Game service management method and system based on Internet

Abstract

The invention discloses a game service management method and system based on the Internet, and relates to the technical field of the Internet. The method comprises the steps that player behavior data and game resource data are collected, integrated, cleaned, encrypted and stored, and the privacy of players is protected. Personalized recommendation and prediction are carried out by using the decision tree model, and the player experience is improved. Meanwhile, game resources are monitored in real time according to historical resource consumption data, and a supplementary mechanism is automatically triggered. A risk assessment model is established, abnormal behaviors are monitored, an early warning signal is sent, and the operation risk is reduced. The system comprises a data acquisition and processing module, a data encryption and desensitization module, a personalized recommendation prediction module, a game resource management module, a risk early warning module and the like, and functional modular design is realized. According to the method, the security and individuation level of game services are improved, reasonable allocation of resources is ensured, the operation risk is reduced, and safer and more convenient game experience is provided for players.

Description

Technical Field

[0001] The present invention belongs to the field of Internet technology, and in particular relates to a game service management method and system based on the Internet. Background Art

[0002] With the rapid development of Internet technology, the online game industry is booming and attracting a large number of players. However, with the continuous expansion of the online game market and the increasing complexity of games, game service management faces many challenges. Traditional management methods can no longer meet the needs of modern game operations. The shortcomings include:

[0003] Data sources are wide and in various formats, making it difficult to collect and integrate, and there is a large amount of duplicate, erroneous and abnormal data, which affects the accuracy of analysis. At the same time, players' sensitive data is easily leaked during storage and transmission, and data analysis without desensitization will also violate privacy.

[0004] The lack of accurate recommendations makes it difficult to push appropriate virtual props, activities, and social friends based on players’ personalized needs. In addition, due to the lack of effective prediction methods, it is impossible to accurately predict players’ future needs, resulting in an imbalance in resource allocation.

[0005] Resource allocation lacks scientific basis and does not take into account the differences in needs of different players, which can easily lead to excess or shortage of resources for some players. When resources are scarce, it is difficult to replenish them in time, affecting the smoothness of the game and player satisfaction.

[0006] Account theft and illegal trading of virtual items are common, disrupting the order of game operations. Existing management methods lack real-time monitoring and abnormal early warning mechanisms, making it difficult to handle abnormal behaviors in a timely manner, which can easily lead to increased losses.

[0007] To solve the above problems, this patent proposes an Internet-based game service management method and system, through multi-dimensional technologies such as real-time data collection and preprocessing, data encryption and desensitization, personalized recommendation and prediction, game resource management and risk warning. Summary of the invention

[0008] In order to overcome the above-mentioned shortcomings and deficiencies of the prior art, the first purpose of the present invention is to provide a game service management method based on the Internet; the second purpose of the present invention is to provide a game service management system based on the Internet.

[0009] The first object of the present invention adopts the following technical solution:

[0010] The process of Internet-based game service management method is as follows:

[0011] Step 1: Data collection and preprocessing: Collect all kinds of player behavior data and in-game resource data in real time, integrate and clean the collected data from different sources and in different formats, remove duplications, errors and outliers, store the cleaned and integrated data in a high-performance database using a distributed storage architecture, and store them in categories according to data type and frequency of use;

[0012] Step 2: Data encryption and desensitization: Encrypt and store players’ sensitive data and desensitize them, record access and operation behaviors to sensitive data, assign permissions based on player roles, use attribute-based access control, and implement strong password policies;

[0013] Step 3: Personalized recommendation and prediction: Build a decision tree model, and make personalized recommendations and predict players’ future gaming needs based on the output of the decision tree model;

[0014] Step 4: Game resource management: Analyze historical game resource consumption data based on the decision tree model, monitor in-game resource data, compare game resource inventory with preset amounts, and adjust resources based on the comparison results;

[0015] Step 5: Risk warning: Establish a risk assessment model, monitor abnormal behavior based on the model results, establish a risk warning level system and take corresponding measures.

[0016] Preferably, in the data collection and preprocessing steps, the player behavior data includes game login time, online time, game operation records, social interaction data, and consumption data; the in-game resource data includes the number of game coins, the attributes and quantity of virtual props, map scene information, and monster data.

