Game equipment configuration method and device, equipment and storage medium

Abstract

The invention provides a game equipment configuration method and device, equipment and a storage medium, and the method comprises the steps: obtaining real-time game data of a current game session and an initial equipment scheme set, screening from the initial equipment scheme set according to the real-time game data to obtain a target equipment scheme, and updating the current equipment according to the target equipment scheme. According to the technical scheme, screening is carried out from a large number of equipment schemes in combination with the real-time game data, the equipment scheme most suitable for the current game environment and the player operation style can be rapidly determined, the equipment scheme recommendation accuracy is improved, and the equipment scheme selection efficiency of players is also improved for complex and changeable game scenes.

Description

Technical Field

[0001] This application relates to the technical field of electronic games, and more particularly, to a method, device, equipment, and storage medium for configuring game equipment. Background Art

[0002] In shooting games, players need to change the game equipment configuration according to the real-time battlefield environment. Among them, the process of changing the game equipment configuration includes: selecting an equipment combination, selecting equipment, and adjusting weapon attributes.

[0003] In the prior art, in the game equipment system, generally, multiple optional equipment combinations are pre-configured for players to select, or a static decision tree model is used to select multiple selectable equipment combinations.

[0004] However, the pre-configured equipment combinations are difficult to adapt to the complex changes in the battlefield environment, and the static decision tree model is difficult to perceive the real-time game environment. Therefore, there is also a problem of low accuracy in the recommendation of equipment combinations. Summary of the Invention

[0005] The purpose of this application is to provide a method, device, equipment, and storage medium for configuring game equipment to solve the problem of low accuracy in the recommendation of equipment combinations in the prior art for the above-mentioned deficiencies in the prior art.

[0006] To achieve the above purpose, the technical solution adopted in this application is as follows:

[0007] In a first aspect, this application provides a method for configuring game equipment, the method including:

[0008] Obtaining real-time game data of the current game session and an initial equipment plan set, the real-time game data including at least one of the following: game environment data, equipment data, and player operation data;

[0009] Screening a target equipment plan from the initial equipment plan set according to the real-time game data, and updating the current equipment according to the target equipment plan.

[0010] In a second aspect, this application provides a device for configuring game equipment, the device including:

[0011] An obtaining module, configured to obtain real-time game data of the current game session and an initial equipment plan set, the real-time game data including at least one of the following: game environment data, equipment data, and player operation data;

[0012] A screening module, configured to screen a target equipment plan from the initial equipment plan set according to the real-time game data, and update the current equipment according to the target equipment plan.

[0013] In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to perform the steps of a configuration method for a game equipment as described in any one of the first aspect.

[0014] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is run by a processor, it executes the steps of a configuration method for a game equipment as described in any one of the first aspect.

[0015] The beneficial effects of the present application are as follows: Obtain the real-time game data of the current game session and the initial equipment plan set, screen out the target equipment plan from the initial equipment plan set according to the real-time game data, and update the current equipment according to the target equipment plan. By screening from a large number of equipment plans in combination with real-time game data, the equipment plan that is most suitable for the current game environment and the player's operation style can be quickly determined, improving the accuracy of equipment plan recommendation, and also improving the efficiency of the player's selection of equipment plans for complex and changeable game scenarios.

[0016] To make the above objects, features, and advantages of the present application more obvious and understandable, the following specifically enumerates preferred embodiments and, in conjunction with the accompanying drawings, details are described as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 Shows a flowchart of a configuration method for a game equipment provided by an embodiment of the present application;

[0019] Figure 2 Shows a flowchart of determining a target equipment plan provided by an embodiment of the present application;

[0020] Figure 3 Shows another flowchart of determining a target equipment plan provided by an embodiment of the present application;

[0021] Figure 4 Shows a flowchart of determining a player feature label provided by an embodiment of the present application;

[0022] Figure 5 Shows another flowchart for determining player feature tags provided by an embodiment of the present application;

[0023] Figure 6 Shows a schematic structural diagram of an artificial intelligence model provided by an embodiment of the present application;

[0024] Figure 7 Shows a flowchart for obtaining a first set of solutions provided by an embodiment of the present application;

[0025] Figure 8 Shows a flowchart for obtaining a second set of solutions provided by an embodiment of the present application;

[0026] Figure 9 Shows a flowchart for obtaining a target equipment solution provided by an embodiment of the present application;

[0027] Figure 10 Shows a flowchart for adjusting the target equipment solution when the price changes provided by an embodiment of the present application;

[0028] Figure 11 Shows a flowchart for adjusting the target equipment solution when the version is updated provided by an embodiment of the present application;

[0029] Figure 12 Shows a schematic structural diagram of a game equipment recommendation device provided by an embodiment of the present application;

[0030] Figure 13 Shows a schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Apparently, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and illustrated herein can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application to be protected, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0032] It should be noted that the term "including" will be used in the embodiments of the present application to indicate the presence of the features stated hereinafter, but does not exclude the addition of other features.

