Implementation method and device of game incentive mechanism and electronic equipment

Abstract

The invention provides an implementation method and device of a game incentive mechanism and electronic equipment, and relates to the field of data processing. The method comprises the following steps: acquiring behavior data, emotion data, social data and game environment data when a user plays a target game; based on the behavior data, the emotion data, the social data and the game environment data, determining game incentive features; inputting the game excitation features into a GANs generation model, and generating a target game task corresponding to the target game; obtaining feedback data after the user performs the target game task; and processing the feedback data by adopting a DRL model to obtain an incentive scheme so as to realize a game incentive mechanism. By implementing the technical scheme provided by the invention, the implementation effect of a game incentive mechanism can be improved.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to a method, apparatus, and electronic device for implementing a game incentive mechanism. Background Art

[0002] In recent years, with the rapid development of the game industry and the intensification of market competition, the game incentive mechanism has become one of the important means to improve the player experience and increase user stickiness. Among them, the game incentive mechanism is often realized through a game incentive plan.

[0003] Currently, the formulation of game incentive plans usually relies on a preset fixed mode, that is, at the game design stage, experts design a unified task template, reward mechanism, and incentive strategy based on experience and intuition. Such plans often only consider the historical behavior data and static user portraits of players during design, resulting in a lack of flexibility and pertinence in the implementation of the incentive mechanism. Further, due to the characteristics of the game player group such as diverse personalities and rapid behavior changes, it is difficult for the incentive plan under the fixed mode to accurately capture the immediate needs of players, thus resulting in poor implementation effects of the game incentive mechanism.

[0004] Therefore, there is an urgent need for a method, apparatus, and electronic device for implementing a game incentive mechanism. Summary of the Invention

[0005] This application provides a method, apparatus, and electronic device for implementing a game incentive mechanism, which is convenient for improving the implementation effect of the game incentive mechanism.

[0006] In the first aspect of this application, a method for implementing a game incentive mechanism is provided. The method includes: obtaining behavior data, emotional data, social data, and game environment data when a user plays a target game; determining game incentive features based on the behavior data, the emotional data, the social data, and the game environment data; inputting the game incentive features into a GANs generation model to generate a target game task corresponding to the target game; obtaining feedback data after the user completes the target game task; and processing the feedback data using a DRL model to obtain an incentive plan to implement the game incentive mechanism.

[0007] By adopting the above technical solution, the behavior data, emotional data, social data, and game environment data of the user in the target game are collected, and multi-dimensional information is fused to extract personalized game incentive features. Subsequently, the features are input into the GANs generation model to generate a target game task that matches the user's state. After the user completes the task, the system further collects the feedback data and uses the DRL model to process the feedback, dynamically optimizing and outputting a personalized incentive plan, thereby constructing an intelligent game incentive mechanism centered on data-driven and adaptive adjustment. Compared with the prior art, it can provide a game incentive plan that better conforms to the user's personalization, thereby improving the implementation effect of the game incentive mechanism.

[0008] Optionally, the obtaining of the behavior data, emotional data, social data, and game environment data of the user when playing the target game specifically includes: through game log recording, obtaining the operation trajectory, task operation record, operation frequency, and game duration of the user in the target game to obtain the behavior data; obtaining the voice and / or text chat records of the user in the target game, extracting the user's emotional state through an emotion analysis algorithm, and performing quantization processing on the emotion value to obtain the emotional data; obtaining the interaction records, communication frequencies, comments, and likes of the user in the target game and the game social platform to obtain the social data; obtaining the current game level, scene characteristics, real-time events, task difficulty, and dynamic environment variables of the target game to obtain the game environment data.

[0009] By adopting the above technical solution, key information of the user in the target game is comprehensively obtained through multi-source data collection means: First, the operation trajectory, task execution record, operation frequency, and game duration of the user are recorded through game logs to extract behavior data; Second, the voice or text chat records of the user in the game are processed using an emotion analysis algorithm to extract and quantify their emotional state to form emotional data; Third, the interaction records of the user on the game and its social platform are collected, including communication frequencies, comments, and likes, etc., to construct social data; Finally, by combining factors such as the current game level setting, scene characteristics, real-time events, task difficulty, and environmental changes, game environment data is extracted, thereby constructing a multi-dimensional and dynamic description of the user's game state.

[0010] Optionally, determining the game incentive features based on the behavior data, the sentiment data, the social data, and the game environment data specifically includes: performing outlier filtering, missing value processing, and time alignment on the behavior data, the sentiment data, the social data, and the game environment data to obtain multi-source data; extracting click frequency features and operation response time features from the behavior data corresponding to the multi-source data; extracting sentiment tendency score features and emotional fluctuation amplitude features from the sentiment data corresponding to the multi-source data; extracting social influence features and interaction density features from the social data corresponding to the multi-source data; extracting environment matching degree features from the game environment data corresponding to the multi-source data; and determining the game incentive features based on the click frequency features, the operation response time features, the sentiment tendency score features, the emotional fluctuation amplitude features, the social influence features, the interaction density features, and the environment matching degree features.

[0011] By adopting the above technical solution, by preprocessing the behavior data, the sentiment data, the social data, and the game environment data, including outlier filtering, missing value filling, and time alignment, a unified multi-source data set is constructed; subsequently, key features are extracted from each type of data respectively: click frequency and operation response time features are extracted from the behavior data to reflect the player's operation rhythm and concentration; sentiment tendency score and emotional fluctuation amplitude are extracted from the sentiment data to depict the player's emotional state and its change trend; social influence and interaction density are extracted from the social data to measure their social activity level and dissemination ability; environment matching degree is extracted from the game environment data to reflect the adaptability of the task setting to the player's ability. Finally, based on the above multi-dimensional features, the game incentive features for personalized incentive generation are comprehensively determined, providing data support for subsequent intelligent generation tasks and incentive strategies.