[0017] Preferably, in the data encryption and desensitization steps, the AES encryption algorithm is used to encrypt and store player sensitive data including ID number, bank card information, and recharge records. Sensitive data is desensitized when the data is used. Corresponding permissions are assigned according to player roles, including ordinary players, guild administrators, and game anchors. Access rights are dynamically determined based on the player's geographic location and login time. Players are required to set complex passwords that contain numbers, letters, special characters and are of sufficient length and change them regularly.

[0018] As a preferred embodiment, the process of establishing a decision tree model in the personalized recommendation and prediction step is: inputting player data and in-game resource data into the decision tree model, taking 70% of the historical data as a training set and 30% as a test set; calculating the Gini coefficient G of each attribute based on the training set data, sorting the Gini coefficients from small to large, taking the attribute corresponding to the smallest Gini coefficient as the root node, and then calculating the Gini coefficients of the remaining attributes, taking the attribute corresponding to the minimum value as the internal node, and repeating the calculation until the remaining attributes are 0 to obtain the decision tree model to be tested; testing the decision tree model to be tested based on the test set data, and comparing the test accuracy Q with the preset test accuracy Q 0 Compare, when Q ≥ Q 0 When Q 0 When , the training set data is expanded to continue training until the conditions are met and then output.

[0019] Preferably, in the personalized recommendation and prediction steps, personalized recommendation is to recommend virtual props, game activities, and social friends in the game based on the player behavior patterns and resource association rules output by the decision tree model using a recommendation algorithm; demand prediction is to predict players' future demand for virtual props or game activities based on time series analysis combined with players' historical game behavior and consumption data.

[0020] Preferably, in the game resource management step, resource prediction and analysis is to analyze historical game resource consumption data based on a decision tree model, monitor in-game resource data, compare the game coin inventory K with the preset game coin inventory, and the virtual prop inventory with the preset virtual prop inventory, to determine whether the current game resources can meet future needs; resource adjustment is when the game coin inventory K is greater than or equal to the preset game coin inventory or the virtual prop inventory meets the preset requirements, it is determined that the resources can meet future needs, and when K is less than the preset game coin inventory or the virtual prop inventory is insufficient, the resource replenishment mechanism is triggered, and the game coins are replenished by adjusting the in-game task rewards and monster drops, and the virtual props are replenished by adjusting the output probability or from the reserve.

[0021] Preferably, in the risk warning step, a risk assessment model is established based on the players' gaming behavior patterns, consumption behaviors and social interactions, combined with in-game transaction rules and security policies to determine risk assessment indicators, and the indicators are input into the transaction risk assessment model to obtain results as a basis for monitoring abnormal behavior; risk warning and processing is to establish a risk warning level system based on the detected abnormal behavior, low-risk abnormal behavior may not send a warning signal but record the situation, and high-risk abnormal behavior immediately takes emergency measures and sends a warning signal.

[0022] The second object of the present invention adopts the following technical solution:

[0023] ​An Internet-based game service management system, which is used to implement an Internet-based game service management method, wherein the system includes a data collection and processing module, a data encryption and desensitization module, a personalized recommendation and prediction module, a game resource management module, and a risk warning module;

[0024] Data collection and processing module: used to collect various types of player behavior data in the game in real time, including game login time, online time, game operation records, social interaction data, consumption data, and in-game resource data; integrate and clean the collected data and store it in the database;

[0025] Data encryption and desensitization module: used to encrypt and store players' sensitive data, and perform desensitization processing during data use; assign corresponding permissions according to players' roles, and adopt attribute-based access control to improve account security;

[0026] Personalized recommendation prediction module: used to analyze player data by building a decision tree model, output player behavior patterns and resource association rules; use recommendation algorithms to make personalized recommendations for virtual props, game activities, and social friends in the game, and predict players' future game needs based on time series analysis;

[0027] Game resource management module: used to analyze historical game resource consumption data based on the decision tree model, monitor the resource data in the game in real time, and determine whether the current game resources can meet the game needs of future players. If the resources are insufficient, the resource replenishment mechanism is automatically triggered;

[0028] Risk warning module: used to establish a risk assessment model, analyze players' gaming behavior patterns, consumption behaviors and social interactions, and determine risk assessment indicators; establish a risk warning level system based on detected abnormal behaviors and send risk warning signals.