[0033] The configuration method of the gaming equipment in one embodiment of the present disclosure may be run on a local terminal device or a server. When the configuration method of the gaming equipment is run on a server, the method may be implemented and executed based on a cloud interaction system, wherein the cloud interaction system includes a server and a client device.

[0034] In an optional implementation, various cloud applications can be run under the cloud interaction system, such as cloud games. Taking cloud games as an example, cloud games refer to a game mode based on cloud computing. In the operation mode of cloud games, the operating body of the game program and the main body of the game screen presentation are separated. The storage and operation of the configuration method of the game equipment are completed on the cloud game server. The role of the client device is used for receiving and sending data and presenting the game screen. For example, the client device can be a display device with data transmission function close to the user side, such as a mobile terminal, a TV, a computer, a handheld computer, etc.; but the cloud game server in the cloud is used for information processing. When playing the game, the player operates the client device to send an operation instruction to the cloud game server. The cloud game server runs the game according to the operation instruction, encodes and compresses the game screen and other data, and returns it to the client device through the network. Finally, the client device decodes and outputs the game screen.

[0035] In an optional embodiment, taking a game as an example, a local terminal device stores a game program and is used to present a game screen. The local terminal device is used to interact with the player through a graphical user interface, that is, the game program is downloaded and installed by an electronic device and run conventionally. The local terminal device may provide the graphical user interface to the player in a variety of ways, for example, it may be rendered and displayed on a display screen of the terminal, or provided to the player through a holographic projection. For example, the local terminal device may include a display screen and a processor, the display screen is used to present a graphical user interface, the graphical user interface includes a game screen, and the processor is used to run the game, generate a graphical user interface, and control the display of the graphical user interface on the display screen.

[0036] In a possible implementation, an embodiment of the present invention provides a method for configuring gaming equipment, providing a graphical user interface through a terminal device, wherein the terminal device may be the local terminal device mentioned above, or may be a client device in the cloud interaction system mentioned above.

[0037] In shooting games, if the player needs to switch the game map, or the price or parameter information of the current weapon is updated, in order to ensure the winning rate of the game, the player needs to adjust the current equipment plan, such as adjusting the equipment combination in the equipment plan.

[0038] Currently, players can choose one of the multiple weapon combinations recommended by the system as their own equipment plan. However, the fixed equipment combinations limit the flexibility of the equipment plan and cannot adapt to the complex changes in the battlefield environment. In addition, players need to manually compare the equipment parameters of the multiple recommended equipment combinations, which also greatly reduces the decision-making efficiency of players.

[0039] In another implementation, an equipment plan can be generated through a static decision tree model. The specific process is as follows: The optimal weapon combination is generated through a static decision tree based on the parameter information of the current weapon. However, this method also lacks real-time perception of the game environment. Especially in scenarios where weapon prices can change and game maps change, the accuracy of the equipment plan generated by the static decision tree model is also relatively low.

[0040] In summary, based on the existing technology, there is a problem of relatively low accuracy when recommending equipment combinations in complex and changeable game scenarios.

[0041] Based on this, the present application proposes a configuration method for game equipment, which comprehensively judges according to real-time game environment data, equipment data, and the operation style of players, so as to recommend a more cost-effective equipment plan that suits the player's style and the current game environment. The present application can improve the accuracy of equipment plan recommendation in complex and changeable game scenarios and further reduce the decision-making cost of players for equipment plans.

[0042] Next, in combination with Figure 1 , the configuration method of the game equipment in the present application will be described. As Figure 1 shown, the method includes:

[0043] S101. Obtain the real-time game data of the current game session and the initial equipment plan set. The real-time game data includes: game environment data, equipment data, and player operation data.

[0044] After the player starts the current game session and enters the game map, the device can automatically obtain the real-time game data of the current game session and the player's current initial equipment plan set.

