Game user touch classification management method, system and equipment and medium

Abstract

The invention relates to the technical field of game operation, in particular to a game user reach classification management method, system and device and a medium, and the method specifically comprises the following steps: obtaining user reach data of each business line, and carrying out traffic dyeing on the user reach data through a preset classification label to form a unified business reach classification standard system; based on the service reach classification standard system, performing association analysis on an abnormal reach event occurring in a reach process and mark information of each link of the reach process, and determining an abnormal reason and a disposal strategy of the abnormal reach event; an association rule mining algorithm is utilized to analyze the arrival relevance between different service lines, and the arrival strategy is optimized by identifying the possible repeated arrival condition and excessive disturbance condition. According to the method, the pertinence and effectiveness of touch are improved, the visualization and transparency of the whole touch process are also realized, and the game operation efficiency and the user experience are greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of game operation technology, and in particular to a game user reach classification management method, system, device and medium. Background Art

[0002] With the rapid development of the gaming industry, gaming operation scenarios are becoming increasingly diverse, and user reach capabilities are also becoming more diverse. In the process of gaming operations, in order to effectively convey information, improve user experience, and increase user stickiness, gaming platforms need to interact with users through a variety of reach methods (such as SMS, outbound calls, WeChat messages, emails, etc.). However, with the increase in reach methods, users may be perceived by a variety of reach capabilities at any time, which not only brings users a rich information receiving experience, but may also cause users to feel excessively disturbed.

[0003] From the user's perspective, frequent and disorderly contact may cause users to feel disgusted and even cause user loss. Therefore, users expect to receive valuable information at the right time and through the right contact method. This requires the game platform to clearly understand the distribution of user contact and formulate personalized contact strategies based on user preferences and needs.

[0004] From the perspective of the game platform, facing complex business needs and marketing goals, how to effectively manage user reach has become an urgent problem to be solved. The game platform needs to clearly understand the reach of each business port to users, including the reach method, reach content, reach time, etc., to ensure the effectiveness of the reach strategy. At the same time, in extreme cases, the platform also needs to be able to quickly locate the cause of abnormal reach to users so that timely measures can be taken to correct it.

[0005] However, existing user reach management methods often lack unified standards and systems, resulting in the inability to effectively integrate and analyze reach data between different business lines. In addition, existing reach strategies are often formulated based on experience, lack data support and personalized processing, and are difficult to meet users' diverse needs for reach methods and content. Summary of the invention

[0006] The purpose of the present invention is to provide a game user reach classification management method, system, device and medium, which not only improves the pertinence and effectiveness of the reach, but also realizes the visualization and transparency of the entire reach process, greatly improves the efficiency of game operations and user experience, so as to solve at least one of the above-mentioned prior art problems.

[0007] In a first aspect, the present invention provides a method for managing game user reach classification, the method specifically comprising:

[0008] Obtain user reach data for each business line, and color the user reach data using pre-set classification labels to form a unified business reach classification standard system;

[0009] Based on the business contact classification standard system, perform personalized contact behavior standard processing for each business line;

[0010] Based on the business contact classification standard system, the abnormal contact events occurring during the contact process and the tag information of each link of the contact process are correlated and analyzed to determine the abnormal causes and handling strategies of the abnormal contact events;

[0011] Use association rule mining algorithms to analyze the reach correlation between different business lines, and optimize reach strategies by identifying possible repeated reach and excessive interruptions.

[0012] By modeling and analyzing user behavior data, the reach preference and acceptance level of each user are determined, and the reach strategy is optimized according to the reach preference and acceptance level;

[0013] Based on the key indicator data obtained from each link in the reaching process, the reaching effect in different scenarios is evaluated, and the reaching strategy is optimized based on the reaching effect.

[0014] In a second aspect, the present invention provides a game user reach classification management system, the system specifically comprising:

[0015] The first management module is used to obtain user reach data of each business line, and to color the user reach data with preset classification labels to form a unified business reach classification standard system;

[0016] A second management module is used to perform personalized contact behavior standard processing on each business line based on the business contact classification standard system;

[0017] The third management module is used to perform correlation analysis on abnormal contact events occurring during the contact process and tag information of each link of the contact process based on the business contact classification standard system, and determine the abnormal cause and disposal strategy of the abnormal contact event;

[0018] The fourth management module is used to analyze the reach correlation between different business lines using association rule mining algorithms, and optimize the reach strategy by identifying possible repeated reach and excessive interruption situations;

[0019] A fifth management module is used to determine the reach preference and acceptance level of each user by modeling and analyzing the user behavior data, and optimize the reach strategy according to the reach preference and acceptance level;

[0020] The sixth management module is used to evaluate the reaching effects in different scenarios based on the key indicator data obtained in each link of the reaching process, and optimize the reaching strategy based on the reaching effects.

[0021] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor and a computer program stored in the memory, and when the computer program is executed on the processor, it implements the game user reach classification management method as described in any one of the above methods.

[0022] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the game user reach classification management method as described in any one of the above methods is implemented.

[0023] Compared with the prior art, the present invention has at least one of the following technical effects:

[0024] 1. The present invention not only improves the pertinence and effectiveness of reaching, but also realizes the visualization and transparency of the entire reaching process, greatly improving the efficiency of game operation and user experience.

[0025] 2. The present invention uses pre-set classification labels to color user reach data traffic, forming a unified business reach classification standard system, which solves the problem that reach data between different business lines cannot be effectively integrated and analyzed.

[0026] 3. Based on the business reach classification standard system, the present invention performs personalized reach behavior standard processing on each business line to ensure that the reach strategy meets the user's preferences and needs.

