Collaborative strategy generation method, system and equipment for multi-player game

Abstract

The invention is applicable to the technical field of computers, and provides a collaborative strategy generation method for a multi-player game, which comprises the following steps: distributing a role customization scheme for an initial player based on a player role generalization model; distributing a game session for the initial player based on the role customization scheme and the role participation mode, and marking the initial player as a game player; acquiring game behavior data of each game player, and analyzing the game behavior data to obtain a cooperation mode of the game players; dynamically adjusting game element attribute parameters according to the role customization scheme, the game behavior data and the cooperation mode of the game player; inputting the real-time game behavior data and the game element attribute parameters into a pre-trained deep learning model to generate collaborative strategy suggestions; and providing the collaboration strategy suggestions to the corresponding game players, and dynamically adjusting resource allocation of each game player according to the collaboration strategy suggestions. According to the method, the ability of the multi-player game in the aspects of deep strategy collaboration and personalized game experience can be improved.

Description

Technical Field

[0001] The present application belongs to the field of computer technology, and more particularly to a method, system and system for generating a collaborative strategy for a multiplayer game. Background Art

[0002] Multiplayer games are a type of video game in which thousands of players can exist in the same game at the same time. Players can fight or cooperate with players from all over the world in real time to compete for game resources, territories or complete specific tasks, which makes the game lively, interesting and highly interactive.

[0003] In multiplayer games of existing technologies, players' decision-making process is often independent of others. Even if they are in the same game world, it is difficult for them to feel close connection and collaborative support with each other. This isolation not only weakens the strategic depth of the game, but also limits the personalized development of players in the game. In addition, in terms of strategy generation, existing multiplayer games often lack intelligent tools to assist players in formulating and optimizing strategies. Players need to rely on their own experience and intuition to make decisions, which undoubtedly increases the difficulty of the game and also limits the innovation and diversity of strategies. At the same time, the game environment is not flexible enough in adapting and responding to changes in player strategies. When players adopt new strategies or actions in the game, the game environment often cannot respond in time, resulting in the inability to effectively verify and optimize the player's strategy. This game environment lacking dynamic feedback not only weakens the strategic nature of the game, but also reduces the player's gaming experience.

[0004] To sum up, multiplayer games still need to improve in terms of deep strategic collaboration and personalized gaming experience. Summary of the invention

[0005] The embodiments of the present application provide a method, system and system for generating a collaborative strategy for a multiplayer game, which can solve one of the above-mentioned problems in the prior art.

[0006] In a first aspect, an embodiment of the present application provides a method for generating a collaborative strategy for a multiplayer game, comprising:

[0007] Constructing a player role generalization model, and assigning a role customization scheme to an initial player based on the player role generalization model;

[0008] Acquire a role participation mode of an initial player, assign a game session to the initial player based on the role customization scheme and the role participation mode, and mark the initial player as a game player;

[0009] For multiple game players in the same game session, obtain game behavior data of each game player, analyze the game behavior data, and obtain a collaboration mode of the game players;

[0010] Dynamically adjust game element attribute parameters according to the game player's role customization plan, game behavior data, and collaboration mode;

[0011] Inputting the real-time game behavior data and game element attribute parameters of the game players into a pre-trained deep learning model to generate collaborative strategy recommendations;

[0012] The collaboration strategy suggestion is provided to the corresponding game player, and the resource allocation of each game player is dynamically adjusted according to the collaboration strategy suggestion.

[0013] Furthermore, the constructing of the player role generalization model includes:

[0014] Based on the character customization data of historical players, a clustering algorithm is used to classify the historical players to obtain the character customization preference characteristics of different player groups, wherein the historical players are players who have played multi-player games;

[0015] For each of the player groups, an association rule mining algorithm is used to analyze the association constraint relationship between the player role attributes to form a role attribute combination rule library;

[0016] Based on the role customization preference characteristics and the role attribute combination rule base, a genetic algorithm is used to generate diversified role customization solutions.

[0017] Furthermore, the allocation of a role customization scheme to the initial player based on the player role generalization model includes:

[0018] Acquire historical character customization data of an initial player, and determine the player group of the initial player through a player character preference model;

[0019] The player character generalization model generates a plurality of character customization schemes through the corresponding player group of the initial player;

[0020] An attribute balance check is performed on the plurality of character customization schemes, and a character customization scheme is allocated to the initial player by adjusting the character attribute combinations in the character customization schemes.

[0021] Furthermore, the role participation mode includes a group joining mode and a group building mode;

[0022] The allocating a game session to the initial player comprises:

[0023] For the group joining method, based on the character customization scheme, the initial player is added to a game session of an existing multiplayer game;

[0024] For the group building method, a new game session is created, and game players are assigned to the new game session based on the role customization plan.

[0025] Furthermore, for multiple game players in the same game session, obtaining game behavior data of each game player, analyzing the game behavior data, and obtaining the collaboration mode of each game player includes:

[0026] According to the game behavior data of each game player, the frequency of collaboration between each game player and other game players is counted to obtain the player collaboration frequency matrix;

[0027] Based on the player collaboration frequency matrix, cluster analysis is used to obtain the collaboration mode between different game players;

[0028] Analyze the multiple collaboration modes to obtain collaboration effect indicators of each collaboration mode;

[0029] Based on the collaboration effect index, an association rule mining algorithm is used to obtain the collaboration mode of the role customization schemes of different game players.