[0012] Optionally, determining the game incentive features based on the click frequency features, the operation response time features, the sentiment tendency score features, the emotional fluctuation amplitude features, the social influence features, the interaction density features, and the environment matching degree features specifically includes: adopting an attention mechanism to fuse the click frequency features, the operation response time features, the sentiment tendency score features, the emotional fluctuation amplitude features, the social influence features, the interaction density features, and the environment matching degree features into a comprehensive feature; obtaining the real-time state of the user when playing the target game and determining the game incentive sensitive data of the user; and dynamically allocating weights to the comprehensive feature according to the game incentive sensitive data to generate the vectorized game incentive features.

[0013] By adopting the above technical solutions, when determining the game incentive features, first, the attention mechanism is used to deeply integrate key features from multiple sources, including click frequency, operation response time, sentiment tendency score, emotional fluctuation range, social influence, interaction density, and environmental matching degree, to construct a unified comprehensive feature representation; subsequently, combined with the real-time state of the user during the target game, the current incentive-sensitive data of the user is identified, so as to reflect the user's immediate response tendency to different incentive factors; based on this sensitive data, the system applies dynamic weight adjustment to the integrated features to strengthen the adaptability to the user's current psychological and behavioral characteristics, and finally outputs vectorized game incentive features with personalization and timeliness, providing accurate input for the subsequent generation of personalized tasks and incentive plans.

[0014] Optionally, inputting the game incentive features into the GANs generation model to generate the target game tasks corresponding to the target game specifically includes: through the generator of the GANs generation model, generating text or structured task parameters for describing the target game tasks for the game incentive features; determining the generative tasks according to the task parameters; through the discriminator of the GANs generation model, performing adversarial training on the generative tasks and the real game tasks to meet the game situation and incentive goals, and obtaining the target game tasks.

[0015] By adopting the above technical solutions, the process of using the GANs generation model to generate target game tasks specifically includes: first, inputting the vectorized game incentive features into the generator of the GANs model, which generates text information or structured task parameters for describing the target game tasks to form candidate generative tasks; subsequently, determining the preliminary task content that meets the user characteristics and incentive requirements according to these task parameters; then, through the discriminator in the GANs model, performing adversarial training on the generative tasks and the real tasks in the game, evaluating their rationality and incentive effectiveness in the current game situation, so as to optimize the task generation strategy. The finally output target game tasks not only fit the player's current state and incentive requirements, but also have good game immersion and challenge, realizing the personalization and intelligence of task generation.

[0016] Optionally, the DRL model is used to process the feedback data to obtain an incentive plan to implement the game incentive mechanism, which specifically includes: determining user status data and incentive action data based on the feedback data; generating a short-term incentive plan according to the first dimension through the first incentive function of the DRL model, where the first dimension includes task completion rate, user satisfaction, and emotional improvement; generating a long-term incentive plan according to the second dimension through the second incentive function of the DRL model, where the second dimension includes user retention, repurchase behavior, and social diffusion effect; generating the incentive plan based on the short-term incentive plan and the long-term incentive plan to implement the game incentive mechanism.

[0017] By adopting the above technical solution, the feedback data after the user completes the target game task is processed by the DRL model to generate a personalized incentive plan. Specifically, it includes: first, extracting the user's current status data and incentive-related action data from the feedback data; then using two types of incentive functions set in the DRL model to optimize the strategy from two dimensions: short-term and long-term. The short-term dimension includes indicators such as task completion rate, user satisfaction, and emotional improvement, which are used to quickly evaluate the current incentive effect and generate an incentive plan with immediate feedback; the long-term dimension covers user retention, repurchase behavior, and social dissemination effect, focusing on the continuous value contribution within the user's life cycle. Finally, by integrating the short-term and long-term incentive plans, a final incentive strategy that balances immediate response and continuous guidance is formed to achieve the dynamic optimization and personalized regulation of the game incentive mechanism.

[0018] Optionally, the method further includes: receiving the custom incentive plan uploaded by the user; parsing the custom incentive plan to obtain custom game parameters; and applying the custom game parameters to the target game.

[0019] By adopting the above technical solution, the method supports the user to actively participate in the personalized customization of the incentive mechanism, which specifically includes: receiving the custom incentive plan uploaded by the user, parsing the personalized game preferences and incentive settings included therein, and extracting and structuring the corresponding custom game parameters therefrom; then directly applying these parameters to the target game to achieve personalized adjustment of task content, reward methods, or interaction mechanisms, thereby enhancing the user's sense of control and participation in the game incentive system, and improving the overall game experience and satisfaction.

[0020] In a second aspect of the present application, there is provided an implementation device for a game incentive mechanism. The implementation device includes an acquisition module and a processing module. Among them, the acquisition module is used to acquire the behavior data, emotional data, social data, and game environment data of a user when playing a target game; the processing module is used to determine game incentive features based on the behavior data, emotional data, social data, and game environment data; the processing module is further used to input the game incentive features into a GANs generation model to generate a target game task corresponding to the target game; the acquisition module is further used to acquire feedback data of the user after performing the target game task; the processing module is further used to process the feedback data by using a DRL model to obtain an incentive plan, so as to implement the game incentive mechanism.

[0021] In a third aspect of the present application, there is provided an electronic device. The electronic device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions. Both the user interface and the network interface are used to communicate with other devices. The processor is used to execute the instructions stored in the memory so that the electronic device executes the method described above.

[0022] In a fourth aspect of the present application, there is provided a computer-readable storage medium. The computer-readable storage medium stores instructions, and when the instructions are executed, the method described above is executed.

[0023] In summary, one or more technical solutions provided in the present application have at least the following technical effects or advantages: By collecting the behavior data, emotional data, social data, and game environment data of a user in a target game, integrating multi-dimensional information to extract personalized game incentive features; then inputting the features into a GANs generation model to generate a target game task matching the user's state; after the user completes the task, the system further collects its feedback data and uses a DRL model to process the feedback, dynamically optimizing and outputting a personalized incentive plan, thereby constructing an intelligent game incentive mechanism centered on data-driven and adaptive adjustment. Compared with the prior art, it can provide a game incentive plan that better meets the user's personalization, thereby improving the implementation effect of the game incentive mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic flowchart of a method for implementing a game incentive mechanism provided by an embodiment of the present application; Figure 2 It is another schematic flowchart of a method for implementing a game incentive mechanism provided by an embodiment of the present application; Figure 3 It is a schematic module diagram of an implementation device for a game incentive mechanism provided by an embodiment of the present application; Figure 4 A structural schematic diagram of an electronic device provided by an embodiment of the present application.

