature matrix mining module 20, a game management module construction module 30, a game equilibrium analysis module 40, a Nash equilibrium analysis module 50, and an optimization management module 60.
[0044] The feedback management system establishment module 10 is used to establish a feedback management system based on user segmentation and multi-channel collaboration, wherein the feedback types include active feedback and passive feedback from the user end;
[0045] The feedback feature matrix mining module 20 is used to perform data flow analysis and mine feedback feature matrix in combination with the feedback management system, wherein the feedback features include user dimension and communication dimension;
[0046] The game management module construction module 30 is used to perform sample-driven training and construct a game management module, wherein the game management module includes multiple game branches based on cooperation state, dynamic state, and information state, and the multiple game branches can be coordinated;
[0047] The game equilibrium analysis module 40 is used to transmit the feedback feature matrix to the game management module, perform multi-objective game equilibrium analysis, and determine the SMS management strategy;
[0048] The Nash equilibrium analysis module 50 is used to identify the SMS management strategy, perform Nash equilibrium analysis based on the user end and the sender end for SMS customization, and determine the personalized strategy;
[0049] The optimization management module 60 is used for the sending end to optimize the management of SMS information based on the SMS management strategy and the personalization strategy.
[0050] The specific configuration of the feedback management system establishment module 10 will be described in detail below. As described above, a feedback management system is established based on user segmentation and multi-channel collaboration, wherein the feedback type includes active feedback and passive feedback from the user end, and the feedback management system establishment module 10 further includes: a first weight distribution determination unit, the first weight distribution determination unit is used to obtain SMS elements, divide multiple user classes based on individual user characteristics, and weight the multiple user classes to determine the first weight distribution; a second weight distribution determination unit, the second weight distribution determination unit is used to traverse multiple SMS channels, perform channel weighting, and determine the second weight distribution; an information flow framework determination unit, the information flow framework determination unit is used to combine the multiple user classes and the multiple SMS channels, weight the first weight distribution and the second weight distribution to obtain the average, and determine the information flow framework; a feedback management system establishment unit, the feedback management system establishment unit is used to establish the feedback management system based on the information flow framework.
[0051] The specific configuration of the game management module construction module 30 will be described in detail below. As described above, sample-driven training is performed to construct a game management module, the game management module includes multiple game branches based on cooperative state, static state, and information state, and the multiple game branches can be coordinated, and the game management module construction module 30 further includes: a game branch composition unit, the game branch composition unit is used for a one-element game branch including a cooperative game block and a non-cooperative game block, a binary game branch including a static game block and a dynamic game block, and a ternary game branch including a complete information game block and an incomplete information game block; a game characteristic determination unit, the game characteristic determination unit is used to traverse the feedback feature matrix and determine the game characteristics based on the feature decision standard; an initialization unit, the initialization unit is used to perform branch fitting processing based on the game characteristics and initialize the game management module.
[0052] The specific configuration of the game equilibrium analysis module 40 will be described in detail below. As described above, the feedback feature matrix is transmitted to the game management module to perform multi-objective game equilibrium analysis to determine the SMS management strategy. The game equilibrium analysis module 40 further includes: a management target determination unit, which is used to traverse the feedback feature matrix, divide the feedback features based on the feature decision criteria, and determine multiple management targets, wherein the multiple management targets correspond to the SMS management dimension, and the feature decision criteria are determined based on the SMS management dimension; a management strategy set generation unit, which is used to traverse the multiple management targets, determine the management strategy set of each management target, and generate multiple groups of management strategy sets; and a balanced game optimization unit, which is used to combine the multiple groups of management strategy sets to perform management equilibrium game optimization based on the multiple management targets to determine the SMS management strategy.
[0053] Among them, in combination with the multiple groups of management strategy sets, a management equilibrium game optimization based on the multiple management objectives is performed to determine the SMS management strategy, and the equilibrium game optimization unit further includes: an equilibrium game expression acquisition subunit, and the equilibrium game expression acquisition subunit is used to obtain an equilibrium game expression: ;in, For the The management policy set of the management target is relative to The cost function of the management strategy set for the remaining management objectives, The total amount of multiple management objectives. Indicates The set of strategies that management targets can choose from, Indicates Management target selection Strategy in The probability distribution of Indicates that except In addition to the management objectives, the combined strategy set of other management objectives, For the The cost of losses incurred by the remaining management objectives under the management strategy set of the management objective.