[0030] Furthermore, the analysis of the plurality of collaboration modes to obtain collaboration effect indicators of each collaboration mode includes:

[0031] Obtaining team performance data and collaboration frequency corresponding to each collaboration in each of the collaboration modes;

[0032] Inputting the team performance data, the collaboration frequency and the collaboration mode into a pre-trained collaboration effect evaluation model to obtain the influence of different collaboration modes on the collaboration effect;

[0033] Based on the influence rules, the effect weights of the collaboration effects of various collaboration modes are determined, and collaboration effect indicators are generated.

[0034] Furthermore, dynamically adjusting the game element attribute parameters according to the game player's role customization scheme, game behavior data, and collaboration mode includes:

[0035] Based on the character customization scheme and game behavior data of the game player, a player game experience evaluation model is constructed;

[0036] Extracting game element attribute parameters from a plurality of preset game configuration schemes, inputting the game element attribute parameters into the player game experience evaluation model, and obtaining a fitness score for each game configuration scheme;

[0037] Analyze the game element attribute parameters in each game configuration scheme respectively to obtain a balance index of each game configuration scheme;

[0038] A search space for game configuration solutions is constructed, game element attribute parameters are used as optimization variables, the fitness score and the balance index are optimized, and the game element attribute parameters are dynamically adjusted.

[0039] Furthermore, the player game experience evaluation model is constructed based on the game player's character customization scheme and game behavior data, including:

[0040] Obtaining the character customization scheme and game behavior data of the game player, and generating a training data set through data preprocessing and feature engineering;

[0041] The supervised learning algorithm is used to input the training data set into the player game experience evaluation model for training, so that the player game experience evaluation model can evaluate the player game experience under different game configuration schemes and generate a fitness score.

[0042] In a second aspect, an embodiment of the present application provides a collaborative strategy generation system for a multiplayer game, comprising:

[0043] A first processing module is used to construct a player role generalization model, and assign a role customization plan to an initial player based on the player role generalization model;

[0044] A second processing module: used for obtaining a role participation mode of an initial player, allocating a game session to the initial player based on the role customization scheme and the role participation mode, and marking the initial player as a game player;

[0045] The third processing module is used for obtaining game behavior data of each game player in the same game session, analyzing the game behavior data, and obtaining a cooperation mode of the game players;

[0046] The fourth processing module is used to dynamically adjust the game element attribute parameters according to the role customization scheme, game behavior data and cooperation mode of the game player;

[0047] A fifth processing module: used for inputting the real-time game behavior data and game element attribute parameters of the game players into a pre-trained deep learning model to generate collaborative strategy recommendations;

[0048] The sixth processing module is used to provide the collaboration strategy suggestion to the corresponding game player, and dynamically adjust the resource allocation of each game player according to the collaboration strategy suggestion.

[0049] In a third aspect, an embodiment of the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for generating collaborative strategies for multiplayer games when executing the computer program.

[0050] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0051] The present application discloses a method for generating collaborative strategies for multiplayer games. By constructing a player role generalization model, the character customization data generated by historical players in a multiplayer game system is generalized, so that the commonalities and differences of player preferences can be discovered, thereby providing a more accurate user portrait. Based on the above generalization results, a character customization scheme that is more in line with their personal preferences is provided to new players or returning players, thereby improving the personalized experience of the game. In addition, by obtaining the game behavior data of multiple game players in the same game session and conducting in-depth analysis of these data, the collaborative mode of the game players can be identified and extracted, thereby helping the players to better understand each other's behavioral tendencies and collaborative methods in the game, thereby adjusting their own strategies, improving the efficiency of team collaboration, and ultimately improving the overall game experience. In addition, by comprehensively considering the game player's role customization scheme, game behavior data and collaborative mode, the game element attribute parameters are dynamically adjusted, thereby dynamically adjusting the configuration of the game background, which helps to optimize the difficulty and balance of the game. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present application, 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0053] Figure 1 It is a flowchart of a method for generating a collaborative strategy for a multiplayer game provided by an embodiment of the present invention;

[0054] Figure 2 It is a structural diagram of a collaborative strategy generation system for multiplayer games provided by one embodiment of the present invention;

[0055] Figure 3 It is a structural diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0056] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0057] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0058] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0059] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0060] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0061] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0062] See also Figure 1 As shown, the present invention is a method for generating a collaborative strategy for a multiplayer game, comprising the following steps:

[0063] S100, constructing a player role generalization model, and assigning a role customization plan to an initial player based on the player role generalization model;

[0064] The collaborative strategy generation method of the present application is applied to a multiplayer game system, and can provide collaborative strategy suggestions and reallocation of system resources for the gaming process of a group game played by multiple people, ensuring that each player can obtain resources that match their contribution, thereby improving the strategy, depth and fun of the game, while optimizing the allocation of system resources and game performance, bringing players a better and richer gaming experience.

[0065] In this application, by constructing a player character generalization model and generalizing the character customization data generated by historical players in a multiplayer game system, it is possible to discover the commonalities and differences in player preferences, and thus provide a more accurate user portrait. Based on the above generalization results, new or returning players can be provided with character customization solutions that are more in line with their personal preferences, thereby enhancing the personalized experience of the game.