[0025] Explanation of reference numerals: 31, acquisition module; 32, processing module; 41, processor; 42, communication bus; 43, user interface; 44, network interface; 45, memory. Specific implementation manners

[0026] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0027] In the description of the embodiments of the present application, words such as "for example" or "for illustration" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "for example" or "for illustration" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "for example" or "for illustration" is intended to present related concepts in a specific manner.

[0028] In the description of the embodiments of the present application, the meaning of the term "plurality" refers to two or more. For example, a plurality of systems refers to two or more systems, and a plurality of screen terminals refers to two or more screen terminals. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the technical features indicated. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The terms "include", "comprise", "have" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0029] With the rapid development of the game industry and the continuous intensification of market competition, game incentive mechanisms have become a key means to enhance the player experience and increase user stickiness. Among them, the core of the incentive mechanism usually relies on specific game incentive programs for implementation.

[0030] However, most of the current incentive programs are still formulated in a preset fixed mode, that is, during the game development stage, designers set unified task templates, reward methods, and incentive strategies based on experience and rules. Such programs often only refer to the historical behaviors or static portraits of players, lacking the ability to perceive and respond to dynamic changes, resulting in the lack of flexibility and personalized adaptation of incentive measures. Especially when facing a player group with increasingly diverse user needs and rapidly changing behavior patterns, the traditional fixed mode is difficult to accurately identify and respond to the incentive needs of players in different situations, thereby limiting the effectiveness of the incentive mechanism.

[0031] To solve the above technical problems, this application provides a method for implementing a game incentive mechanism, referring to Figure 1 , Figure 1 which is a schematic flowchart of a method for implementing a game incentive mechanism provided by an embodiment of this application. The method for implementing this game incentive mechanism is applied to a server and includes steps S110 to S150. The above steps are as follows: S110. Obtain the behavior data, emotional data, social data, and game environment data of the user when playing a target game.

[0032] Specifically, during the process of a user participating in a certain target game, the server collects and records different types of information in real time for subsequent data analysis and implementation of personalized services. Behavior data refers to the operation records of the user in the game, such as the user's movement trajectory, clicks, shootings, task operation records, operation frequency, and game duration. These data reflect the behavior patterns and habits of players in the game. Example: When a player is playing a shooting game, the server will record the player's specific operations such as firing, aiming, and moving each time, as well as the time he stays in each level, so as to judge his game style (such as being more offensive or defensive). Emotional data is obtained by collecting the voice chats or text messages of the user in the game, and using emotional analysis algorithms to judge and quantify the user's emotional state, resulting in the user's emotional changes. Example: In a role-playing game, the user may express excitement, nervousness, or frustration when chatting. The server can analyze these voices or texts to judge whether the player is currently in a high-spirited or low-spirited state, so as to prepare for subsequent personalized incentives.

[0033] Social data covers the interaction records of users within the game or on related social platforms, such as friend additions, comments, likes, teaming information, and other social behaviors among players. This part of the data can be used to measure the activity and influence of players in the community. Example: In a multiplayer online competitive game, the server records players' friend interactions, teaming situations, and comments posted on the game forum. These data can reflect the social activity level of players and provide a reference basis for game community management and incentive strategies. Game environment data refers to the real-time status information of the game itself, such as the current game level, scene characteristics, real-time events, task difficulty, and dynamically changing environmental variables, etc. These data help describe the game situation where the player is located. Example: If a player is in a high-difficulty level of an adventure game, the server will record the specific settings of this level and the special events occurring currently, which can be used to judge the challenges faced by the player in the game, so as to adjust the subsequent task and reward difficulties.

[0034] In summary, by integrating these multi-dimensional data, the server can comprehensively understand the state and environment of players in the game, provide accurate data support for subsequent task generation and incentive plan formulation, and achieve a more personalized and real-time responsive game experience.

[0035] In a possible implementation manner, the behavior data, emotional data, social data, and game environment data of the user during the target game are obtained. Specifically, including: through game log recording, obtaining the operation trajectory, task operation records, operation frequency, and game duration of the user in the target game to obtain behavior data; obtaining the voice and / or text chat records of the user in the target game, extracting the user's emotional state through an emotion analysis algorithm, and quantifying the emotion value to obtain emotional data; obtaining the interaction records, communication frequency, comments, and likes of the user in the target game and the game social platform to obtain social data; obtaining the current game level, scene characteristics, real-time events, task difficulty, and dynamic environmental variables of the target game to obtain game environment data.

[0036] Specifically, the behavior data is collected by recording the operations of the user in the target game through game logs. It includes the operation trajectory, which is used to record operations such as the movement path, aiming, and shooting of the user in the game; task operation records: the specific processes and steps of the user executing tasks in the game; operation frequency: recording the number or frequency of the user's operations; game duration: recording the specific time the user stays in the game. Example: In an action adventure game, the server records the operations of the user's every movement, jump, attack, and item pickup through logs, and at the same time counts the duration the player stays in each level, so as to judge the player's game proficiency and preferred playing methods.

[0037] Emotional data is collected by obtaining the voice and / or text chat records of users in the target game. The collected content includes processing the chat records using sentiment analysis algorithms to extract the emotional states of users during the game (such as joy, tension, anger, or depression); quantifying the emotional values, that is, converting the emotional states of users into numerical indicators for subsequent comparison and analysis. For example: In a multiplayer online game, when a player cheers or complains loudly in a voice chat, the system analyzes the emotional fluctuations through an emotion recognition algorithm and records the emotional state in numerical form (for example, from 1 to 10 points) to judge the current emotional tendency and intensity of the player.