[0027] 4. The present invention can quickly locate the cause of abnormal reaching of users by associating and analyzing abnormal reaching events that occur during the reaching process and the marking information of each link in the reaching process, thus providing strong support for timely correction.

[0028] 5. The present invention uses association rule mining algorithms to analyze the reach correlation between different business lines, identify repeated reaches and excessive interruptions, and determine users' reach preferences and acceptance levels through user behavior data modeling and analysis, thereby optimizing reach strategies and improving user experience.

[0029] 6. The present invention evaluates the reaching effects in different scenarios based on the key indicator data obtained in each link of the reaching process, and provides data support for further optimizing the reaching strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0031] Figure 1 It is a flowchart of a method for managing game user reach classification provided by an embodiment of the present invention;

[0032] Figure 2 It is a structural diagram of a game user reach classification management system provided by an embodiment of the present invention;

[0033] Figure 3 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0035] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0036] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0037] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0038] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0039] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0040] In the embodiment of the present application, the execution subject of the process includes a terminal device, which includes but is not limited to: a server, a computer, a smart phone, a tablet computer, and other devices capable of executing the method disclosed in the present application. Figure 1 A flowchart of a method for managing game user reach classification disclosed in an embodiment of the present invention is shown, and the details are as follows:

[0041] S101, obtaining user reach data of each business line, coloring the user reach data with pre-set classification labels, and forming a unified business reach classification standard system.

[0042] In this embodiment, according to the complexity of the game business scenario, the user contact methods of each business line are uniformly classified and managed. For different contact methods, corresponding classification tags are set, and the contact link is colored by classification tags, marking key information such as business parties, scenarios, trigger sources, and contact methods to achieve visualization of the entire contact process. The user contact data is colored by preset classification tags, making the data more intuitive and easy to understand. The company can clearly see the user contact situation of each business line and the response differences of different user groups.

[0043] In some embodiments, in the above step S101, the user reach data is subjected to flow coloring by using a preset classification label to form a unified service reach classification standard system, specifically including:

[0044] Obtaining a user reach mode for each business line, and setting a first classification label according to the user reach mode;

[0045] Based on the first classification label, a naive Bayes algorithm is used to classify the user reach data to obtain a second classification label;

[0046] Based on the second classification label, the business party, scenario, trigger source, and contact method of the user contact data are marked, and a K-means algorithm is used to perform cluster analysis on the user contact data to obtain clustering results of different business lines;

[0047] Based on the clustering results, a unified business reach classification standard system is formed.

[0048] In this embodiment, a first classification label is set according to the user reach method of each business line, and the user reach data is preliminarily divided according to the first classification label. The naive Bayes algorithm is used to classify the user reach data based on the first classification label to obtain a more fine-grained second classification label. According to the second classification label, the business party, scenario, trigger source and reach method of the user reach data are marked to prepare for subsequent clustering analysis. The marked user reach data is used as input, and the K-means algorithm is used for clustering analysis to obtain clustering results for different business lines. Analyze the clustering results, identify the commonalities and differences in user reach of different business lines, and extract key features and rules. Based on the clustering results and the extracted key features, a unified business reach classification standard is formulated to form a standardized classification system. The business reach classification standard system is applied to the user reach practice of each business line to guide business operations and decision-making, and improve the accuracy and effectiveness of user reach.

[0049] Exemplarily, first obtain the user contact methods of each business line, such as email, SMS, in-app message, website notification, etc., and set the first classification label according to these contact methods, such as marking email as "mail" and SMS as "SMS". Then, based on the first classification label, the naive Bayes algorithm is used to classify the user contact data to obtain the second classification label, such as subdividing "mail" into "promotional email", "notification email", etc., and subdividing "SMS" into "verification code SMS", "notification SMS", etc. The naive Bayes algorithm predicts the category of new data by calculating the probability of different features appearing in each category. For example, if a piece of user contact data contains words such as "discount" and "discount", it is more likely to be classified as "promotional email". Then, based on the second classification label, the business party, scenario, trigger source and contact method of the user contact data are marked, and the K-means algorithm is used to perform cluster analysis on the user contact data. The K-means algorithm continuously updates the cluster center through iteration and assigns the data points to the nearest cluster center until the clustering results converge. Different feature combinations can be selected during the clustering process, such as the combination of business parties and trigger sources, the combination of scenarios and contact methods, etc., to obtain clustering results of user contact behaviors in different dimensions. Finally, based on the clustering results, a unified business contact classification standard system is formed, such as dividing promotional contacts into new user discounts, holiday promotions, etc., and dividing notification contacts into logistics notifications, account arrival reminders, etc., to standardize and guide the user contact strategies and practices of each business line.

[0050] S102: Based on the business reach classification standard system, perform personalized reach behavior standard processing for each business line.

[0051] In this embodiment, based on the unified reach classification standards and in combination with business characteristics, corresponding user reach management specifications and processes are formulated to standardize and constrain the reach behaviors of each business line, and improve business reach efficiency and accuracy through process automation and intelligence.

[0052] In some embodiments, in the above step S102, the personalized contact behavior standard processing is performed on each business line based on the business contact classification standard system, specifically including:

[0053] Based on the business reach classification standard system, a clustering algorithm is used to analyze the business characteristic data of each business line to obtain reach management specification data and reach process templates for different business lines;

[0054] Use text classification algorithms to analyze the historical reach data of each business line and extract the reach behavior feature data;

[0055] Establishing a behavior specification model through a logistic regression algorithm according to the reach management specification data, the reach process template and the reach behavior feature data;

[0056] Based on the behavior specification model, personalized contact behavior specification processing is performed on each business line.