[0066] In some embodiments, constructing the player character generalization model includes:

[0067] Based on the character customization data of historical players, a clustering algorithm is used to classify the historical players to obtain the character customization preference characteristics of different player groups, wherein the historical players are players who have played multiplayer games;

[0068] For each of the player groups, an association rule mining algorithm is used to analyze the association constraint relationship between the player role attributes to form a role attribute combination rule library;

[0069] Based on the role customization preference characteristics and the role attribute combination rule base, a genetic algorithm is used to generate diversified role customization solutions.

[0070] In this embodiment, the character customization data is determined by the type of multiplayer game. For example, for competitive multiplayer games, the character customization data includes character attributes and skill attributes, wherein the character attributes include the character's occupation, gender and age, and the skill attributes include the skill type, skill level and skill effect.

[0071] In this embodiment, by generalizing the character customization data of historical players, key features and commonalities are extracted to form a player character generalization model. The player character generalization model can generate diverse character customization plans for initial players, where the initial players are players who are not assigned to the corresponding team.

[0072] In this embodiment, the character customization data of historical players is obtained, the above character customization data is preprocessed, outliers are cleaned, missing values ​​are filled, and data integrity is ensured. The K-means clustering algorithm is used to group the players according to the above character customization data to obtain the character customization preference characteristics of different player groups.

[0073] Specifically, taking the above-mentioned competitive multiplayer game as an example, the Euclidean distance of each player is calculated according to the attributes such as the character's occupation, gender and age, and the initial cluster center is determined. Specifically, the above-mentioned character occupation, gender and age are attribute-encoded to obtain the corresponding attribute values. For each historical player, a feature vector is constructed according to the customized character attributes of the character occupation, gender and age. For example, the feature vector of a historical player may be [1 (warrior), 0 (male), 25 (age)]. The Euclidean distance formula is used to calculate the distance between the player and each cluster center. The specific formula is: d(x, y) = sqrt((x1-y1) 2 +(x2-y2) 2 +...+(xn-yn) 2 ), where x and y represent the n-dimensional vectors of historical players and cluster centers respectively, x = (x1, x2, ..., xn), y = (y1, y2, ..., yn), each historical player is assigned to the cluster to which the nearest cluster center belongs, and then the center of each cluster is recalculated, that is, the average value of the attributes of all players in the cluster is calculated as the new cluster center, through multiple iterations, until the cluster center no longer changes significantly or reaches the preset number of iterations, multiple clusters are determined, each cluster represents a player group. For each player group, the character customization preference characteristics of the players in the group are extracted, such as occupation preference, skill configuration preference, etc. Specifically, based on skill attributes such as skill type, level, and effect, the above player groups are further subdivided, and the hierarchical clustering algorithm is used to optimize the preliminary clustering results to improve the accuracy of group division.

[0074] In one embodiment, a player character customization preference feature library is constructed for each of the above-mentioned player groups to record the preference features of each player group and provide data support for subsequent character recommendations and personalized services.

[0075] In this embodiment, the association rule mining algorithm is used to analyze the association of the role customization of different players in the player group, discover potential role customization trends, and provide reference for game design. Specifically, the Apriori algorithm is used to mine the data of the role attributes between the two player groups, set the minimum support threshold and the minimum confidence threshold, and identify frequent item sets; based on the frequent item sets, association rules are generated to determine the association constraint relationship between different role attributes to form a preliminary rule base; through the preset rationality evaluation model, the rules in the preliminary rule base are rationally verified, unreasonable rules are eliminated, and reasonable rules are retained; combined with the characteristics of the player group, the retained reasonable rules are refined and optimized to form a role attribute combination rule base for different player groups.

[0076] In this embodiment, in order to further optimize the combination of each character attribute in the character attribute combination rule library, a genetic algorithm is used to optimize the combination of character attributes and generate diversified character customization schemes. Specifically, the character attribute combination rule library contains value ranges and combination constraints of different character attributes. The character attributes and constraints in the character attribute combination rule library are converted into genetic codes. Each gene bit represents the value of a character attribute. Multiple gene bits form a chromosome, which represents a complete character attribute combination scheme. Then, an initial population is randomly generated, and the fitness of each chromosome individual is calculated by a fitness function, wherein the fitness function needs to consider the rationality of the character attribute combination and the player preference matching degree; The initial population is subjected to selection, crossover and mutation operations; the selection operation selects individuals with higher fitness from the current population; the crossover operation recombines the chromosomes of two parent individuals in a certain way to generate new offspring individuals; the mutation operation randomly changes certain gene positions of individuals to introduce new attribute combinations; the newly generated individuals are screened according to fitness, the individuals with lower fitness are eliminated, and the individuals with higher fitness are retained to enter the next generation population; until the preset number of iterations is reached or the termination condition is met; during the iteration process, the individuals with the highest fitness that have appeared are recorded as candidate optimal solutions; the optimal solution obtained by the genetic algorithm is decoded into a character attribute combination scheme to generate a character customization scheme that meets the player's preference characteristics.