[0038] Social data is collected by obtaining the interaction records of users within the target game and on related game social platforms. Interaction records such as forming teams, adding friends, voice or text communication; communication frequency counts the frequency of a player's participation in discussions and chats; comments and likes record the comment and like behaviors of players on social platforms or game communities. For example: In a competitive game, the server records the teaming information of a player with other players, the number of likes, comment content, and comment frequency after each game, and these data can be used to evaluate the activity level and influence of a player in the social network.

[0039] Game environment data is collected by obtaining the current running status information of the target game. Game levels: the levels or stages where the current player is located; scene characteristics: record the layout, style, difficulty, and other characteristics of the game scene; real-time events: such as time-limited events, sudden tasks, or dynamic events within the game; task difficulty: the difficulty level of the current level or task; dynamic environment variables: parameters that change continuously in the game, such as weather, time, the number of monsters, etc. For example: In an adventure game, the system records the level information that the player is challenging, the complexity of the current scene, and whether there are time-limited events or sudden events, so as to judge the challenge and urgency of the current game environment, which is very crucial for adjusting the incentive strategy.

[0040] In summary, this passage details how the server collects various types of information of users in the game through multiple channels, from behavioral, emotional, social to environmental data, providing basic data support for subsequent data fusion, situation awareness, and the generation of personalized incentive programs.

[0041] S120. Determine game incentive characteristics based on behavioral data, emotional data, social data, and game environment data.

[0042] Specifically, the server will first clean, calibrate, and normalize these data to ensure consistency in terms of time and format. The server will extract meaningful metrics from various types of data. For example, it will extract operation frequency or task completion efficiency from behavioral data, obtain emotional tendencies and fluctuations from emotional data, measure interaction density or social influence from social data, and extract level difficulty or environmental suitability from environmental data. Then, using technical means (such as attention mechanisms, deep neural networks, or other multi-modal fusion methods), these scattered metrics will be integrated into a unified game incentive feature. This feature can be regarded as a multi-dimensional vector that comprehensively describes the player's current state and the surrounding environment, revealing their sensitivity to different incentive strategies and needs.

[0043] For example, assume a player is playing a multiplayer online tactical game, and the information collected by the server is as follows: The behavioral data record shows that the player has a high operation frequency and a fast task completion speed, but there are frequent failure records in high-difficulty levels. The emotional data shows that the player's voice chat reveals emotions of anxiety and frustration. The social data shows that the player actively participates in discussions in the team chat and also receives a lot of encouragement and praise from teammates. The game environment data shows that the current player is in a high-difficulty and challenging level with complex scene designs and many real-time events. By fusing and processing the above data, the game incentive feature extracted by the server may indicate that the player has certain pressure under high-intensity operations, has positive feedback in social interactions, but lacks sufficient confidence when facing high-difficulty tasks. Based on this feature, the game system can customize incentive strategies, such as generating more friendly and assisting tasks, increasing immediate encouragement rewards, or adjusting task difficulty, to boost the player's confidence and promote a continuous gaming experience. In summary, this passage describes the integration and refinement of various multi-dimensional data into a comprehensive and dynamically reflective feature of the player's game state and needs to guide the formulation of subsequent personalized game incentive plans.

[0044] In one possible implementation, game incentive features are determined based on behavioral data, emotional data, social data, and game environment data, specifically including: filtering outliers, processing missing values, and time aligning the behavioral data, emotional data, social data, and game environment data to obtain multi-source data; extracting click frequency features and operation response time features from the behavioral data corresponding to the multi-source data; extracting emotional tendency score features and emotional fluctuation amplitude features from the emotional data corresponding to the multi-source data; extracting social influence features and interaction density features from the social data corresponding to the multi-source data; extracting environment matching features from the game environment data corresponding to the multi-source data; determining game incentive features based on click frequency features, operation response time features, emotional tendency score features, emotional fluctuation amplitude features, social influence features, interaction density features, and environment matching features.

[0045] Specifically, the server will first clean the collected user behavior data, emotional data, social data and game environment data, including outlier filtering (removing extremely abnormal data), missing value filling (making up for data omissions caused by network or equipment, etc.) and time alignment (ensuring that different types of data can be matched and compared on the timeline), thereby forming a structured multi-source data set. Then, key features are extracted from different data: for example, "click frequency" and "operation response time" are extracted from behavioral data to indicate operation proficiency; "emotional tendency score" and "emotional fluctuation amplitude" are extracted from emotional data to reflect emotional state and stability; "social influence" and "interaction density" are extracted from social data to measure their social activity and influence; "environmental matching" is extracted from game environment data, that is, the degree of adaptation of the current level or scene to the player's ability. After integrating the features of the above multiple dimensions, a comprehensive "game incentive feature" can be formed for the subsequent generation of targeted incentive content.

[0046] For example, a player in a shooting game has frequent operations but a large response delay (indicating a little nervousness or lack of proficiency), has a strong negative emotional tendency and large emotional fluctuations in voice chat (may be in an anxious state), interacts with others frequently but in a small circle (limited social influence but high interaction density), and is currently in a difficult map task (low environmental matching). Based on the above analysis, the game incentive feature will combine these dimensions to determine that the player currently needs a combination strategy of "reducing task difficulty + providing positive emotional feedback + social incentives", so as to generate tasks and reward mechanisms more accurately and improve their game enthusiasm and retention rate.

[0047] In a possible implementation manner, game incentive features are determined based on click frequency features, operation response time features, sentiment tendency score features, emotional fluctuation amplitude features, social influence features, interaction density features, and environmental matching degree features. Specifically, it includes: using an attention mechanism to fuse the click frequency features, operation response time features, sentiment tendency score features, emotional fluctuation amplitude features, social influence features, interaction density features, and environmental matching degree features into a comprehensive feature; obtaining the real-time state of the user when playing the target game to determine the user's game incentive sensitive data; and dynamically assigning weights to the comprehensive feature according to the game incentive sensitive data to generate a vectorized game incentive feature.