[0045] Optionally, the game environment data can be the objective conditions and environmental factors in the game scene, including map information, weather information, lighting information, and resource distribution information, etc.

[0046] Optionally, the equipment data can be the attribute data and equipment price data of the equipment in the game at the current time. The attribute data of the equipment includes the damage value, defense value, recoil, etc. of the equipment, and the equipment price data includes the purchase and sale prices of various equipment, including weapon prices, armor prices, consumable prices, etc.

[0047] Optionally, the player operation data can be the records of the player's behaviors and operations in the game, including the player's movement data, combat data, and interaction data. The player operation data can reflect the player's gaming style and behavior pattern.

[0048] Optionally, the initial equipment plan set includes multiple equipment plans, and the initial equipment plan set can be obtained by acquiring the equipment plans of other players from within the game community. Exemplarily, common popular equipment plans can be obtained from within the game community related to the game through an open API (Application Programming Interface) interface to obtain the initial equipment plan set. The equipment plan is used to indicate the weaponry equipped by the player and the parameter information of each piece of equipment, including the attribute parameters of the equipment and the price of each piece of equipment.

[0049] Among them, when generating the initial equipment plan set, the data can be filtered for outliers first, and then the equipment plans can be initially screened in combination with the time information, game version information, and map information of each equipment plan. The equipment plans that are the same as the current game session version and map recently are added to the initial equipment plan set. If the equipment plan is a plan that has been updated within N hours, it is regarded as a recent equipment plan.

[0050] In another implementation, the player can also issue a natural language instruction to interact with the device when needing to update the equipment plan, and at this time the player is either in the game process or before the game starts. After the device receives the player's natural language instruction, it can parse the characteristics of the equipment plan from the instruction and generate the corresponding equipment plan. For example, if the player issues a voice instruction "Please generate a silent plan suitable for the rainforest map", the device can obtain the game environment data corresponding to the rainforest map, the current equipment data, and the player's historical operation data, and obtain the initial equipment plan set corresponding to the rainforest map and the current game version, and generate an equipment plan suitable for the player based on the real-time game data and the initial equipment plan set.

[0051] S102. Screen the target equipment plan from the initial equipment plan set according to the real-time game data, and update the current equipment according to the target equipment plan.

[0052] It should be noted that the real-time game data can not only reflect the player's gaming style, but also reflect the real-time price of the equipment in the game and the real-time environment of the game. Therefore, based on the real-time game data, the optimal equipment plan that conforms to the player's gaming style can be screened from the initial equipment plan set as the target equipment plan.

[0053] Optionally, the initial set of equipment plans includes multiple equipment plans, which include equipment plans that the user can use and equipment plans that the user cannot use. In one possible implementation, multiple screening conditions such as the player's budget limit, weapon assembly weight threshold, and weapon unlocking level can be determined from the real-time game data, and based on the above screening conditions, the equipment plans that the player can use can be screened out from the initial set of equipment plans, and then combined with the real-time game data, the most suitable equipment plan for the player's style can be selected from the equipment plans that the player can use as the target equipment plan.

[0054] Exemplarily, equipment plans that exceed the player's budget limit, equipment plans that exceed the weapon assembly weight threshold, and equipment plans that exceed the weapon unlocking level can be deleted from the initial equipment plans to obtain a candidate plan set, and then the target equipment plan can be further screened out from the candidate plan set.

[0055] Among them, updating the current equipment according to the target equipment plan can be to replace the equipment that does not meet the requirements in the player's current equipment, or to upgrade, strengthen or adjust the configuration of the existing equipment, etc.

[0056] In the embodiment of the present application, the real-time game data of the current game session and the initial set of equipment plans are obtained, the candidate plan set is screened out from the initial set of equipment plans according to the real-time game data, the player feature label is determined according to the player operation data, and at least one target candidate plan is screened out from the candidate plan set according to the player feature label and the real-time game data, the target equipment plan is determined according to at least one target candidate plan, and the current equipment is updated according to the target equipment plan. By screening from a large number of equipment plans in combination with the real-time game data, the most suitable equipment plan for the current game environment and the player's operation style can be quickly determined, which improves the accuracy of equipment plan recommendation, and also improves the efficiency of the player's selection of equipment plans for complex and changeable game scenarios.

[0057] The following is a further description of screening the target equipment plan from the initial set of equipment plans according to the real-time game data as Figure 2 shown, the above step S102 includes:

[0058] S201. Determine the screening conditions according to the real-time game data, and screen out the candidate plan set that meets the screening conditions from the initial set of equipment plans based on the improved genetic algorithm.