[0077] In one embodiment, the rationality of the character attribute combination and the player preference matching degree need to be considered in the process of setting the fitness function of the above-mentioned genetic algorithm. Specifically, the rationality score of the character attribute combination is calculated, and the specific calculation formula is as follows:

[0078]

[0079] Among them, RS represents the rationality score, w i represents the role attribute weight, S i Indicates the player's attributes for the character, R i represents the ideal value of the character's attribute, P(S i ) is a compensation term that represents the proportional relationship of attribute weights;

[0080] Calculate the matching score between the character attribute combination and the player preference vector. The specific calculation formula is as follows:

[0081]

[0082] Among them, MS represents the matching score, w i represents the role attribute weight, S i Indicates the player's attributes for the character, M i Indicates the expected value of the character's attribute;

[0083] Finally, RS and MS are weighted to obtain the fitness function:

[0084] FS=α×RS+β×MS

[0085] Wherein, α and β represent the weight coefficients of RS and MS in FS respectively, and α+β=1.

[0086] In some embodiments, the assigning of a role customization scheme to the initial player based on the player role generalization model includes:

[0087] Acquire historical character customization data of an initial player, and determine the player group of the initial player through a player character preference model;

[0088] The player character generalization model generates a plurality of character customization schemes through the corresponding player group of the initial player;

[0089] An attribute balance check is performed on the plurality of character customization schemes, and a character customization scheme is allocated to the initial player by adjusting the character attribute combinations in the character customization schemes.

[0090] In this embodiment, the ID of the initial player and his historical role customization data are obtained, the historical role customization data is input into the player role preference model, the player group of the initial player is output, and multiple customization schemes are generated based on the player group and the player role generalization model. The player role preference model is pre-trained, and is a model that can identify the player group preferences of different players after modeling the historical role customization data of a large number of historical players.

[0091] In this embodiment, a property balance check is performed on the generated multiple character customization schemes to determine whether the values ​​of the character attributes in each scheme are balanced, and then the character attribute customization scheme is further adjusted to ensure the balance of the character attributes in the final character customization scheme that can be selected by the player. Specifically, the first value of each character attribute in the character customization scheme is obtained, and a character attribute value matrix is ​​constructed; the deviation between the first value of each character attribute and the attribute balance threshold is calculated by a preset attribute balance threshold, and a deviation list is generated; by determining whether there is an attribute exceeding the attribute balance threshold in the deviation list, the character attribute is dynamically adjusted to generate a second value. For the second value generated after the dynamic adjustment, a new attribute value matrix is ​​regenerated, the balance of the new attribute value matrix is ​​calculated, and it is determined whether it meets the preset attribute balance threshold requirement. If not, the dynamic adjustment process is repeated until a balanced state is reached, and the character customization scheme that finally meets the balance requirement is output to the initial player, so that the initial player can select the corresponding character customization scheme, and the attribute change data during the adjustment process is recorded for subsequent analysis.

[0092] S200, obtaining a role participation mode of an initial player, assigning a game session to the initial player based on the role customization scheme and the role participation mode, and marking the initial player as a game player;

[0093] In some embodiments, the role participation mode includes a group joining mode and a group building mode;

[0094] The allocating a game session to the initial player comprises:

[0095] For the group joining method, based on the character customization scheme, the initial player is added to a game session of an existing multiplayer game;

[0096] For the group building method, a new game session is created, and game players are assigned to the new game session based on the role customization plan.

[0097] In this embodiment, newly joined game players can choose corresponding role participation methods according to their own ideas, and then match corresponding game sessions and role customization plans for game players, thereby improving player satisfaction and loyalty. Specifically, role participation methods include joining a team and establishing a team.

[0098] In this embodiment, game players belonging to the same team are added to the same game session to ensure that multiple players can interact online at the same time and share game resources and status. Therefore, for the initial players who choose to join the team, they can directly join any existing game session according to their selected character customization plan to realize joining the team in the multiplayer game. For the initial players who choose to build a team, a new game session is created, and then game players are assigned to the new game session based on their selected character customization plan to form a new group.

[0099] S300, for multiple game players in the same game session, obtaining game behavior data of each game player, analyzing the game behavior data, and obtaining a collaboration mode of the game players;

[0100] In this embodiment, by obtaining the game behavior data of multiple game players in the same game session and conducting in-depth analysis of these data, it is possible to identify and extract the game players' collaboration patterns, thereby helping players to better understand each other's behavioral tendencies and collaboration methods in the game, thereby adjusting their own strategies, improving the efficiency of team collaboration, and ultimately enhancing the overall game experience.

[0101] In some embodiments, the step S300 includes:

[0102] According to the game behavior data of each game player, the frequency of collaboration between each game player and other game players is counted to obtain the player collaboration frequency matrix;

[0103] Based on the player collaboration frequency matrix, cluster analysis is used to obtain the collaboration mode between different game players;

[0104] Analyze the multiple collaboration modes to obtain collaboration effect indicators of each collaboration mode;

[0105] Based on the collaboration effect index, an association rule mining algorithm is used to obtain the collaboration mode of the role customization schemes of different game players.