[0057] In this embodiment, according to the business contact classification standard system, a clustering algorithm is used to analyze the business characteristic data of each business line to obtain the contact management specification data of different business lines. The contact management specification data obtained by cluster analysis is used to determine the standardized contact process template of each business line. The text classification algorithm is used to analyze the historical contact data of each business line and extract the contact behavior feature data. The contact management specification data, the contact process template and the contact behavior feature data are preprocessed, including data cleaning, feature selection and data standardization operations to obtain a standardized data set. According to the preprocessed data set, the contact behavior feature data is modeled using a logistic regression algorithm, and a behavior specification model is obtained through training. In the modeling process, appropriate feature combinations and model parameters are selected to ensure the accuracy and generalization ability of the model. For the contact behavior feature data of different business lines, the established behavior specification model is used for prediction and evaluation. According to the results of the model output, it is judged whether the contact behavior meets the specification requirements, and corresponding specification processing suggestions are given. For the contact behavior that does not meet the specification requirements, a personalized contact behavior specification processing solution is generated based on the contact management specification data and the contact process template. The plan should include specific regulatory adjustment measures and implementation steps to ensure that the contact behavior complies with management regulations. Apply personalized contact behavior regulatory processing solutions to the corresponding business lines to optimize and adjust the contact process. Through real-time monitoring of contact behavior data, dynamically evaluate the regulatory processing effect, and iteratively optimize the solution based on feedback results. During the regulatory processing process, collect and analyze contact behavior data to obtain user feedback and business effect indicators. Use these data as new training samples to regularly update and optimize the behavioral regulatory model to improve the adaptability and accuracy of the model. Establish a contact behavior regulatory management platform that integrates behavioral regulatory models and personalized processing solutions. Through visual interfaces and automated tools, real-time monitoring of contact behaviors, regulatory evaluation, and processing solution generation are achieved to provide intelligent regulatory management services for business lines.

[0058] Exemplarily, after obtaining a unified business contact classification standard system, the K-means clustering algorithm is used to analyze the business characteristic data of each business line. First, extract the user contact frequency, contact time period, contact content and other features of each business line to construct a feature vector. Then, use the Euclidean distance as the similarity measure, set the number of clusters to 5, and iteratively optimize the cluster center until the clustering results converge. Finally, obtain the contact management specification data and contact process templates of different business lines, such as the contact frequency of high-frequency contact business lines should be controlled within 3 times a day, and the contact time is concentrated between 8 and 10 pm. Then, the FastText text classification algorithm is used to analyze the historical contact data of each business line. Extract the TF-IDF features and N-gram features of the contact text, construct a text feature vector, and use the SoftMax function for multi-classification training to extract the contact behavior feature data, such as promotional contact texts usually contain keywords such as "discount" and "discount", and the contact time is mostly around holidays. Based on the contact management specification data, contact process templates, and contact behavior feature data, a behavior specification model is established through a logistic regression algorithm. The business line characteristics, contact characteristics, and compliance with the specifications are used as the input variables of the model, and the logistic regression model is trained to obtain the weight of the impact of each feature on the contact behavior specification. Finally, based on the behavior specification model, personalized contact behavior specification processing is performed on each business line. For promotional business lines with high frequency of contact, the contact frequency is appropriately increased during holidays, and the discount strength is highlighted in the contact text; for service notification business lines with low frequency of contact, contact information is sent at noon on weekdays, and the text content emphasizes the service expiration reminder to improve user satisfaction.

[0059] S103, based on the business touch classification standard system, perform correlation analysis on the abnormal touch events occurring in the touch process and the marking information of each link in the touch process to determine the abnormal causes and handling strategies of the abnormal touch events.

[0060] In this embodiment, in the game operation, abnormal touch events may include but are not limited to: users not receiving push notifications, push content errors, inappropriate push time causing user disgust, etc. These events will have a negative impact on user experience and even cause user loss. In order to effectively manage these abnormal events, it is necessary to define and classify abnormal touch events, such as dividing them according to factors such as the frequency of occurrence, scope of impact, and severity of the event.

[0061] In the process of reaching out, game companies need to mark each link for subsequent correlation analysis. These marking information includes: reaching method: such as APP push, SMS, email, etc.; reaching time: such as the specific time point or time period of push; user feedback: such as whether the user clicks, whether he participates in the activity, etc.

[0062] Based on the business touch classification standard system and the marking information of the touch process, game companies can conduct correlation analysis on abnormal touch events, including: collecting marking information of each link in the touch process and abnormal touch event data; cleaning the collected data to remove invalid or redundant information; using data mining or machine learning algorithms to conduct correlation analysis on the cleaned data to find the correlation between abnormal touch events and each link in the touch process; and determining the abnormal cause of the abnormal touch event based on the results of the correlation analysis.

[0063] According to the abnormal reason, the game company can formulate corresponding disposal strategies. For example, if the abnormal reason is the failure of the push system, the push system can be optimized or the push service provider can be changed. If the abnormal reason is that the push content does not meet the user's interests, the push content strategy can be optimized to improve the relevance and attractiveness of the content. If the abnormal reason is that the push time is inappropriate, the push time strategy can be adjusted to avoid push during the user's inactive time period.

[0064] In this embodiment, through the formulation and implementation of correlation analysis and disposal strategies, abnormal touch events can be more accurately identified and resolved, thereby improving user experience. By analyzing and handling abnormal touch events, push strategies can be continuously optimized to improve push effects and user engagement.

[0065] In some embodiments, in the above step S103, based on the business reach classification standard system, the abnormal reach event occurring in the reach process and the tag information of each link of the reach process are correlated and analyzed to determine the abnormal cause and handling strategy of the abnormal reach event, specifically including:

[0066] Based on the business access classification standard system, key information of each link in the access process is marked and encoded to form marked information;

[0067] Obtain a historical abnormal touch event data set, and use the FP-growth algorithm to analyze the correlation between each abnormal event and the marking information of each link in the historical abnormal touch event data set to form abnormal touch correlation information;

[0068] According to the abnormal contact related information, determine the key links and abnormal causes that trigger each abnormal contact event;

[0069] According to the key links and the abnormal causes, the optimal handling strategy is determined from a preset strategy library.