[0029] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0030] 1. The present invention achieves accurate analysis of player behavior patterns and resource association rules by establishing a decision tree model, thereby providing personalized recommendation services for players. At the same time, combined with time series analysis, this method can predict players' future game needs and provide powerful decision-making support for game operators. This innovative personalized recommendation and demand prediction mechanism not only improves the accuracy of recommendations, but also enhances players' gaming experience.

[0031] 2. The present invention realizes real-time monitoring and risk assessment of players' gaming behaviors, consumption behaviors and social interactions by establishing a risk assessment model. When abnormal behaviors are detected, the method can immediately trigger the risk warning mechanism and take emergency measures to deal with them. This innovative risk warning and processing capability not only improves the security of game services, but also effectively prevents potential security risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0033] Figure 1 A flow chart showing the Internet-based game service management method of the present invention is shown;

[0034] Figure 2 A module diagram of the Internet-based game service management system of the present invention is shown. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0036] In addition, the described features, structures or characteristics may be combined in one or more example embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the example embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, steps, etc. may be adopted. In other cases, well-known structures, methods, implementations or operations are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0037] Embodiment 1:

[0038] See also Figure 1 As shown, the Internet-based game service management method of this embodiment has the following process:

[0039] Step 1: Data collection and preprocessing.

[0040] Real-time collection of various behavioral data of players in the game, including game login time, online time, game operation records (such as skill release, character movement, prop use, etc.), social interaction data (friend addition, team chat, guild communication, etc.), consumption data (recharge amount, type and quantity of virtual props purchased).

[0041] Collect in-game resource data, such as the number of game coins, the attributes and quantity of virtual props, map scene information, monster data, etc.

[0042] Integrate the collected data from different sources and in different formats and unify the data format to facilitate subsequent analysis and processing.

[0043] Clean the collected data to remove duplicate data, erroneous data and outliers. For example, check the game login time records and remove obviously unreasonable timestamps (such as login time earlier than the game server opening time); verify and correct the abnormal number of skill releases in the game operation records (such as the number of skill releases in a short period of time far exceeding the normal range).

[0044] The cleaned and integrated data is stored in a high-performance database, using a distributed storage architecture to ensure data security and scalability. It can be classified and stored according to data type and frequency of use. For example, the basic information and common operation data of players can be stored in a relational database for quick query and update; historical game records and large-scale social interaction data can be stored in a non-relational database to adapt to the rapid growth of data volume and complex data structure.

[0045] Step 2: Data encryption and desensitization.

[0046] Players' sensitive data, such as ID number, bank card information, recharge records, etc., are encrypted and stored, using advanced encryption algorithms (such as AES encryption algorithm) to ensure the security of data during storage and transmission.

[0047] During data use, sensitive data is desensitized, for example, the middle digits of the ID number are replaced with asterisks, and only the first and last digits of the bank card number are displayed, which not only meets the needs of data analysis, but also protects the privacy of players. At the same time, all access and operation behaviors to sensitive data are recorded in detail, including access time, visitor identity, operation content, etc., for traceability and auditing.

[0048] The corresponding permissions are assigned according to the role of the player in the game (such as ordinary players, guild administrators, game anchors, etc.). Ordinary players can only access their own game data and basic game functions; guild administrators can manage guild member information, organize guild activities, etc.; game anchors can obtain game data and permissions related to live broadcasting.

[0049] Attribute-based access control is used to dynamically determine access rights based on the player's geographic location and login time. For example, players are restricted from performing certain high-risk operations (such as large recharges and sensitive item transactions) in specific areas or during specific time periods to prevent account theft or illegal transactions.

[0050] Implement a strong password policy, requiring players to set complex passwords that contain numbers, letters, special characters and are of sufficient length, and remind players to change their passwords regularly to improve account security.

[0051] Step 3: Personalized recommendations and predictions.

[0052] S31. Establish a decision tree model.