[0106] In this embodiment, in the game data analysis, firstly, by collecting the game behavior data of the players, the frequency of collaboration between the game players is counted using data mining technology, and a player collaboration frequency matrix is ​​constructed. For example, if player A and player B collaborate together 10 times in a game, then the value between A and B in the matrix is ​​10. Next, a hierarchical clustering algorithm is used to classify the collaboration patterns between the game players. Specifically, by calculating the similarity of the collaboration frequencies between the game players, players with similar collaboration patterns are clustered together, thereby identifying collaboration patterns such as "quick raid", "long-term defense", and "resource grabbing".

[0107] In this embodiment, according to the player collaboration effect index data, an association rule mining algorithm is used to mine the association rules between different player roles and skill combinations and collaboration effects. According to the mined association rules, the best collaboration mode under different player roles and skill combinations is determined. According to the best collaboration mode, the player skill combination strategy is optimized to obtain the optimized player role skill combination combination.

[0108] In some embodiments, the analyzing of the plurality of collaboration modes to obtain the collaboration effect index of each collaboration mode includes:

[0109] Obtaining team performance data and collaboration frequency corresponding to each collaboration in each of the collaboration modes;

[0110] Inputting the team performance data, the collaboration frequency and the collaboration mode into a pre-trained collaboration effect evaluation model to obtain the influence of different collaboration modes on the collaboration effect;

[0111] Based on the influence rules, the effect weights of the collaboration effects of various collaboration modes are determined, and collaboration effect indicators are generated.

[0112] In this embodiment, after each game player collaboration ends, the team performance data, collaboration frequency and collaboration method generated by the collaboration are automatically recorded and used as the data basis for obtaining collaboration effect indicators.

[0113] In this embodiment, the team performance data, the frequency of collaboration between each game player and the corresponding collaboration mode in the team collaboration process are obtained. The team performance data includes the resource collection, enemy killing and task completion of the game players in the team, etc. The above data are preprocessed to remove repeated, erroneous or incomplete data records, and the key collaboration features in the original data are extracted and used as input variables to construct a collaboration effect evaluation model. The above key collaboration features include collaboration frequency features, collaboration mode features and collaboration effect features. The collaboration frequency feature is a player collaboration frequency matrix constructed based on the collaboration frequency between game players. The collaboration mode feature is to encode different collaboration modes to obtain a collaboration mode matrix. The collaboration effect feature is to quantify various performance conditions of the team performance data to generate a collaboration effect matrix. Based on the nature of the problem of the influence of different collaboration modes on the collaboration effect and the availability of key collaboration features, a suitable evaluation model, such as a linear regression model, is selected. Through the extracted data of multiple groups of team collaboration processes, a feature data set containing multiple key collaboration features is generated, and a collaboration effect evaluation model for predicting the team collaboration effect is constructed. Through this collaboration effect evaluation model, the influence of different collaboration modes on the collaboration effect can be obtained.

[0114] In this embodiment, based on the influence of different collaboration modes on the collaboration effect, an expert scoring method is used to construct a judgment matrix between different collaboration modes and collaboration effects. The judgment matrix should be able to reflect the importance of each collaboration mode to the collaboration effect. A weight calculation method is used to calculate the weight of each collaboration mode according to the judgment matrix, thereby quantifying the influence of different collaboration modes on the collaboration effect. Based on the determined effect weight and collaboration effect definition, a collaboration effect index system is constructed, and the above-mentioned collaboration effect index system includes quantitative indicators of various collaboration effects, that is, the effect weight corresponding to each collaboration effect.

[0115] S400, dynamically adjusting game element attribute parameters according to the game player's role customization scheme, game behavior data, and collaboration mode;

[0116] In this embodiment, by comprehensively considering the game players' character customization schemes, game behavior data, and collaboration patterns, the game element attribute parameters are dynamically adjusted, thereby dynamically adjusting the configuration of the game background, which helps to optimize the difficulty and balance of the game.

[0117] In one embodiment, the step S400 includes:

[0118] Based on the character customization scheme and game behavior data of the game player, a player game experience evaluation model is constructed;

[0119] Extracting game element attribute parameters from a plurality of preset game configuration schemes, inputting the game element attribute parameters into the player game experience evaluation model, and obtaining a fitness score for each game configuration scheme;

[0120] Analyze the game element attribute parameters in each game configuration scheme respectively to obtain a balance index of each game configuration scheme;

[0121] A search space for game configuration solutions is constructed, game element attribute parameters are used as optimization variables, the fitness score and the balance index are optimized, and the game element attribute parameters are dynamically adjusted.

[0122] In this embodiment, the game behavior data and character customization data of the player are obtained, and a training data set of the player game experience evaluation model is constructed through data preprocessing and feature engineering. The player game experience evaluation model is trained using a supervised learning algorithm to obtain model parameters, and the game element attribute parameters in each preset game configuration scheme are extracted, where the game element attribute parameters include maps, monsters, items, and weather. The above game element attribute parameters are input into the trained player game experience evaluation model to obtain the fitness score of the configuration scheme.

[0123] Specifically, when creating a game session for a group of players, the collaborative strategy generation method for multiplayer games selects a game configuration scheme from multiple preset game configuration schemes as the game background of the game, wherein the game configuration scheme includes game element attribute parameters, and the game element attribute parameters include maps, monsters, items, and weather. The map includes terrain type, map size, resource distribution, and environmental effects; monsters include monster types and levels, behavior patterns, and refresh mechanisms; items include equipment, consumables, quest items, and special items; weather includes weather types, weather changes, weather effects, and weather and monster / quest associations.