[0070] In this embodiment, according to the business touch classification standard system, the key information of each link in the touch process is obtained, and standardized marking information data is generated through information marking and coding processing. The historical abnormal touch event data set is obtained, the data set is preprocessed, and the noise data and invalid data are eliminated to obtain the cleaned historical abnormal touch event data set. The FP-growth algorithm is used to analyze the cleaned historical abnormal touch event data set, and the frequent patterns and association rules between each abnormal event and the marking information of each link are mined to obtain abnormal touch association information. According to the association information, the key link that causes the abnormal touch event is determined through the association rule mining algorithm. The root cause analysis is performed on the key link, and the decision tree algorithm is used to determine the cause of the abnormal touch event. According to the abnormal cause, the disposal strategy most similar to the abnormal cause is selected from the preset disposal strategy library through the similarity matching algorithm. If multiple disposal strategies with high similarity are matched, the optimal strategy is selected from them according to the historical data through the reinforcement learning algorithm. The optimal disposal strategy is applied to the business process involved in the abnormal touch event, and the disposal operation defined by the strategy is automatically executed through the workflow engine. Track the execution effect of the disposal strategy. If it does not meet expectations, return to reselect the disposal strategy. If the disposal is effective, mark the abnormal touch event as resolved and update the strategy library and knowledge base.

[0071] Exemplarily, based on the business contact classification standard system, the key information in the contact process is marked and encoded. Regular expressions are used to match keywords in the contact information, such as "coupon", "discount", "free", etc., to extract the features of promotional contact, and these features are one-hot encoded to form the marking information of the contact link. At the same time, a historical abnormal contact event data set is collected, including 5,000 abnormal samples, each of which contains the marking information of the contact link and the abnormal type. The FP-growth algorithm is used to mine the frequent patterns of the abnormal event data set, setting the minimum support to 05 and the minimum confidence to 8, to obtain the association rules between abnormal events and the contact link, and to form abnormal contact association information. Through the association rules, the key links and causes of the abnormality are determined, such as excessive contact frequency leading to user complaints, inappropriate contact content leading to user disgust, etc. According to the key links and abnormal causes, the optimal handling strategy is matched from the preset strategy library. The strategy library contains 100 common abnormal situations and their corresponding processing solutions. By calculating the similarity between the abnormal situation and each strategy in the strategy library, the strategy with the highest similarity is selected as the optimal strategy. Finally, the optimal strategy is applied to the business contact process, and the identified abnormal contact events are handled in real time, such as reducing the contact frequency, adjusting the contact content, etc., to improve the contact effect and reduce customer churn rate.

[0072] S104: Analyze the reach correlation between different business lines using association rule mining algorithms, and optimize the reach strategy by identifying possible repeated reach situations and excessive interruptions.

[0073] In this embodiment, the reach data of different business lines are collected from the game operation system, including reach time, reach method (such as email, SMS, APP push, etc.), reach content, user ID, etc. Then, the data is preprocessed, such as removing duplicate data, processing missing values, converting data types, etc., to ensure the quality and consistency of the data. The preprocessed data is mined using an association rule mining algorithm (such as the Apriori algorithm). Set appropriate support and confidence thresholds to filter out meaningful association rules. For example, the support threshold can be set to 0.05 (indicating that the rule appears in at least 5% of transactions) and the confidence threshold can be set to 0.7 (indicating that the rule is established in 70% of cases). Through mining, association rules such as "If a user receives an email push from business line A, there is a 75% probability that the user will also receive a SMS push from business line B" can be obtained. These rules reveal the correlation between the reach of different business lines. According to the mined association rules, possible repeated reach and excessive disturbance situations can be identified. For example, if a user frequently receives push notifications from multiple business lines at the same time, and the content of these push notifications is similar or repeated, it can be considered that there is repeated contact. If a user receives a large number of push notifications in a short period of time, it can be considered that there is excessive disturbance.

[0074] Based on the identified repeated reach and excessive interruptions, the reach strategy can be optimized. For example, for repeated reach, similar or repeated push content can be merged to reduce the number of pushes and avoid users feeling redundant and bored. For excessive interruptions, the push frequency and time can be adjusted to avoid pushes during user inactivity or rest time. At the same time, a user feedback mechanism can be introduced to allow users to independently choose which business lines to receive pushes.

[0075] Implement the optimized reach strategy into the game operation system and continuously monitor the reach effect and user feedback. Based on the monitoring results, continuously adjust and optimize the reach strategy to ensure its effectiveness and adaptability.

[0076] In this embodiment, by optimizing the reach strategy, repeated reach and excessive interruption are reduced, and user satisfaction and loyalty are improved. The association rule mining algorithm can quickly identify the reach correlation between different business lines, providing strong support for optimizing the reach strategy and improving operational efficiency.

[0077] In some embodiments, in the above step S104, the use of an association rule mining algorithm to analyze the reach correlation between different business lines and optimize the reach strategy by identifying possible repeated reach situations and excessive interruptions specifically includes:

[0078] Obtain user contact records for each business line to form a contact log data set;

[0079] By extracting features from the reach log data set, counting key features within a preset time window, and constructing a user reach feature data set, the key features include reach frequency, reach channel distribution, and reach content theme;

[0080] According to the preset support threshold and confidence threshold, an association rule mining algorithm is used to analyze the user reach feature data set to determine the association rules between different business lines and reach elements;

[0081] constructing an evaluation index for each association rule according to the key features, and determining a comprehensive risk degree of each association rule according to the evaluation index;

[0082] By comparing the comprehensive risk with a preset comprehensive risk threshold, it is determined whether the contact mode corresponding to each association rule will lead to repeated contact and excessive interruption, thereby forming a contact mode analysis result;

[0083] According to the reach pattern analysis results, the reach strategy is optimized.