[0053] Player data (such as game preferences, consumption habits, social relationships, etc.) and in-game resource data (such as virtual item attributes, game level difficulty, etc.) are input into the decision tree model. When building the decision tree model, 70% of the historical data is divided into a training set and 30% is divided into a test set.

[0054] According to the training set data, the Gini coefficient G of each attribute is calculated, k is set as the total number of categories, i is the category, P(i) is the probability of label=i, and the attributes are sorted in order from small to large according to the Gini coefficient G, and the attribute corresponding to the minimum Gini coefficient G is used as the root node. Then the Gini coefficient of the remaining attributes after the root node is set is calculated, and the attribute corresponding to the minimum Gini coefficient of the remaining attributes is used as the internal node, and the calculation is repeated until the remaining attributes are 0, and the decision tree model to be tested is obtained.

[0055] The decision tree model to be tested is tested according to the test set data, and the test accuracy Q is compared with the preset test accuracy Q 0 When Q ≥ Q 0 When Q 0 When , it is determined that the expanded training set data continues to train the decision tree model to be tested until the test accuracy Q is greater than or equal to the preset test accuracy Q 0 After the judgment conditions are met, the decision tree model to be tested is output as a decision tree model.

[0056] S32. Personalized recommendations.

[0057] According to the player behavior patterns and resource association rules output by the decision tree model, the recommendation algorithm is used to make personalized recommendations for virtual props, game activities, social friends, etc. in the game. For example, if the decision tree model analyzes that a player likes PVP competitive gameplay and often buys attack props, the system can recommend newly launched PVP competitive activities and related attack enhancement props to the player. ​

[0058] S33. Demand forecast.

[0059] Based on time series analysis, combined with players’ historical gaming behaviors and consumption data, players’ future gaming needs can be predicted. For example, players can be predicted to have demand for certain types of virtual props or types of gaming activities they may participate in over a period of time in the future, so that game operators can prepare resources and plan activities in advance.

[0060] Step 4: Game resource management.

[0061] S41. Resource forecasting and analysis.

[0062] The decision tree model is used to analyze historical game resource consumption data (such as game currency usage, consumption and acquisition of virtual props) and monitor the resource data in the game in real time. For example, the demand patterns of game coins for players of different levels and the preferences and usage frequencies of virtual props for different types of players are analyzed.

[0063] Comprehensively analyze the prediction results and current game resource inventory data, and compare the game coin inventory K and the preset game coin inventory K 0 Compare the virtual item inventory with the preset virtual item inventory to determine whether the current game resources can meet the game needs of future players.

[0064] S42. Resource adjustment.

[0065] When the game coin inventory K ≥ K 0 , or when the inventory of virtual props meets the preset requirements, it is determined that the current game resources can meet future needs.

[0066] When K <K 0 , or when the inventory of virtual props is insufficient, it is determined that the current game resources cannot meet future needs, and the resource replenishment mechanism is automatically triggered. For game coins, they can be replenished by adjusting in-game task rewards, monster drops, etc.; for virtual props, the prop output probability can be adjusted, or the corresponding props can be replenished from the game resource reserve.

[0067] Step 5: Risk warning.

[0068] S51. Establish a risk assessment model.

[0069] Risk assessment indicators are determined based on the analysis of players' gaming behavior patterns (such as frequent logging into abnormal locations, large-scale transactions of the same virtual item in a short period of time), consumption behaviors (such as abnormally large recharges, abnormal sources of recharges) and social interactions (such as frequent interactions with multiple abnormal accounts), combined with the transaction rules and security policies within the game.

[0070] Input the risk assessment indicators into the transaction risk assessment model, obtain the results of the transaction risk assessment model, and use them as the basis for monitoring abnormal behaviors during the game. For example, the model can determine whether the player's account is at risk of being stolen or whether there is illegal behavior in the transaction based on indicators such as the frequency of changes in the player's login IP address and the fluctuation range of the transaction amount.

[0071] S52. Risk warning and handling.

[0072] According to the abnormal behaviors detected, a risk warning level system is established and risk warning signals are sent. When the risk level is low-risk abnormal behavior, such as players occasionally log in from abnormal locations but have no other abnormal behaviors, risk warning signals may not be sent, but relevant situations will be recorded for subsequent observation.