[0054] Among them, the game equilibrium analysis module 40 further includes: a text message customization factor determination unit, the text message customization factor determination unit is used to identify the text message management strategy, determine the text message customization factor and initialize it; a game constraint condition determination unit, the game constraint condition determination unit is used to determine the game constraint condition based on the text message customization factor, wherein the game constraint condition includes an avoidance coefficient pair, the avoidance coefficient pair corresponds to the text message customization factor one by one, and each avoidance coefficient pair corresponds to the user end and the sending end respectively; a multi-objective game equilibrium analysis unit, the multi-objective game equilibrium analysis unit is used to perform a multi-objective game equilibrium analysis based on the initialized text message customization factor and the game constraint condition.
[0055] The specific configuration of the optimization management module 60 will be described in detail below. As described above, the sending end performs optimization management of SMS information based on the SMS management strategy and the personalization strategy, and the optimization management module 60 further includes: a feedback trend curve determination unit, which is used to determine the feedback trend curve of the upper feedback and the lower feedback for the neighborhood feedback cycle; a strongly correlated feature mining unit, which is used to trace and mine the strongly correlated features of the user end based on the feedback trend curve; and a strategy compensation unit, which is used to perform strategy compensation based on the strongly correlated features.
[0056] The SMS information optimization processing system based on user operation feedback provided in the embodiment of the present invention can execute the SMS information optimization processing method based on user operation feedback provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.
[0057] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.
[0058] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
Claims
1. A method for optimizing SMS information based on user operation feedback, characterized in that: The method comprises: A feedback management system is established based on user segmentation and multi-channel collaboration, where feedback types include active feedback and passive feedback from the user side; In combination with the feedback management system, data flow analysis is performed to mine a feedback feature matrix, wherein the feedback features include user dimensions and communication dimensions; Conduct sample-driven training and construct a game management module, wherein the game management module includes multiple game branches based on cooperation state, dynamic state, and information state, and the multiple game branches can work together; Transmitting the feedback feature matrix to the game management module to perform multi-objective game equilibrium analysis and determine the SMS management strategy; Identify the SMS management strategy, perform Nash equilibrium analysis based on the user end and the sender end for SMS customization, and determine a personalized strategy; The sending end performs optimized management of SMS information based on the SMS management strategy and the personalization strategy; The establishment of a feedback management system includes: Acquire SMS elements, divide the elements into multiple user classes based on individual user characteristics, and assign weights to the multiple user classes to determine a first weight distribution; Traverse multiple SMS channels, assign channel weights, and determine the second weight distribution; Combining the multiple user classes and the multiple SMS channels, weighting and averaging the first weight distribution and the second weight distribution, and determining an information flow framework; Based on the information flow framework, establishing the feedback management system; Conduct multi-objective game equilibrium analysis to determine SMS management strategies, including: Traversing the feedback feature matrix, dividing the feedback features based on feature decision criteria, and determining a plurality of management objectives, wherein the plurality of management objectives correspond to SMS management dimensions, and the feature decision criteria are determined based on the SMS management dimensions; Traversing the multiple management targets, determining a management policy set for each management target, and generating multiple groups of management policy sets; In combination with the multiple groups of management strategy sets, a management equilibrium game optimization based on the multiple management objectives is performed to determine the SMS management strategy.
2. The method for optimizing and processing short message information based on user operation feedback according to claim 1, characterized in that: Construct a game management module, including: The univariate game branch includes cooperative game blocks and non-cooperative game blocks, the binary game branch includes static game blocks and dynamic game blocks, and the ternary game branch includes complete information game blocks and incomplete information game blocks; Traversing the feedback feature matrix to determine the game characteristics based on the feature decision criteria; Based on the game characteristics, branch fitting processing is performed to initialize the game management module.