[0084] In this embodiment, the user reach records of each business line are obtained to form a reach log data set. According to the preset time window, feature extraction is performed on the reach log data set, and key features such as reach frequency, reach channel distribution, and reach content theme are counted to construct a user reach feature data set. According to the preset support threshold and confidence threshold, the Apriori association rule mining algorithm is used to analyze the user reach feature data set to obtain a frequent item set. Based on the frequent item set, the association rules between different business lines and reach elements are determined by calculating the confidence. If the confidence of the association rule is greater than or equal to the preset confidence threshold, the association rule is added to the association rule set. According to the association rule set, the correlation between the user reach features of different business lines is determined, and the key reach element combination is identified.

[0085] For each rule, a comprehensive risk assessment index system is constructed based on key features such as the frequency of contact, distribution of contact channels, and the theme of contact content. The index system includes repeated contact risk indicators and excessive disturbance risk indicators. The weighted summation method is used to weight the scores of each risk indicator to obtain the comprehensive risk of each association rule. The preset comprehensive risk threshold is obtained from the rule base, and the comprehensive risk of each rule is compared with the threshold. If the comprehensive risk exceeds the threshold, it is judged that the contact mode corresponding to the association rule has the risk of repeated contact or excessive disturbance, and it is added to the high-risk contact mode set. The patterns in the high-risk contact mode set are analyzed to find out the key factors that cause risks and form contact mode optimization suggestions. Adjust the original contact strategy according to the optimization suggestions to reduce the risk of repeated contact and excessive disturbance, and improve the accuracy of contact and user experience.

[0086] Exemplarily, first obtain the user contact records of each business line from the business system, including contact time, contact channel, contact content and other information, to form a contact log data set. Perform feature extraction on the contact log data set, count the key features within the preset time window (such as the last month), including contact frequency, contact channel distribution and contact content theme, and construct a user contact feature data set. For example, count the number of times a user is contacted in the last month as the contact frequency feature, calculate the proportion of the number of times contacted by different channels (such as SMS, email, APP push, etc.) as the contact channel distribution feature, and perform a topic model (such as LDA) analysis on the contact content to extract the theme as the contact content theme feature. Then, set the minimum support threshold to 05 and the minimum confidence threshold to 8, and use the Apriori association rule mining algorithm to analyze the user contact feature data set to obtain the association rules between different business lines and contact elements, such as "If the user of the credit card business line has a contact frequency of more than 10 times in the last month, and the proportion of SMS channel contact exceeds 80%, it is easy to cause user complaints." Next, evaluation indicators are constructed for each association rule based on key features, including the frequency of contact indicator, channel distribution indicator, and content theme indicator. The rationality of the association rules is evaluated from the three dimensions of frequency of contact, channel distribution, and content theme, and weights are set based on expert experience to calculate the comprehensive risk. For example, the frequency of contact indicator can be set to indicate that the higher the frequency of contact, the greater the risk; the channel distribution indicator can be set to indicate that the proportion of contact through a single channel is too high, and the content theme indicator can be set to indicate that the risk of contacting a specific theme too many times is too high. Finally, the weighted sum of the three indicators is used to obtain the comprehensive risk, ranging from 0 to 100. By comparing the comprehensive risk with the preset comprehensive risk threshold (such as 80), it is determined whether the contact mode corresponding to each association rule will lead to repeated contact and excessive disturbance, and the contact mode analysis results are formed. Finally, according to the contact mode analysis results, the contact strategy is optimized, such as reducing the contact frequency for high-risk contact modes, adjusting the distribution of contact channels and content themes, so as to improve the contact effect and reduce the customer complaint rate.

[0087] S105, determining the reach preference and acceptance level of each user by modeling and analyzing the user behavior data, and optimizing the reach strategy based on the reach preference and the acceptance level.

[0088] In this embodiment, modeling and analyzing user behavior data to determine each user's reach preference and acceptance level, and optimizing the reach strategy accordingly, is a key means to improve user experience and operational efficiency in game operations. First, user behavior data is comprehensively collected from the game system. These data should cover various interactive behaviors of users in the game, such as login time, game duration, click behavior, purchase records, etc. Next, the data is preprocessed, including data cleaning, format conversion and standardization processing to ensure the quality and consistency of the data. Based on the preprocessed user behavior data, a user behavior model is constructed. This model can be a complex machine learning model, such as a deep learning model or a random forest model, which is used to capture potential patterns and trends in user behavior. The goal of the model is to identify the user's reach preferences, such as which type of push messages (such as game updates, event notifications, discount information, etc.) are preferred, and the user's acceptance of different reach methods (such as email, SMS, APP push, etc.). Using the constructed user behavior model, the reach preference and acceptance of each user are quantitatively analyzed. This can be achieved by calculating indicators such as the click-through rate and conversion rate of users for different types of push messages. At the same time, you can also analyze the user's acceptance level in different time periods and situations to obtain a more comprehensive user portrait.

[0089] Optimize the reach strategy based on the user's reach preferences and acceptance level. Specifically, the following measures can be taken: According to the user's reach preferences, push customized message content to increase the attractiveness of the message and the user's click-through rate. According to the user's acceptance level, intelligently adjust the push time and frequency to avoid pushing during the user's inactive or resting time to reduce interruptions and improve user experience. Combined with the user's reach preferences and acceptance level, select the most appropriate push channel, such as email, SMS or APP push, to improve the message arrival rate and user satisfaction.