[0059] Among them, the screening conditions include: price conditions, weight conditions, level conditions, etc. The price condition can be the budget limit of the equipment plan, the weight condition can be the weight threshold of the equipment plan, and the level condition can be the unlocking level of the equipment plan that the player can unlock, etc.

[0060] Optionally, the filtering conditions can be used as constraint conditions to screen the initial equipment plan based on the improved genetic algorithm, and select the equipment plan within the player's budget, within the bearable weight, and within the unlockable level to obtain a set of candidate plans.

[0061] By screening the equipment plan set based on the improved genetic algorithm, a set of candidate plans available to the player can be quickly determined from a large number of popular plans, improving the speed of plan screening.

[0062] S202. Screen the target equipment plan from the set of candidate plans according to multiple screening dimensions, and the multiple screening dimensions include: equipment performance, equipment value, game environment, and player style.

[0063] The set of candidate plans includes multiple equipment plans that the player can use. By further screening the set of candidate plans, the equipment plan that best suits the player's game style and the current game environment can be selected from the plans that the player can use as the target equipment plan.

[0064] Optionally, screening the set of candidate plans according to multiple screening dimensions such as equipment performance, equipment value, game environment, and player style can ensure that the obtained target candidate plan is an equipment plan suitable for the player's style and needs, and ensure that the obtained target candidate plan is an equipment plan suitable for the current game environment and the current equipment price.

[0065] Optionally, the equipment performance dimension can indicate the attack performance, defense performance, movement performance, etc. of each equipment in the candidate plan. The equipment value dimension can indicate the price and price fluctuations of each equipment in the candidate plan. The game environment dimension can indicate the requirements, weakening coefficient, and enhancement coefficient of the current game scenario for the equipment in the candidate plan. The player style dimension can indicate the player's game preferences in historical game data. For example, the player style can be attack type, mobility type, defense type, economy type, etc.

[0066] As a possible implementation, each candidate plan can be scored from the above multiple screening dimensions, sorted according to the score value, and the candidate plan with the highest ranking is used as the target equipment plan.

[0067] As another possible implementation, it is also possible to combine an artificial intelligence model for screening. The set of candidate plans, real-time game data, and screening dimensions are used as input parameters of the artificial intelligence model. The artificial intelligence model screens from multiple screening dimensions based on the real-time game data to screen the target equipment plan from the set of candidate plans.

[0068] Optionally, at least one target candidate solution can be determined from the set of candidate solutions first, and the player can select the target equipment solution from the target candidate solutions. When recommending at least one target candidate solution to the player, the scores, advantages, and disadvantages of each target candidate solution can also be output, so that the player can quickly select the equipment solution based on the advantages and disadvantages of each target candidate solution.

[0069] As a possible implementation, for each target candidate solution, an equipment code of the target candidate solution can be generated. The equipment code includes the weapon model, weapon identification, and check code in the equipment solution. If the player selects the current target candidate solution as the target equipment solution, the equipment code of the target candidate solution can be copied, so as to update the current equipment to the target equipment solution indicated by the equipment code.

[0070] The following is a further description of the above-mentioned method for screening the target equipment solution from the set of candidate solutions according to multiple screening dimensions, as Figure 3 shown, the above step S202 includes:

[0071] S301. Determine the player characteristic label according to the player operation data.

[0072] Optionally, the player operation data can characterize the player's style and interaction habits in the game. According to the player operation data, the preference of the player for selecting the equipment solution can be determined, so as to determine the dimension that the equipment solution needs to be optimized. Among them, the optimization dimensions of the equipment solution include at least one of the following: equipment weight, equipment attack power, equipment range, equipment defense power, equipment price, equipment cost performance, equipment speed bonus, etc.

[0073] For example, if the player operation data characterizes the player's style as exploratory, the player characteristic label can indicate that the dimensions that the equipment solution needs to be optimized are equipment weight and equipment speed bonus. If the player operation data characterizes the player's style as aggressive, the player characteristic label can indicate that the dimensions that the equipment solution needs to be optimized are equipment attack power and equipment range. If the player operation data characterizes the player's style as defensive, the player characteristic label can indicate that the dimension that the equipment solution needs to be optimized is equipment defense power. If the player operation data characterizes the player's style as economical, the player characteristic label can indicate that the dimensions that the equipment solution needs to be optimized are equipment price and equipment cost performance.