[0124] In this embodiment, the balance index includes a difficulty balance index, a resource allocation balance index and a player role balance relationship index, such as calculating the ratio between the monster's health and the skill level of the skill attribute as the difficulty balance index, and calculating the variance of the character attributes of different players as the role balance relationship index.

[0125] In this embodiment, a search space for game configuration schemes is constructed, and fitness and balance are used as optimization targets. The simulated annealing algorithm is used to explore the search space, and the optimal game configuration scheme is found by dynamically adjusting the game element attribute parameters in the game configuration scheme, such as the initial temperature, cooling rate, and termination condition. Taking the ambient temperature as an example, in each iteration, a new game configuration scheme is generated according to the current temperature, and its comprehensive indicators of fitness and balance are calculated. If the comprehensive indicators of the new scheme are better than the current optimal scheme, it will be used as the new optimal scheme; otherwise, according to the Metropolis criterion, a worse scheme is accepted with a certain probability to jump out of the local optimum. As the temperature gradually decreases, the algorithm will gradually converge to the vicinity of the global optimal solution. When the termination condition is met, the current optimal configuration scheme combination is output as the latest environmental configuration scheme for the game.

[0126] In some embodiments, constructing a player game experience evaluation model based on the game player's character customization scheme and game behavior data includes:

[0127] Obtaining the character customization scheme and game behavior data of the game player, and generating a training data set through data preprocessing and feature engineering;

[0128] The supervised learning algorithm is used to input the training data set into the player game experience evaluation model for training, so that the player game experience evaluation model can evaluate the player game experience under different game configuration schemes and generate a fitness score.

[0129] In this embodiment, the construction process of the player game experience evaluation model is as follows: determine the evaluation indicators of the player game experience evaluation, the player character customization plan, and the player game behavior data, and build a model framework that can comprehensively evaluate the player game experience based on the evaluation indicators. The above model framework includes multiple sub-models, each of which is responsible for evaluating one or more specific evaluation indicators, and uses historical data to train the model, and then optimizes the model based on the training results. Extract the game element attribute parameters from the preset multiple game configuration schemes, and input the extracted game element attribute parameters into the constructed player game experience evaluation model. The above player game experience evaluation model calculates the fitness score of each game configuration scheme based on the input game element attribute parameters, thereby reflecting the pros and cons of the scheme in terms of the player game experience.

[0130] S500, inputting the real-time game behavior data and game element attribute parameters of the game player into a pre-trained deep learning model to generate a collaborative strategy suggestion;

[0131] In this embodiment, by collecting historical game behavior data of game players, time series analysis methods such as long short-term memory networks are used to identify player behavior patterns. For example, if a player frequently fails but continues to try when facing difficult levels, it is identified that the player has a high spirit of challenge and persistence. The above behavior patterns are associated with game element attribute parameters, and a convolutional neural network is used to analyze the background environment to capture the key features of the background environment during the game. Based on the above analysis, the deep learning model can predict the collaboration strategies that players may adopt, and ultimately generate collaboration strategy recommendations, thereby improving overall game performance and satisfaction.

[0132] In this embodiment, the real-time game behavior data and game element attribute parameters of the players are intelligently analyzed through the deep learning model, which can generate more accurate and efficient collaborative strategy suggestions. These strategy suggestions are not only based on the current status of the players, but also take into account the overall dynamics of the game environment, thereby improving the scientificity and practicality of the strategy. At the same time, the game environment can be dynamically adjusted through the deep learning model. In turn, game resources can be allocated more reasonably, enhancing the game experience of game players and improving players' satisfaction and loyalty to the game.

[0133] S600: Provide the collaboration strategy suggestion to corresponding game players, and dynamically adjust resource allocation of each game player according to the collaboration strategy suggestion.

[0134] It can be understood that the generated collaboration strategy suggestions are provided to the corresponding game players, and each game player can obtain at least one collaboration strategy suggestion. Different game players may adopt different collaboration strategy suggestions for the above multiple collaboration strategy suggestions. Therefore, this embodiment also needs to dynamically adjust the resource allocation of each game player according to the collaboration strategy suggestions to maintain the overall balance of the game.

[0135] Specifically, a game theory model is used to construct a multi-player strategy combination matrix, multiple collaboration strategy suggestions from each game player are combined into a strategy set, each game player selects a collaboration strategy suggestion from the corresponding strategy set, and the collaboration strategy suggestions selected by multiple game players are combined to generate a strategy combination.

[0136] In this embodiment, the above-mentioned strategy set is preliminarily estimated through the profit function, and finally the profit of each game player under different strategy sets is obtained through iterative calculation, and the profit data of the strategy set after each iteration is recorded. Using the definition of Nash equilibrium, it is determined whether there is a strategy combination that prevents any single game player from obtaining higher profits by unilaterally changing the strategy. If it exists, the strategy combination is determined to be a Nash equilibrium point. The resource allocation plan corresponding to the Nash equilibrium point is used as a reference to ensure the fairness of each orderly player in resource allocation and avoid excessive advantages or disadvantages of a single player. Through the preset resource allocation adjustment algorithm, the initial resource allocation plan is optimized to make it close to the allocation state corresponding to the Nash equilibrium point, and finally the optimized resource allocation plan is output.