[0090] Implement the optimized reach strategy into the game operation system and continuously monitor its effect. By collecting user feedback and observing changes in user behavior data, continuously adjust and optimize the reach strategy to ensure its adaptability and effectiveness.

[0091] In this embodiment, through personalized push and intelligent scheduling, unnecessary interruptions and redundant information are reduced, and user satisfaction and loyalty are improved. The optimized reach strategy can reach target users more accurately, improve the message arrival rate and conversion rate, thereby reducing operating costs and improving operating efficiency.

[0092] In some embodiments, in the above step S105, determining each user's reach preference and acceptance level by modeling and analyzing the user behavior data specifically includes:

[0093] Obtain various types of user behavior data in the game and build a user behavior log data set;

[0094] Obtain the historical reach feedback data corresponding to each user, and build a user behavior prediction model using a decision tree algorithm or a logistic regression algorithm based on the user behavior log data set and the historical reach feedback data;

[0095] Based on the user behavior prediction model, determine each user's reach preference and acceptance level.

[0096] In this embodiment, various types of user behavior data in the game are obtained, including login time, online time, activity, consumption amount, etc., to construct a user behavior log data set; historical contact feedback data corresponding to each user is obtained, including click-through rate and conversion rate of pushed information, etc., and associated with the user behavior log data set; a decision tree algorithm is used to train a user behavior prediction model based on user behavior data and historical contact feedback data to obtain classification rules for predicting the user's acceptance of contact information; if the prediction effect of the decision tree algorithm is not good, a logistic regression algorithm is used to train a user behavior prediction model based on user behavior data and historical contact feedback data to obtain a probability model for predicting the user's acceptance of contact information; based on the trained user behavior prediction model, the behavior data of each user is predicted to obtain the user's acceptance probability of different types of contact information; based on the user's acceptance probability of different types of contact information, the contact preference of each user is determined, and the contact type with an acceptance probability higher than a preset threshold is used as the user's preferred contact method; based on each user's contact preference, when pushing information to the user, the contact method preferred by the user is given priority to improve the user's acceptance and the contact effect.

[0097] Exemplary, first obtain various types of user behavior data from the game server log, including login time, online time, number of completed levels, amount of props purchased, and other information, and obtain user feedback data on historical reach activities from the reach feedback log, such as push click-through rate, participation activity conversion rate, etc., and aggregate these data to form a user behavior log data set and a historical reach feedback data set. Then, select the CART classification tree model in the decision tree algorithm, use the features in the user behavior log data set as input, and use the user feedback in the historical reach feedback data set as the output label to train the user behavior prediction model. During the model training process, grid search is used to optimize the key parameters of the decision tree, such as the maximum depth of the tree, the minimum number of samples of the leaf node, etc., and the model performance is evaluated by 5-fold cross validation, and the parameter combination with the highest F1 value is selected. Then, use the trained CART classification tree model to predict each user, judge the user's acceptance of different types of reach activities according to the user's behavioral characteristics, obtain the user's reach preference, and estimate the possibility of the user accepting each type of activity according to the classification probability of the leaf node. For example, for user A, the model predicts that his acceptance of mall promotions is 80%, his acceptance of game activities is 60%, and his acceptance of social sharing is 40%. Therefore, user A's contact preference is determined to be mall promotions > game activities > social sharing. Finally, the user's contact preference and acceptance level are used as the basis for subsequent personalized contact. When formulating contact strategies, the contact frequency, contact content, and reward intensity are set for different users to improve the accuracy and conversion rate of contact.

[0098] S106, evaluating the reaching effects in different scenarios based on the key indicator data obtained in each link of the reaching process, and optimizing the reaching strategy based on the reaching effects.

[0099] In this embodiment, data scoring technology is used to collect and record data at each link of the reach link to obtain key data indicators in the reach process, such as reach time, reach object, reach content, etc. By performing multi-dimensional analysis on the collected data, the reach effect in different scenarios is evaluated, and the reach methods with better and worse effects are identified to provide data support for business decisions.

[0100] In some embodiments, in the above step S106, evaluating the reach effects in different scenarios based on the key indicator data obtained in each link of the reach process, and optimizing the reach strategy based on the reach effects, specifically includes:

[0101] Obtain key indicator data for each link in the reaching process, including reaching time, reaching object, reaching content, and reaching event;

[0102] Setting a reach dimension, wherein the reach scenario dimension includes a scenario dimension, an object dimension, and a content dimension;

[0103] Aggregate the key indicator data and the reach dimensions and divide them into scenarios to obtain multiple reach scenarios;

[0104] The association rule mining algorithm is used to set the minimum support and minimum confidence for each contact scenario, and to mine the frequent item sets and association rules between contact time, contact object, contact content and contact event to form association results;

[0105] According to the association results, the reach effects in different reach scenarios are evaluated to obtain reach effect evaluation data, and the reach strategy is optimized through the reach effect evaluation data.

[0106] In this embodiment, the original data of each link in the reaching process is obtained, and key indicator data such as reaching time, reaching object, reaching content and reaching event are extracted. According to the preset reaching dimension model, multiple reaching dimensions such as scene dimension, object dimension and content dimension are set. The key indicator data are classified and labeled according to the reaching dimension to obtain indicator data with dimension labels. A clustering algorithm is used to aggregate the indicator data with dimension labels, and the data is divided into multiple clusters according to the similarity of the data. For each cluster, the characteristics of its internal data are analyzed, and the key features that can represent the cluster are extracted as the basis for the division of the reaching scene. It is judged whether the data volume of each cluster exceeds the preset threshold. If it exceeds, the cluster is determined as an independent reaching scene. The data of each reaching scene is further analyzed, and information such as scene portrait, object portrait and content portrait is extracted to complete the scene division.