[0074] It should be noted that the player can have multiple non-contradictory styles in the game. For example, the player can have both an aggressive style and an economical style, or can have both an exploratory and a defensive style at the same time. At this time, the player characteristic label includes the player characteristic labels corresponding to all the styles that the player has.

[0075] S302. Input the player characteristic label, the set of candidate solutions, and the real-time game data into the artificial intelligence model.

[0076] Among them, the artificial intelligence model may include multiple sub-models for screening the candidate solution set multiple times.

[0077] As a possible implementation, the process of determining the player feature label according to the player operation data can also be completed by a sub-model in the artificial intelligence model. At this time, the real-time game data and the candidate solution set can be input into the artificial intelligence model.

[0078] S303. The bidirectional LSTM (Long Short-Term Memory) model in the artificial intelligence model screens the candidate solution set from the equipment value dimension to obtain the first solution set.

[0079] Among them, the bidirectional LSTM model can screen from the equipment value dimension based on the equipment data in the real-time game data. For example, feature extraction is performed based on the equipment data, and the equipment value of each equipment solution is determined according to the extracted features, so as to screen each candidate solution set based on the equipment value to obtain the first solution set.

[0080] Optionally, the equipment data can be in the form of a time series, representing the change in equipment price at different time points.

[0081] Exemplarily, dependency capture can be performed on the equipment data, and then feature extraction is performed on the captured dependencies, so as to determine the equipment value of each equipment solution based on the extracted features. In this process, the bidirectional LSTM model is responsible for capturing the time dependencies in the equipment data. The bidirectional LSTM model includes a spatio-temporal attention weighting module, and the spatio-temporal attention weighting module further optimizes the feature representation, highlighting important features, thereby improving the accuracy of screening.

[0082] S304. The time decay weighting model in the artificial intelligence model screens the first solution set from the equipment performance dimension and the game environment dimension to obtain the second solution set.

[0083] Optionally, the equipment solution may exhibit different performances in different game environments. Therefore, the equipment solutions in the first solution set can be screened based on the game environment data, and the equipment solutions suitable for the current game environment are found and added to the second solution set.

[0084] Optionally, the time decay weighting model can perform feature extraction and fusion by combining the current game environment and the performance of each equipment solution, so as to obtain a feature representation that can simultaneously represent the equipment performance dimension and the game environment dimension.

[0085] Among them, the time decay weighted model includes a weight decay function for calculating the weight of each data point. The weight decay function can be, for example, an exponential decay function or a linear decay function.

[0086] S305. The multi-modal cross-attention module and the decision-making module in the artificial intelligence model screen the second set of solutions from the player style dimension based on the player feature tags to obtain the target equipment solution.

[0087] Optionally, the multi-modal cross-attention module is used to extract and fuse the features of equipment performance and player feature tags and generate a comprehensive feature representation. The decision-making module is used to determine the optimization dimension according to the player feature tags, score and screen the equipment solutions, and finally obtain the target equipment solution that better conforms to the player style.

[0088] Next, the steps of determining the player feature tags according to the player operation data are described as follows. Figure 4 As shown, the above S301 step includes:

[0089] S401. Extract the key parameter data from the player operation data.

[0090] Among them, the key parameter data can characterize the player's game style. Taking a shooting game as an example, the key parameter data includes at least one of the following: aiming speed, recoil control path, shooting frequency, shooting accuracy, movement speed, combat distance, preferred weapon type, virtual account balance, etc.

[0091] S402. Determine the player feature tags according to the key parameter data.

[0092] The key parameter data can reflect the operation style preferred by the player. The operation style of the player includes at least one of the following: aggressive, exploratory, defensive, economical, etc.

[0093] For example, when the shooting frequency is relatively high, the movement speed is relatively high, the player prefers submachine guns and the shooting accuracy is relatively low, it can be determined that the player is aggressive; when the movement speed is relatively high, the combat distance is relatively far, the equipment is light and the shooting frequency is relatively low, it can be determined that the player is exploratory; when the shooting frequency is relatively low, the player prefers sniper rifles and the shooting accuracy is relatively high, it can be determined that the player is defensive; if the purchase price of the player's weapons and equipment is relatively low, the selling price is relatively high, and the amount of the player's virtual account balance remains stable, it can be determined that the player is economical.