[0137] Specifically, when constructing a multi-player strategy combination matrix, assuming that there are 3 players and each player has 3 strategy choices, the strategy combination matrix is ​​a 3×3×3 three-dimensional matrix. For player 1, its strategy set is {A, B, C}, and the corresponding payoff function can be expressed as U1(A)=2x1+3x2, U1(B)=x1+4x2, U1(C)=3x1+x2, where x1 and x2 are the allocations of resource 1 and resource 2, respectively. Similarly, the payoff functions of player 2 and player 3 can be expressed as U2(A)=3x1+2x2, U2(B)=2x1+3x2, U2(C)=x1+4x2 and U3(A)=4x1+x2, U3(B)=2x1+3x2, U3(C)=3x1+2x2, respectively. Through Monte Carlo simulation, 10,000 sets of resource allocation schemes are randomly generated, and the profit value of each player under different strategy combinations is calculated and recorded in the strategy profit matrix. According to the definition of Nash equilibrium, by exhaustively searching the strategy profit matrix, it is found that when player 1 chooses strategy B, player 2 chooses strategy C, and player 3 chooses strategy A, no single player can obtain higher profits by changing strategies. Therefore, this strategy combination (B, C, A) is a Nash equilibrium point. The resource allocation scheme corresponding to the Nash equilibrium point (x1=4, x2=6) is taken as the optimization target, and a resource allocation adjustment algorithm is designed. The resource allocation adjustment algorithm takes the initial resource allocation scheme as the starting point, and continuously adjusts the resource allocation ratio through the gradient descent method to gradually approach the target allocation scheme. After 100 iterations, the optimized resource allocation scheme (x1=38, x2=62) is finally obtained, which is close to the allocation scheme corresponding to the Nash equilibrium point. The allocation scheme is then output as the resource allocation scheme of the game.

[0138] See also Figure 2 As shown, the present invention also provides a collaborative strategy generation system for multiplayer games, the system comprising:

[0139] The first processing module 201 is used to construct a player role generalization model and assign a role customization plan to an initial player based on the player role generalization model;

[0140] The second processing module 202 is used to obtain the role participation mode of the initial player, allocate a game session to the initial player based on the role customization scheme and the role participation mode, and mark the initial player as a game player;

[0141] The third processing module 203 is used for obtaining game behavior data of each game player in the same game session, analyzing the game behavior data, and obtaining a cooperation mode of the game players;

[0142] The fourth processing module 204 is used to dynamically adjust the game element attribute parameters according to the role customization scheme, game behavior data and cooperation mode of the game player;

[0143] The fifth processing module 205 is used to input the real-time game behavior data and game element attribute parameters of the game player into the pre-trained deep learning model to generate a collaborative strategy suggestion;

[0144] The sixth processing module 206 is used to provide the collaboration strategy suggestion to the corresponding game player, and dynamically adjust the resource allocation of each game player according to the collaboration strategy suggestion.

[0145] It is understandable that if Figure 1 The contents of the collaborative strategy generation method embodiment for multiplayer games shown in the figure are all applicable to the collaborative strategy generation system embodiment for multiplayer games. The functions specifically implemented by the collaborative strategy generation system embodiment for multiplayer games are similar to those of the embodiment shown in the figure. Figure 1 The embodiment of the collaborative strategy generation method for multiplayer games shown in FIG. 1 is the same as that shown in FIG. 1 , and the beneficial effects achieved are the same as those of FIG. Figure 1 The beneficial effects achieved by the embodiment of the cooperative strategy generation method for multiplayer games shown are also the same.

[0146] It should be noted that the information interaction, execution process and other contents between the above-mentioned systems are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0147] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0148] See also Figure 3 As shown, an embodiment of the present invention further provides a computer device 3, comprising: a memory 302 and a processor 301 and a computer program 303 stored on the memory 302. When the computer program 303 is executed on the processor 301, a collaborative strategy generation method for multiplayer games as described in any one of the above methods is implemented.

[0149] The computer device 3 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer device 3 may include, but is not limited to, a processor 301 and a memory 302. Those skilled in the art will appreciate that Figure 3 It is only an example of computer device 3 and does not constitute a limitation on computer device 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components, for example, it may also include input and output devices, network access devices, etc.

[0150] The processor 301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0151] In some embodiments, the memory 302 may be an internal storage unit of the computer device 3, such as a hard disk or memory of the computer device 3. In other embodiments, the memory 302 may also be an external storage device of the computer device 3, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device 3. Further, the memory 302 may also include both an internal storage unit of the computer device 3 and an external storage device. The memory 302 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program, etc. The memory 302 may also be used to temporarily store data that has been output or is to be output.

[0152] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for generating a collaborative strategy for a multiplayer game as described in any one of the above methods is implemented.