[0073] When the risk level is high-risk abnormal behavior, such as detecting obvious signs of player account theft or large-scale illegal transactions, emergency measures will be taken immediately, such as freezing relevant accounts, suspending trading functions, and sending risk warning signals to game operators and relevant security departments for timely processing.

[0074] The beneficial effects of this embodiment are as follows: this method effectively improves the security and personalization level of game services, ensures the rational allocation of game resources, reduces operational risks, and enhances player experience through measures such as real-time data collection and preprocessing, data encryption and desensitization, personalized recommendation and prediction, game resource management, and risk warning.

[0075] Embodiment 2:

[0076] See also Figure 2 As shown, the Internet-based game service management system of this embodiment includes a data collection and processing module, a data encryption and desensitization module, a personalized recommendation prediction module, a game resource management module and a risk warning module.

[0077] Data collection and processing module: used to collect various types of player behavior data in the game in real time, including game login time, online time, game operation records, social interaction data, consumption data, etc., as well as in-game resource data; integrate and clean the collected data, and store it in a high-performance database.

[0078] Data encryption and desensitization module: used to encrypt and store players' sensitive data, and to perform desensitization during data use. At the same time, corresponding permissions are assigned according to the player's role, and attribute-based access control is adopted to improve account security.

[0079] Personalized recommendation prediction module: used to analyze player data by establishing a decision tree model, output player behavior patterns and resource association rules; use recommendation algorithms to make personalized recommendations for virtual props, game activities, social friends, etc. in the game, and predict players' future game needs based on time series analysis.

[0080] Game resource management module: used to analyze historical game resource consumption data based on the decision tree model, monitor the resource data in the game in real time, and determine whether the current game resources can meet the future game needs of players. If the resources are insufficient, the resource replenishment mechanism will be automatically triggered.

[0081] Risk warning module: used to establish a risk assessment model, analyze players' gaming behavior patterns, consumption behaviors and social interactions, and determine risk assessment indicators; establish a risk warning level system based on detected abnormal behaviors and send risk warning signals.

[0082] The beneficial effects of this embodiment are as follows: through modular design, the system realizes data collection, encryption and desensitization, personalized recommendation, resource management and risk warning functions, effectively improves the security and personalization level of game services, ensures the rational allocation of game resources, reduces operational risks, and provides players with a safer and more convenient gaming experience.

[0083] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0084] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for managing game services based on the Internet, characterized in that: The method process is as follows: Step 1, Data collection and preprocessing: Real-time collect various types of player behavior data and in-game resource data during the game. Integrate and clean the collected data from different sources and in different formats, remove duplicates, errors, and outliers, and store the cleaned and integrated data in a high-performance database using a distributed storage architecture, and classify and store it according to data types and usage frequencies; Step 2, Data encryption and desensitization: Encrypt and store sensitive player data and perform desensitization processing, record access and operation behaviors on sensitive data, allocate permissions according to player roles, adopt attribute-based access control, and implement a strong password policy; Step 3, Personalized recommendation and prediction: Establish a decision tree model, and perform personalized recommendation and prediction of players' future game needs according to the output of the decision tree model; Step 4, Game resource management: Analyze historical game resource consumption data according to the decision tree model, monitor in-game resource data, compare the game resource inventory with the preset amount, and adjust resources according to the comparison results; Step 5, Risk warning: Establish a risk assessment model, monitor abnormal behaviors according to the model results, establish a risk warning level system and perform corresponding processing.

2. The Internet-based game service management method according to claim 1, characterized in that: In the data collection and preprocessing step, the player behavior data includes game login time, online duration, game operation records, social interaction data, and consumption data; the in-game resource data includes the number of game coins, the attributes and quantities of virtual items, map scene information, and monster data.

3. The Internet-based game service management method according to claim 1, characterized in that: In the data encryption and desensitization step, use the AES encryption algorithm to encrypt and store sensitive player data including ID card numbers, bank card information, and recharge records, desensitize sensitive data when using the data, allocate corresponding permissions according to player roles including ordinary players, guild administrators, and game streamers, dynamically determine access permissions based on the player's geographical location and login time, and require players to set complex passwords containing numbers, letters, and special characters and change them regularly.