3. The method for optimizing and processing short message information based on user operation feedback according to claim 1, characterized in that: The method further comprises: Get the equilibrium game expression: ; in, For the The management policy set of the management target is relative to The cost function of the management strategy set for the remaining management objectives, The total amount of multiple management objectives. Indicates The set of strategies that management targets can choose from, Indicates Management target selection Strategy in The probability distribution of Indicates that except In addition to the management objectives, the combined strategy set of other management objectives, For the The cost of losses incurred by the remaining management objectives under the management strategy set of the management objective.
4. The method for optimizing and processing short message information based on user operation feedback according to claim 1, characterized in that: Before the multi-objective game equilibrium analysis is performed, the following steps are included: Identify the SMS management strategy, determine SMS customization elements and initialize them; Determining a game constraint condition based on the SMS customization element, wherein the game constraint condition includes an avoidance coefficient pair, the avoidance coefficient pair corresponds to the SMS customization element one by one, and each avoidance coefficient pair corresponds to a user end and a sending end respectively; Based on the initialized SMS customization factors and the game constraints, a multi-objective game equilibrium analysis is performed.
5. The method for optimizing and processing short message information based on user operation feedback according to claim 1, characterized in that: The method further comprises: According to the neighborhood feedback cycle, determine the feedback trend curves of upper feedback and lower feedback; Based on the feedback trend curve, trace back and mine the strongly relevant features of the user end; Based on the strongly correlated features, strategy compensation is performed.
6. A short message information optimization processing system based on user operation feedback, characterized in that: The system is used to implement the SMS information optimization processing method based on user operation feedback according to any one of claims 1 to 5, and the system includes: A feedback management system establishment module, which is used to establish a feedback management system based on user segmentation and multi-channel collaboration, wherein the feedback types include active feedback and passive feedback from the user end; A feedback feature matrix mining module, wherein the feedback feature matrix mining module is used to perform data flow analysis and mine feedback feature matrices in combination with the feedback management system, wherein the feedback features include user dimensions and communication dimensions; A game management module construction module, wherein the game management module construction module is used to perform sample-driven training and construct a game management module, wherein the game management module includes multiple game branches based on cooperation state, dynamic state, and information state, and the multiple game branches can be coordinated; A game equilibrium analysis module, which is used to transmit the feedback feature matrix to the game management module, perform multi-objective game equilibrium analysis, and determine the SMS management strategy; A Nash equilibrium analysis module, wherein the Nash equilibrium analysis module is used to identify the SMS management strategy, perform Nash equilibrium analysis based on the user end and the sender end for SMS customization, and determine a personalized strategy; An optimization management module, wherein the optimization management module is used for the sending end to optimize the management of SMS information based on the SMS management strategy and the personalization strategy; The feedback management system establishment module also includes: a first weight distribution determination unit, the first weight distribution determination unit is used to obtain SMS elements, divide multiple user classes based on user individual characteristics, and assign weights to the multiple user classes to determine a first weight distribution; a second weight distribution determination unit, the second weight distribution determination unit is used to traverse multiple SMS channels, assign channel weights, and determine a second weight distribution; an information flow framework determination unit, the information flow framework determination unit is used to combine the multiple user classes with the multiple SMS channels, assign weights to the first weight distribution and the second weight distribution, and determine the information flow framework; a feedback management system establishment unit, the feedback management system establishment unit is used to establish the feedback management system based on the information flow framework; The game equilibrium analysis module also includes: a management target determination unit, which is used to traverse the feedback feature matrix, divide the feedback features based on the feature decision criteria, and determine multiple management targets, wherein the multiple management targets correspond to SMS management dimensions, and the feature decision criteria are determined based on the SMS management dimensions; a management strategy set generation unit, which is used to traverse the multiple management targets, determine the management strategy set of each management target, and generate multiple groups of management strategy sets; and a balanced game optimization unit, which is used to combine the multiple groups of management strategy sets, perform management equilibrium game optimization based on the multiple management targets, and determine the SMS management strategy.