[0048] Specifically, first, the server fuses data features from multiple different sources, including "click frequency", "operation response time", "sentiment tendency score", "emotional fluctuation amplitude", "social influence", "interaction density", and "environmental matching degree", etc. These features describe the player's operation habits, emotional state, social interactions, and the game environment they are in. Through the attention mechanism, these features are fused into a comprehensive feature. The role of the attention mechanism here is to automatically adjust the weights of each feature in the fusion process according to the importance of each feature, ensuring that more important features contribute more to the final incentive feature. Next, the server determines which features are "game incentive sensitive data" based on the player's current game state (for example, the player's performance in the current level, emotional changes, etc.). These data reflect the player's immediate reaction or demand for the incentive mechanism, such as the current sense of challenge, emotional fluctuations, etc. According to the player's real-time state and sensitive data, the server dynamically assigns weights to the fused features. That is to say, if the player's emotional fluctuation is large at a certain moment, the system may increase the weight of the "emotional fluctuation amplitude feature"; if the player encounters difficulties in the task, the weight of the "operation response time feature" may be increased. Finally, after dynamic weight adjustment, a vectorized "game incentive feature" is generated. This feature contains the player's current needs and state and can provide precise input for the subsequent design of the incentive mechanism.

[0049] For example, assume that a player is engaged in a multiplayer battle game, with frequent operations but a slightly slow reaction (a relatively large operation response time feature), and shows a certain degree of anxiety in the voice (a relatively high emotional fluctuation amplitude feature). At this time, the player may need more positive encouragement and task adjustment. Through the attention mechanism, the system combines this data to generate a comprehensive feature, and finds that the player has a large emotional fluctuation and a relatively low environmental matching degree (high difficulty) of the current game task. Therefore, the system will increase the weights of the "emotional fluctuation" and "environmental matching degree" features. The finally generated "game incentive feature" will particularly focus on emotional regulation and task difficulty adjustment, so as to provide a more suitable incentive plan for the player, which may be to appropriately reduce the task difficulty, increase emotional support, etc.

[0050] S130. Input the game incentive feature into the GANs generation model to generate a target game task corresponding to the target game.

[0051] Specifically, the server will use the "game incentive feature" after feature extraction, fusion, and weighting as input data and send it into the GANs model. This feature contains information such as the player's operation habits, emotional state, and social interactions, reflecting the player's current needs and state. GANs (Generative Adversarial Networks) is a deep learning model that undergoes adversarial training through a generator (generating tasks) and a discriminator (evaluating whether the tasks are reasonable). In this scenario, the generator of GANs will generate game tasks (such as specific levels, goals, challenges, etc.) that match the input "game incentive feature". These tasks are personalized and aim to enhance the player's sense of participation, challenge, and satisfaction.

[0052] For example, assume that in a shooting game, due to a large emotional fluctuation (anxiety) and a slightly slow operation reaction (a long response time) of the player, the "game incentive feature" generated by the system indicates that the player may need a moderately challenging task. The generator of GANs generates a task based on this incentive feature, such as reducing the difficulty of the current level and adding auxiliary goals (such as hints or additional help), so that the task is more friendly and challenging to the player without making the player feel overly frustrated or defeated. In this way, the generated task is personalized and can be adjusted according to the player's real-time needs and state.

[0053] In a possible implementation manner, inputting the game incentive feature into the GANs generation model to generate a target game task corresponding to the target game specifically includes: through the generator of the GANs generation model, generating text or structured task parameters for describing the target game task for the game incentive feature; determining the generative task according to the task parameters; through the discriminator of the GANs generation model, performing adversarial training on the generative task and the real game task to meet the game situation and incentive goal, and obtaining the target game task.

[0054] Specifically, first, the generator of the GANs model converts the input "game incentive features" into specific task descriptions. These task descriptions may be in text form (e.g., "The player defeats 5 enemies within 5 minutes") or structured task parameters (e.g., "Task difficulty = medium, Reward = 100 gold coins"). These task descriptions or parameters directly reflect the player's current needs and status, generating personalized game tasks based on the incentive features. According to the generated task descriptions or task parameters, the system determines specific "generative tasks", that is, actual game tasks based on these parameters. These tasks can be newly designed and specifically set to meet the player's personalized needs. Next, the discriminator of the GANs compares the generated tasks with real game tasks for adversarial training. The role of the discriminator is to judge whether the generated tasks are reasonable and in line with the game scenario and incentive goals. For example, it will check whether the generated tasks match the player's skill level, emotional state, and challenge requirements, etc. If the generated tasks do not meet the standards, the discriminator will feedback to the generator, and the generator will then adjust the task generation process to optimize the quality of the tasks.

[0055] For example, assume that a player shows anxiety and has a slow operation reaction when completing a game task. Based on these "game incentive features", the generator of the GANs may generate a task, such as "Defeat 3 enemies within 5 minutes (reduce difficulty)", and attach a task reward (e.g., provide additional hints or enhancement items). Then, the discriminator will compare this task with the real tasks in the game to judge whether it matches the player's emotional state and skill level. If the discriminator believes that this task is too simple or not challenging enough, it will adjust the task difficulty or reward until a task that meets the incentive goals and can stimulate the player's participation is generated.

[0056] S140. Obtain the feedback data after the user performs the target game task.

[0057] Specifically, the feedback data can include the player's completion status (such as whether the task is successfully completed, the time taken to complete the task, the resources used, etc.), the player's emotional reactions (such as whether they feel happy, frustrated, or challenged), and other behavioral data (such as whether they continue to play the game, whether they interact with other players, etc.). These data can reflect the effectiveness of the task, the player's satisfaction, and may even reveal whether the task or reward mechanism needs to be adjusted.

[0058] For example, assume that a player has completed a task which requires them to defeat 5 enemies in the game. After the task is completed, the server will collect the player's feedback data, which may include: whether the player successfully defeated 5 enemies, how much time it took, whether certain items were used during the task, and whether the player showed satisfaction or frustration after the task. If the player felt relaxed and showed a happy mood during the task, the feedback data will show this. If the player failed, the feedback data may include their sense of frustration or feedback on the task difficulty. Based on this feedback data, the system can further adjust the task design or reward mechanism to optimize the player's gaming experience.