4. The Internet-based game service management method according to claim 1, characterized in that: The process of establishing the decision tree model in the personalized recommendation and prediction step is as follows: Input player data and in-game resource data into the decision tree model, take 70% of the historical data as the training set and 30% as the test set; calculate the Gini coefficient G of each attribute according to the training set data, sort them from small to large according to the Gini coefficient, take the attribute corresponding to the minimum Gini coefficient as the root node, then calculate the Gini coefficient of the remaining attributes, take the attribute corresponding to the minimum value as the internal node, repeat the calculation until the remaining attributes are 0 to obtain the decision tree model to be tested; test the decision tree model to be tested according to the test set data, compare the test accuracy Q with the preset test accuracy Q0, when Q≥Q0, output the decision tree model to be tested as the decision tree model, when Q<Q0, expand the training set data and continue training until the condition is met and then output.

5. The Internet-based game service management method according to claim 1, characterized in that: In the personalized recommendation and prediction steps, personalized recommendation is to recommend virtual props, game activities, and social friends in the game based on the player behavior patterns and resource association rules output by the decision tree model using a recommendation algorithm; demand prediction is to predict players' future demand for virtual props or game activities based on time series analysis combined with players' historical game behavior and consumption data.

6. The Internet-based game service management method according to claim 1, characterized in that: In the game resource management step, resource prediction and analysis is to analyze historical game resource consumption data according to the decision tree model, monitor in-game resource data, compare the game coin inventory K with the preset game coin inventory, the virtual prop inventory with the preset virtual prop inventory, and judge whether the current game resources can meet future needs; resource adjustment is when the game coin inventory K is greater than or equal to the preset game coin inventory or the virtual prop inventory meets the preset requirements, it is judged that the resources can meet future needs, when K is less than the preset game coin inventory or the virtual prop inventory is insufficient, the resource replenishment mechanism is triggered, the game coins are replenished by adjusting the in-game task rewards and monster drops, and the virtual props are replenished by adjusting the output probability or from the reserve.

7. The Internet-based game service management method according to claim 1, characterized in that: In the risk warning step, the risk assessment model is established based on the game behavior patterns, consumption behaviors and social interactions of the players, combined with the in-game transaction rules and security strategies to determine the risk assessment indicators, and the indicators are input into the transaction risk assessment model to obtain the results as the basis for monitoring abnormal behaviors; Risk warning and processing is to establish a risk warning level system based on the detected abnormal behavior. Low-risk abnormal behavior may not send a warning signal but the situation will be recorded. High-risk abnormal behavior will immediately take emergency measures and send a warning signal.

8. An Internet-based game service management system, used to implement the Internet-based game service management method as claimed in claim 1, characterized in that: The system includes a data collection and processing module, a data encryption and desensitization module, a personalized recommendation prediction module, a game resource management module and a risk warning module; Data collection and processing module: used to collect various types of player behavior data in the game in real time, including game login time, online time, game operation records, social interaction data, consumption data, and in-game resource data; Integrate, clean and store the collected data in the database; Data encryption and desensitization module: used to encrypt and store players' sensitive data, and perform desensitization processing during data use; Assign corresponding permissions according to the player's role and adopt attribute-based access control to improve account security; Personalized recommendation prediction module: used to analyze player data by building a decision tree model, output player behavior patterns and resource association rules; use recommendation algorithms to make personalized recommendations for virtual props, game activities, and social friends in the game, and predict players' future game needs based on time series analysis; Game resource management module: used to analyze historical game resource consumption data based on the decision tree model, monitor the resource data in the game in real time, and determine whether the current game resources can meet the game needs of future players. If the resources are insufficient, the resource replenishment mechanism is automatically triggered; Risk warning module: used to establish a risk assessment model, analyze players' gaming behavior patterns, consumption behaviors and social interactions, and determine risk assessment indicators; Establish a risk warning level system based on the detected abnormal behaviors and send risk warning signals.

Images