[0107] For different reach scenarios, set the corresponding minimum support and minimum confidence thresholds to filter frequent item sets and association rules. Adopt the Apriori algorithm to mine frequent item sets between reach time, reach object, reach content and reach event according to the set minimum support and minimum confidence. Generate association rules between reach time, reach object, reach content and reach event based on the mined frequent item sets to obtain association results. According to the association results, statistically analyze reach effect indicators under different reach scenarios, such as conversion rate, click-through rate, etc., to obtain reach effect evaluation data. Adopt the decision tree algorithm, take the reach scenario as the decision node, the association rule as the branch condition, and the reach effect evaluation data as the target variable to generate a reach strategy optimization decision tree. According to the branch condition and node threshold of the decision tree, perform scenario judgment and rule matching on the new reach task to obtain the optimized reach time, reach object, and reach content combination to form a dynamically optimized reach strategy.

[0108] For example, first obtain the key indicator data of the contact process from the tracking log of the game server, including the contact time, the user ID of the contact object, the activity ID of the contact content, and the contact events such as clicks, participation, etc., and set the scene dimension, object dimension and content dimension as the contact dimension. Then use Spark to aggregate the key indicator data and the contact dimension, and group and count the data according to different dimension combinations, such as dividing different contact scenes according to login events, activity pages, etc. in the scene dimension. Then use the FP-Growth association rule mining algorithm, set the minimum support to 05 and the minimum confidence to 8, and mine frequent item sets and association rules in different contact scenes. It is found that when reaching new users on the activity page, the click rate can reach 5% and the participation rate is 2%, forming an associated result of the contact effect. Finally, based on key indicators in the associated results, such as click-through rate and participation rate, the effects of different reach scenarios are evaluated, and the average click-through rate, participation rate and other reach effect evaluation data of each scenario are obtained. The reach strategy is optimized according to the evaluation results. For example, for new user reach on the activity page, the reach frequency can be appropriately increased, and the copywriting and creativity of the reach content can be optimized to further improve the reach effect.

[0109] Reference Figure 2 An embodiment of the present invention provides a game user reach classification management system 2, and the system 2 specifically includes:

[0110] The first management module 201 is used to obtain user reach data of each business line, and perform flow coloring on the user reach data using preset classification labels to form a unified business reach classification standard system;

[0111] A second management module 202 is used to perform personalized contact behavior standard processing on each business line based on the business contact classification standard system;

[0112] The third management module 203 is used to perform correlation analysis on the abnormal contact events occurring in the contact process and the tag information of each link of the contact process based on the business contact classification standard system, and determine the abnormal cause and handling strategy of the abnormal contact event;

[0113] The fourth management module 204 is used to analyze the reach correlation between different business lines by using the association rule mining algorithm, and optimize the reach strategy by identifying possible repeated reach situations and excessive interruptions;

[0114] The fifth management module 205 is used to determine the reach preference and acceptance level of each user by modeling and analyzing the user behavior data, and optimize the reach strategy according to the reach preference and the acceptance level;

[0115] The sixth management module 206 is used to evaluate the reaching effects in different scenarios based on the key indicator data obtained in each link of the reaching process, and optimize the reaching strategy based on the reaching effects.

[0116] It is understandable that if Figure 1 The contents of the game user reach classification management method embodiment shown in the figure are all applicable to the game user reach classification management system embodiment of the present game. The functions specifically implemented by the game user reach classification management system embodiment are similar to those in the present game. Figure 1 The game user reach classification management method embodiment shown is the same as that shown in the embodiment, and the beneficial effects achieved are the same as those shown in the embodiment. Figure 1 The beneficial effects achieved by the game user reach classification management method embodiment shown are also the same.

[0117] It should be noted that the information interaction, execution process and other contents between the above-mentioned systems are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0118] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0119] Reference Figure 3 The embodiment of the present invention further provides a computer device 3, comprising: a memory 302 and a processor 301 and a computer program 303 stored in the memory 302. When the computer program 303 is executed on the processor 301, a game user reach classification management method as described in any one of the above methods is implemented.

[0120] The computer device 3 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 3 may include, but is not limited to, a processor 301 and a memory 302. Those skilled in the art will appreciate that Figure 3It is only an example of computer device 3 and does not constitute a limitation on computer device 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components, for example, it may also include input and output devices, network access devices, etc.

[0121] The processor 301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0122] In some embodiments, the memory 302 may be an internal storage unit of the computer device 3, such as a hard disk or memory of the computer device 3. In other embodiments, the memory 302 may also be an external storage device of the computer device 3, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device 3. Further, the memory 302 may also include both an internal storage unit of the computer device 3 and an external storage device. The memory 302 is used to store an operating system, an application program, a boot loader (Boot Loader), data, and other programs, such as the program code of the computer program, etc. The memory 302 may also be used to temporarily store data that has been output or is to be output.

[0123] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the game user reach classification management method as described in any one of the above methods is implemented.

[0124] In this embodiment, if the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0125] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0126] 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.

[0127] In the embodiments disclosed in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or 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.

[0128] 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.

Claims

1. A method for managing game user reach classification, characterized in that: The method specifically comprises: Obtain user reach data for each business line, and color the user reach data using pre-set classification labels to form a unified business reach classification standard system; Based on the business contact classification standard system, perform personalized contact behavior standard processing for each business line; Based on the business contact classification standard system, the abnormal contact events occurring during the contact process and the tag information of each link of the contact process are correlated and analyzed to determine the abnormal causes and handling strategies of the abnormal contact events; Use association rule mining algorithms to analyze the reach correlation between different business lines, and optimize reach strategies by identifying possible repeated reach and excessive interruptions. By modeling and analyzing user behavior data, the reach preference and acceptance level of each user are determined, and the reach strategy is optimized according to the reach preference and acceptance level; Based on the key indicator data obtained from each link in the reaching process, the reaching effect in different scenarios is evaluated, and the reaching strategy is optimized based on the reaching effect.