[0153] In this embodiment, if the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / computer device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, USB flash drive, mobile hard disk, disk or optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0154] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for generating a collaborative strategy for a multiplayer game, characterized in that: include: Constructing a player role generalization model, and assigning a role customization scheme to an initial player based on the player role generalization model; Acquire a role participation mode of an initial player, assign a game session to the initial player based on the role customization scheme and the role participation mode, and mark the initial player as a game player; For multiple game players in the same game session, obtain game behavior data of each game player, analyze the game behavior data, and obtain a collaboration mode of the game players; Dynamically adjust game element attribute parameters according to the game player's role customization plan, game behavior data, and collaboration mode; Inputting the real-time game behavior data and game element attribute parameters of the game players into a pre-trained deep learning model to generate collaborative strategy recommendations; The collaboration strategy suggestion is provided to the corresponding game player, and the resource allocation of each game player is dynamically adjusted according to the collaboration strategy suggestion.

2. The method according to claim 1, characterized in that The constructing of the player role generalization model includes: Based on the character customization data of historical players, a clustering algorithm is used to classify the historical players to obtain the character customization preference characteristics of different player groups, wherein the historical players are players who have played multiplayer games; For each of the player groups, an association rule mining algorithm is used to analyze the association constraint relationship between the player role attributes to form a role attribute combination rule library; Based on the role customization preference characteristics and the role attribute combination rule base, a genetic algorithm is used to generate diversified role customization solutions.

3. The method according to claim 2, characterized in that The step of allocating a role customization scheme to the initial player based on the player role generalization model includes: Acquire historical character customization data of an initial player, and determine the player group of the initial player through a player character preference model; The player character generalization model generates a plurality of character customization schemes through the corresponding player group of the initial player; An attribute balance check is performed on the plurality of character customization schemes, and a character customization scheme is allocated to the initial player by adjusting the character attribute combinations in the character customization schemes.

4. The method according to claim 1, characterized in that The role participation methods include joining a group and establishing a group; The allocating a game session to the initial player comprises: For the group joining method, based on the character customization scheme, the initial player is added to a game session of an existing multiplayer game; For the group building method, a new game session is created, and game players are assigned to the new game session based on the role customization plan.

5. The method according to claim 1, characterized in that The method of acquiring game behavior data of each game player for multiple game players in the same game session, analyzing the game behavior data, and obtaining the cooperation mode of each game player includes: According to the game behavior data of each game player, the frequency of collaboration between each game player and other game players is counted to obtain the player collaboration frequency matrix; Based on the player collaboration frequency matrix, cluster analysis is used to obtain the collaboration mode between different game players; Analyze the multiple collaboration modes to obtain collaboration effect indicators of each collaboration mode; Based on the collaboration effect index, an association rule mining algorithm is used to obtain the collaboration mode of the role customization schemes of different game players.

6. The method according to claim 5, characterized in that The analyzing of the plurality of collaboration modes to obtain collaboration effect indicators of each collaboration mode includes: Obtaining team performance data and collaboration frequency corresponding to each collaboration in each of the collaboration modes; Inputting the team performance data, the collaboration frequency and the collaboration mode into a pre-trained collaboration effect evaluation model to obtain the influence of different collaboration modes on the collaboration effect; Based on the influence rules, the effect weights of the collaboration effects of various collaboration modes are determined, and collaboration effect indicators are generated.

7. The method according to claim 1, characterized in that The dynamically adjusting the game element attribute parameters according to the game player's role customization scheme, game behavior data, and collaboration mode includes: Based on the character customization scheme and game behavior data of the game player, a player game experience evaluation model is constructed; Extracting game element attribute parameters from a plurality of preset game configuration schemes, inputting the game element attribute parameters into the player game experience evaluation model, and obtaining a fitness score for each game configuration scheme; Analyze the game element attribute parameters in each game configuration scheme respectively to obtain a balance index of each game configuration scheme; A search space for game configuration solutions is constructed, game element attribute parameters are used as optimization variables, the fitness score and the balance index are optimized, and the game element attribute parameters are dynamically adjusted.

8. The method according to claim 7, characterized in that The constructing of a player game experience evaluation model based on the game player's character customization scheme and game behavior data includes: Obtaining the character customization scheme and game behavior data of the game player, and generating a training data set through data preprocessing and feature engineering; The supervised learning algorithm is used to input the training data set into the player game experience evaluation model for training, so that the player game experience evaluation model can evaluate the player game experience under different game configuration schemes and generate a fitness score.

9. A collaborative strategy generation system for multiplayer games, characterized in that: include: A first processing module is used to construct a player role generalization model, and assign a role customization plan to an initial player based on the player role generalization model; A second processing module: used for obtaining a role participation mode of an initial player, allocating a game session to the initial player based on the role customization scheme and the role participation mode, and marking the initial player as a game player; The third processing module is used for obtaining game behavior data of each game player in the same game session, analyzing the game behavior data, and obtaining a cooperation mode of the game players; The fourth processing module is used to dynamically adjust the game element attribute parameters according to the role customization scheme, game behavior data and cooperation mode of the game player; A fifth processing module: used for inputting the real-time game behavior data and game element attribute parameters of the game players into a pre-trained deep learning model to generate collaborative strategy recommendations; The sixth processing module is used to provide the collaboration strategy suggestion to the corresponding game player, and dynamically adjust the resource allocation of each game player according to the collaboration strategy suggestion.

10. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 8 is implemented.

Images