[0059] S150. Use the DRL model to process the feedback data to obtain an incentive plan to implement the game incentive mechanism.

[0060] Specifically, the DRL model will evaluate the effectiveness of the current incentive strategy based on the feedback data (such as whether the player completed the task, emotional reactions, gaming behaviors, etc.). Then, it will continuously adjust the rewards, task difficulty, or other incentive elements in the game according to this feedback to form a dynamically adjusted incentive plan. This way can ensure that the incentive mechanism is personalized and optimized as the player's behaviors and emotions change.

[0061] For example, assume that a certain player shows frustration after completing a task and fails to complete the task successfully. The server will analyze the player's feedback data (such as mood swings, reasons for failure, etc.) through the DRL model and generate a new incentive plan accordingly. For example, the DRL model may decide to lower the task difficulty, extend the time limit, or provide more hints to help the player. As the player continues to participate in the game, the DRL model will continuously adjust the incentive strategy based on the feedback of each task completion (such as whether the player is satisfied, whether they continue to challenge more difficult tasks, etc.) to ensure that the player's participation and satisfaction are continuously improved.

[0062] In a possible implementation manner, using the DRL model to process the feedback data to obtain an incentive plan to implement the game incentive mechanism specifically includes: based on the feedback data, determine the user status data and incentive action data; through the first incentive function of the DRL model, generate a short-term incentive plan according to the first dimension, the first dimension includes the task completion rate, user satisfaction, and emotional improvement; through the second incentive function of the DRL model, generate a long-term incentive plan according to the second dimension, the second dimension includes user retention, repurchase behavior, and social diffusion effect; based on the short-term incentive plan and the long-term incentive plan, generate an incentive plan to implement the game incentive mechanism.

[0063] Specifically, the DRL model first determines two important pieces of information based on the player's feedback data, namely user status data and incentive action data. The user status data may include the player's mood, current game progress, task completion status, etc., while the incentive action data is different incentive strategies (such as rewards, task adjustments, etc.) selected according to the player's behavior. Then, the DRL model generates short-term and long-term incentive plans through two different incentive functions. The short-term incentive plan is generated through the first incentive function of the DRL model, based on factors such as task completion rate (such as the speed and success rate of the player completing tasks), user satisfaction (such as whether the player feels happy or satisfied), and emotional improvement (such as whether the player's mood is improved), to generate an immediate incentive plan. These incentives are usually immediate, such as providing rewards, tips, or adjusting task difficulty to increase player engagement in a short period of time. The long-term incentive plan is generated through the second incentive function of the DRL model, based on factors such as user retention (such as whether the player continues to participate in the game), repurchase behavior (such as whether the player makes in-game purchases), and social diffusion effect (such as whether the player shares the game with others or recommends the game), to generate a long-term incentive plan. These incentive plans focus on maintaining the player's long-term activity and enhancing the interaction between the player and the game community. Finally, the system combines the short-term and long-term incentive plans to generate the final game incentive plan to achieve the goal of optimizing the player experience, increasing game engagement, and player satisfaction.

[0064] For example, assume that a player has completed a game task, showing certain emotional fluctuations and not being completely successful. Through the DRL model, the system may generate a short-term incentive plan based on the player's task completion rate and emotional fluctuations, such as providing additional tips or rewards to help the player improve their mood. In the long term, the DRL model may focus on whether the player continues to play the game, whether they make virtual item purchases, and develop long-term incentive strategies based on this data, such as providing exclusive content or social rewards for loyal players to enhance player retention and repurchase behavior.

[0065] In a possible implementation manner, refer to Figure 2 , Figure 2 which is another process schematic diagram of a method for implementing a game incentive mechanism provided by an embodiment of this application. It includes steps S210 to S230, and the above steps are as follows: S210, receive a custom incentive plan uploaded by the user; S220, parse the custom incentive plan to obtain custom game parameters; S230, apply the custom game parameters to the target game.

[0066] Specifically, players can upload a customized incentive plan according to their own needs and preferences, which may include some task designs, reward settings, or incentive strategies. The system will first parse this customized incentive plan and extract the key parameters, which can be specific game task settings, reward rules, player interaction requirements, etc. Then, the system applies these parsed customized game parameters to the target game, adjusting the game mechanism or task settings to provide more suitable incentives according to the personalized needs of the players.

[0067] For example, assume that a player hopes to adjust the reward mechanism in the game through a customized incentive plan. They may upload a plan that requires obtaining additional virtual item rewards after completing a specific task, and this reward is only available to players of a certain level. The system will parse this incentive plan and extract parameters such as the task type, reward conditions, and player level limit. Then, these customized game parameters will be applied to the game, and the corresponding task settings and reward mechanisms will be adjusted to provide a personalized incentive plan for the player. This makes the game experience more in line with the expectations and needs of the players.

[0068] This application also provides an implementation device for a game incentive mechanism. Refer to Figure 3 , Figure 3 , which is a schematic diagram of the modules of an implementation device for a game incentive mechanism provided by an embodiment of this application. This implementation device is a server, and the server includes an acquisition module 31 and a processing module 32. Among them, the acquisition module 31 acquires the behavior data, emotional data, social data, and game environment data of the user when playing the target game; the processing module 32 determines the game incentive characteristics based on the behavior data, emotional data, social data, and game environment data; the processing module 32 inputs the game incentive characteristics into the GANs generation model to generate the target game tasks corresponding to the target game; the acquisition module 31 acquires the feedback data of the user after completing the target game tasks; the processing module 32 processes the feedback data using the DRL model to obtain an incentive plan to implement the game incentive mechanism.