2. The method according to claim 1, characterized in that: The user reach data is subjected to flow coloring by using the preset classification labels to form a unified service reach classification standard system, specifically including: Acquire a user reach mode for each business line, and set a first classification label according to the user reach mode; Based on the first classification label, a naive Bayes algorithm is used to classify the user reach data to obtain a second classification label; Based on the second classification label, the business party, scenario, trigger source, and contact method of the user contact data are marked, and a K-means algorithm is used to perform cluster analysis on the user contact data to obtain clustering results of different business lines; Based on the clustering results, a unified business reach classification standard system is formed.

3. The method according to claim 1, characterized in that: Based on the business contact classification standard system, the personalized contact behavior standard processing is performed on each business line, specifically including: Based on the business reach classification standard system, a clustering algorithm is used to analyze the business characteristic data of each business line to obtain reach management specification data and reach process templates for different business lines; Use text classification algorithms to analyze the historical reach data of each business line and extract the reach behavior feature data; Establishing a behavior specification model through a logistic regression algorithm according to the reach management specification data, the reach process template and the reach behavior feature data; Based on the behavior specification model, personalized contact behavior specification processing is performed on each business line.

4. The method according to claim 1, characterized in that: Based on the business contact classification standard system, the abnormal contact events occurring in the contact process and the tag information of each link of the contact process are correlated and analyzed to determine the abnormal causes and handling strategies of the abnormal contact events, specifically including: Based on the business access classification standard system, key information of each link in the access process is marked and encoded to form marked information; Obtain a historical abnormal touch event data set, and use the FP-growth algorithm to analyze the correlation between each abnormal event and the marking information of each link in the historical abnormal touch event data set to form abnormal touch correlation information; According to the abnormal contact related information, determine the key links and abnormal causes that trigger each abnormal contact event; According to the key links and the abnormal causes, the optimal handling strategy is determined from a preset strategy library.

5. The method according to claim 1, characterized in that The association rule mining algorithm is used to analyze the reach correlation between different business lines, and the reach strategy is optimized by identifying possible repeated reach and excessive interruption situations, including: Obtain user contact records for each business line to form a contact log data set; By extracting features from the reach log data set, counting key features within a preset time window, and constructing a user reach feature data set, the key features include reach frequency, reach channel distribution, and reach content theme; According to the preset support threshold and confidence threshold, an association rule mining algorithm is used to analyze the user reach feature data set to determine the association rules between different business lines and reach elements; constructing an evaluation index for each association rule according to the key features, and determining a comprehensive risk degree of each association rule according to the evaluation index; By comparing the comprehensive risk with a preset comprehensive risk threshold, it is determined whether the contact mode corresponding to each association rule will lead to repeated contact and excessive interruption, thereby forming a contact mode analysis result; According to the reach pattern analysis results, the reach strategy is optimized.

6. The method according to claim 1, characterized in that The user behavior data is modeled and analyzed to determine each user's reach preference and acceptance level, including: Obtain various types of user behavior data in the game and build a user behavior log data set; Obtain the historical reach feedback data corresponding to each user, and build a user behavior prediction model using a decision tree algorithm or a logistic regression algorithm based on the user behavior log data set and the historical reach feedback data; Based on the user behavior prediction model, determine each user's reach preference and acceptance level.

7. The method according to any one of claims 1 to 6, characterized in that The key indicator data obtained from each link in the reaching process is used to evaluate the reaching effects in different scenarios, and the reaching strategy is optimized based on the reaching effects, specifically including: Obtain key indicator data for each link in the reaching process, including reaching time, reaching object, reaching content, and reaching event; Setting a reach dimension, wherein the reach scenario dimension includes a scenario dimension, an object dimension, and a content dimension; Aggregate the key indicator data and the reach dimensions and divide them into scenarios to obtain multiple reach scenarios; The association rule mining algorithm is used to set the minimum support and minimum confidence for each contact scenario, and to mine the frequent item sets and association rules between contact time, contact object, contact content and contact event to form association results; According to the association results, the reach effects in different reach scenarios are evaluated to obtain reach effect evaluation data, and the reach strategy is optimized through the reach effect evaluation data.

8. A game user reach classification management system, characterized in that: The system specifically comprises: The first management module is used to obtain user reach data of each business line, and to color the user reach data with preset classification labels to form a unified business reach classification standard system; A second management module is used to perform personalized contact behavior standard processing on each business line based on the business contact classification standard system; The third management module is used to perform correlation analysis on abnormal contact events occurring during the contact process and tag information of each link of the contact process based on the business contact classification standard system, and determine the abnormal cause and disposal strategy of the abnormal contact event; The fourth management module is used to analyze the reach correlation between different business lines using association rule mining algorithms, and optimize the reach strategy by identifying possible repeated reach and excessive interruption situations; A fifth management module is used to determine the reach preference and acceptance level of each user by modeling and analyzing the user behavior data, and optimize the reach strategy according to the reach preference and acceptance level; The sixth management module is used to evaluate the reaching effects in different scenarios based on the key indicator data obtained in each link of the reaching process, and optimize the reaching strategy based on the reaching effects.

9. A computer device, characterized in that: include: A memory and a processor and a computer program stored in the memory, when the computer program is executed on the processor, implements the game user reach classification management method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the game user reach classification management method as described in any one of claims 1 to 7 is implemented.

Images