[0069] In a possible implementation, behavior data, emotional data, social data, and game environment data of a user during a target game are obtained, specifically including: the processing module 32 obtains the operation trajectory, task operation record, operation frequency, and game duration of the user in the target game through game log recording to obtain behavior data; the acquisition module 31 obtains the voice and / or text chat records of the user in the target game, extracts the user's emotional state through an emotion analysis algorithm, and performs quantization processing on the emotion value to obtain emotional data; the acquisition module 31 obtains the interaction records, communication frequencies, comments, and likes of the user in the target game and the game social platform to obtain social data; the acquisition module 31 obtains the current game level, scene characteristics, real-time events, task difficulty, and dynamic environment variables of the target game to obtain game environment data.

[0070] In a possible implementation, the processing module 32 determines game incentive features based on behavior data, emotional data, social data, and game environment data, specifically including: the processing module 32 performs outlier filtering, missing value processing, and time alignment on the behavior data, emotional data, social data, and game environment data to obtain multi-source data; the processing module 32 extracts a click frequency feature and an operation response time feature from the behavior data corresponding to the multi-source data; the processing module 32 extracts an emotional tendency score feature and an emotional fluctuation amplitude feature from the emotional data corresponding to the multi-source data; the processing module 32 extracts a social influence feature and an interaction density feature from the social data corresponding to the multi-source data; the processing module 32 extracts an environment matching degree feature from the game environment data corresponding to the multi-source data; the processing module 32 determines game incentive features based on the click frequency feature, operation response time feature, emotional tendency score feature, emotional fluctuation amplitude feature, social influence feature, interaction density feature, and environment matching degree feature.

[0071] In a possible implementation, the processing module 32 determines game incentive features based on the click frequency feature, operation response time feature, emotional tendency score feature, emotional fluctuation amplitude feature, social influence feature, interaction density feature, and environment matching degree feature, specifically including: the processing module 32 uses an attention mechanism to fuse the click frequency feature, operation response time feature, emotional tendency score feature, emotional fluctuation amplitude feature, social influence feature, interaction density feature, and environment matching degree feature into a comprehensive feature; the acquisition module 31 obtains the real-time state of the user during the target game and determines the game incentive sensitive data of the user; the processing module 32 performs dynamic weight allocation on the comprehensive feature according to the game incentive sensitive data to generate a vectorized game incentive feature.

[0072] In a possible implementation, the processing module 32 inputs the game incentive feature into the GANs generation model to generate a target game task corresponding to the target game, specifically including: the processing module 32 generates text or structured task parameters for describing the target game task for the game incentive feature through the generator of the GANs generation model; the processing module 32 determines a generative task according to the task parameters; the processing module 32 performs adversarial training on the generative task and the real game task to meet the game scenario and incentive goal through the discriminator of the GANs generation model, and obtains the target game task.

[0073] In a possible implementation, the processing module 32 uses the DRL model to process the feedback data to obtain an incentive plan to implement the game incentive mechanism, specifically including: the processing module 32 determines user state data and incentive action data based on the feedback data; the processing module 32 generates a short-term incentive plan according to the first dimension through the first incentive function of the DRL model, and the first dimension includes task completion rate, user satisfaction, and emotional improvement; the processing module 32 generates a long-term incentive plan according to the second dimension through the second incentive function of the DRL model, and the second dimension includes user retention, repurchase behavior, and social diffusion effect; the processing module 32 generates an incentive plan based on the short-term incentive plan and the long-term incentive plan to implement the game incentive mechanism.

[0074] In a possible implementation, the acquisition module 31 receives a custom incentive plan uploaded by the user; the processing module 32 analyzes the custom incentive plan to obtain custom game parameters; the processing module 32 applies the custom game parameters to the target game.

[0075] It should be noted that when the device provided in the above embodiments implements its functions, only the above-mentioned division of each functional module is used for illustration. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0076] This application also provides an electronic device. Refer to Figure 4 , Figure 4 which is a schematic structural diagram of an electronic device provided by an embodiment of this application. The electronic device may include: at least one processor 41, at least one network interface 44, a user interface 43, a memory 45, and at least one communication bus 42.

[0077] Among them, the communication bus 42 is used to realize the connection and communication between these components.

[0078] Among them, the user interface 43 may include a display screen and a camera. Optionally, the user interface 43 may further include standard wired interfaces and wireless interfaces.

[0079] Among them, the network interface 44 may optionally include standard wired interfaces and wireless interfaces (such as Wi-Fi interfaces).

[0080] Among them, the processor 41 may include one or more processing cores. The processor 41 connects various parts within the entire server through various interfaces and circuits. By running or executing instructions, programs, code sets, or instruction sets stored in the memory 45, and by calling the data stored in the memory 45, the processor 41 performs various functions of the server and processes data. Optionally, the processor 41 may be implemented in at least one of the hardware forms of digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 41 may integrate a combination of one or more of a central processing unit (CPU), a graphics processing unit (GPU), and a modem, etc. Among them, the CPU mainly processes the operating system, user interfaces, and application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 41 and may be implemented separately by a single chip.

[0081] Among them, the memory 45 may include random access memory (RAM) and may also include read-only memory. Optionally, the memory 45 includes a non-transitory computer-readable storage medium. The memory 45 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 45 may include a program storage area and a data storage area. Among them, the program storage area may store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store the data involved in the above-mentioned various method embodiments. Optionally, the memory 45 may further be at least one storage device located far from the aforementioned processor 41. For example Figure 4As shown in the figure, the memory 45, which is a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for an implementation method of a game incentive mechanism.

[0082] In Figure 4 In the electronic device shown in the figure, the user interface 43 is mainly used to provide an interface for the user to input and obtain the data input by the user; while the processor 41 can be used to call the application program for the implementation method of the game incentive mechanism stored in the memory 45. When executed by one or more processors, the electronic device is caused to execute the method as described in one or more of the above embodiments.

[0083] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps can be adopted in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0084] The present application also provides a computer-readable storage medium, and the computer-readable storage medium stores instructions. When executed by one or more processors, the electronic device is caused to execute the method as described in one or more of the above embodiments.

[0085] In the above embodiments, the descriptions of the various embodiments each have their own focuses. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0086] In several embodiments provided by the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, 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 displayed or discussed couplings or direct couplings or communication connections to each other can be through some service interfaces. The indirect couplings or communication connections of the devices or units can be in electrical or other forms.

[0087] 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 can be located in one place, or they can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0088] In addition, in each embodiment of the present application, each functional unit can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0089] 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 memory. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in the various embodiments of the present application. The aforementioned memory includes various media that can store program codes, such as USB flash drives, mobile hard disks, magnetic disks, or optical discs.

[0090] The above are only exemplary embodiments of the present disclosure, and the scope of the present disclosure cannot be limited thereby. That is, any equivalent changes and modifications made in accordance with the teachings of the present disclosure still fall within the scope covered by the present disclosure. After considering the specification and the disclosure of the practical truth, those skilled in the art will easily think of other implementation manners of the present disclosure. The present application aims to cover any variations, uses, or adaptive changes of the present disclosure, and these variations, uses, or adaptive changes follow the general principles of the present disclosure and include the common general knowledge or conventional technical means in the technical field not recorded in the present disclosure. The specification and the embodiments are only regarded as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A method for implementing a game incentive mechanism, characterized in that: The method comprises: Obtaining behavioral data, emotional data, social data, and game environment data of users when playing the target game; Determining game motivational features based on the behavioral data, the emotional data, the social data, and the game environment data; Inputting the game incentive features into the GANs generation model to generate a target game task corresponding to the target game; Obtaining feedback data from the user after performing the target game task; The feedback data is processed using a DRL model to obtain an incentive scheme to implement a game incentive mechanism.

2. The method for implementing the game incentive mechanism according to claim 1, characterized in that: The obtaining of the user's behavior data, emotion data, social data, and game environment data when playing the target game specifically includes: Obtain the user's operation track, task operation record, operation frequency and game duration in the target game through game log records to obtain the behavior data; Acquire the voice and / or text chat records of the user in the target game, extract the user's emotional state through a sentiment analysis algorithm, and quantify the sentiment value to obtain the sentiment data; Obtaining the user's interaction records, communication frequency, comments, and likes in the target game and on the game social platform to obtain the social data; The current game level, scene characteristics, real-time events, task difficulty and dynamic environment variables of the target game are obtained to obtain the game environment data.

3. The method for implementing the game incentive mechanism according to claim 1, characterized in that: The determining of the game motivational features based on the behavior data, the emotion data, the social data and the game environment data specifically includes: Performing outlier filtering, missing value processing, and time alignment on the behavior data, the emotion data, the social data, and the game environment data to obtain multi-source data; Extracting click frequency features and operation response time features from the behavior data corresponding to the multi-source data; Extracting emotional tendency score features and emotional fluctuation amplitude features from the emotional data corresponding to the multi-source data; Extracting social influence features and interaction density features from the social data corresponding to the multi-source data; Extracting environment matching features from the game environment data corresponding to the multi-source data; The game incentive feature is determined based on the click frequency feature, the operation response time feature, the emotional tendency score feature, the emotional fluctuation amplitude feature, the social influence feature, the interaction density feature and the environment matching feature.

4. The method for implementing the game incentive mechanism according to claim 3, characterized in that: The determining of the game incentive feature based on the click frequency feature, the operation response time feature, the emotion tendency score feature, the emotion fluctuation amplitude feature, the social influence feature, the interaction density feature and the environment matching feature specifically includes: The attention mechanism is used to integrate the click frequency feature, the operation response time feature, the sentiment tendency score feature, the sentiment fluctuation amplitude feature, the social influence feature, the interaction density feature and the environment matching feature into a comprehensive feature; Acquire the real-time status of the user when playing the target game, and determine the game incentive sensitive data of the user; According to the game incentive sensitive data, dynamic weight allocation is performed on the comprehensive features to generate vectorized game incentive features.

5. The method for implementing the game incentive mechanism according to claim 1, characterized in that: The step of inputting the game incentive feature into the GANs generation model to generate a target game task corresponding to the target game specifically includes: Generate text or structured task parameters for describing the target game task according to the game incentive features through the generator of the GANs generation model; Determine a generation task according to the task parameters; Through the discriminator of the GANs generation model, the generative task and the real game task are subjected to adversarial training to meet the game context and motivational goals, thereby obtaining the target game task.

6. The method for implementing the game incentive mechanism according to claim 1, characterized in that: The feedback data is processed by the DRL model to obtain an incentive scheme to implement a game incentive mechanism, which specifically includes: Determining user status data and incentive action data based on the feedback data; Generate a short-term incentive plan according to a first dimension through a first incentive function of the DRL model, wherein the first dimension includes task completion rate, user satisfaction, and emotional improvement; Generate a long-term incentive plan according to a second dimension through the second incentive function of the DRL model, where the second dimension includes user retention, repurchase behavior, and social diffusion effect; Based on the short-term incentive plan and the long-term incentive plan, the incentive plan is generated to implement the game incentive mechanism.

7. The method for implementing the game incentive mechanism according to claim 1, characterized in that: The method further comprises: Receiving a customized incentive plan uploaded by the user; Analyze the custom incentive scheme to obtain custom game parameters; Apply the custom game parameters to the target game.

8. A device for implementing a game incentive mechanism, characterized in that: The implementation device comprises an acquisition module (31) and a processing module (32), wherein: The acquisition module (31) is used to acquire the user's behavior data, emotion data, social data and game environment data when playing the target game; The processing module (32) is used to determine game motivational features based on the behavior data, the emotion data, the social data and the game environment data; The processing module (32) is further used to input the game incentive feature into the GANs generation model to generate a target game task corresponding to the target game; The acquisition module (31) is also used to acquire feedback data of the user after performing the target game task; The processing module (32) is also used to process the feedback data using a DRL model to obtain an incentive scheme to implement a game incentive mechanism.

9. An electronic device, characterized in that: The electronic device comprises a processor (41), a memory (45), a user interface (43) and a network interface (44), wherein the memory (45) is used to store instructions, the user interface (43) and the network interface (44) are both used to communicate with other devices, and the processor (41) is used to execute the instructions stored in the memory (45) so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1 to 7 is performed.

Images