[0094] After determining the player's operation style, the dimensions that need to be optimized for the equipment plan can be determined based on the player's operation style. For example, for an attack-type player, the player characteristic label can be "prioritize improving the shooting frequency and shooting accuracy". For a defense-type player, the player characteristic label can be "prioritize improving the defense ability and movement speed". For an economy-type player, the player characteristic label can be "prioritize improving the cost performance of the equipment plan". For an exploration-type player, the player characteristic label can be "prioritize reducing the equipment weight and increasing the movement speed".

[0095] The following is a further description of determining the player characteristic label based on the above key parameter data. As Figure 5 shown, step S402 above includes:

[0096] S501. Score each key parameter data to obtain the parameter scores of each key parameter data.

[0097] Optionally, each key parameter data can be converted based on a preset function to convert the key parameter data into a standard score.

[0098] S502. According to each parameter score and the preset score interval, determine the target score interval of each parameter score, and use the characteristic label corresponding to the target score interval as the player characteristic label.

[0099] Optionally, multiple score intervals can be predefined for each key parameter data. Exemplarily, assuming the total score is 100 points, the score intervals for shooting accuracy can include: high accuracy [80 - 100], medium accuracy [60 - 79], and low accuracy [0 - 59].

[0100] After determining the score intervals where each parameter is located, the player characteristic label can be obtained by synthesizing all the score intervals. Exemplarily, if the player has a high score in shooting accuracy and resource management, but a low score in shooting frequency and movement speed, then the player may be a defense-type player. If the player has a high score in shooting frequency and movement speed, but a low score in shooting accuracy and resource management, then the player may be an attack-type player.

[0101] In the embodiments of the present application, by analyzing the operation data of the player, the player characteristic label can be obtained, so that when screening the equipment plan subsequently, the equipment plan that more conforms to the player's operation style can be selected according to the player characteristic label, improving the accuracy of the equipment plan recommendation.

[0102] Figure 6 is a schematic diagram of the architecture of an artificial intelligence model given in the present application. Refer to Figure 6, The bidirectional LSTM model includes a bidirectional LSTM module, a spatio-temporal attention weighting module, and an output layer connected in sequence. Among them, the input data can be player feature labels, a set of candidate solutions, and real-time game data, or a set of candidate solutions and real-time game data. When the input data is a set of candidate solutions and real-time game data, the artificial intelligence model also includes a label feature extraction module for extracting player feature labels based on the input real-time game data.

[0103] As Figure 7 shown, the process of the bidirectional LSTM model in the above artificial intelligence model screening the set of candidate solutions from the equipment value dimension to obtain the first set of solutions includes:

[0104] S701. The bidirectional LSTM module extracts the dependency relationship of the equipment data in the set of candidate solutions to obtain the context information of the equipment data, and inputs the context information into the spatio-temporal attention weighting module.

[0105] Optionally, the forward LSTM in the bidirectional LSTM module can process data from the start to the end of the sequence to capture the forward dependency relationship of the sequence, and the backward LSTM can process data from the end to the start of the sequence to capture the backward dependency relationship of the sequence. The bidirectional LSTM module finally fuses the output of the forward LSTM and the output of the backward LSTM to obtain the context information of each time step.

[0106] Among them, the context information is used to characterize the semantic information of each equipment price data in the equipment data, as well as the dependency relationship between each equipment price data and other equipment price data in the equipment data. Exemplarily, forward LSTM processing and backward LSTM processing can be performed on the equipment price data in the equipment data, and the results of the forward processing and the backward processing can be concatenated or fused to obtain the context information of the equipment price data.

[0107] S702. The spatio-temporal attention weighting module extracts features from the context information to obtain the first weighted feature, and inputs the first weighted feature into the output layer.

[0108] Optionally, the spatio-temporal attention weighting module can extract features from the context information output by the bidirectional LSTM, calculate the attention weights of each time step, and then weight the features according to the attention weights of each time step to enhance the representation of important features and obtain the first weighted feature of each equipment solution.

[0109] S703. The output layer determines the equipment value of each candidate solution in the set of candidate solutions according to the first weighted feature.

[0110] Optionally, the output layer can calculate the score of each equipment solution in the equipment price dimension according to the first weighted feature, and use the score as the equipment value of each candidate solution.

[0111] S704. The output layer screens the candidate solution set according to the equipment value of each candidate solution to obtain the first solution set.

[0112] Optionally, the output layer can sort the equipment solutions in the candidate solution set according to the equipment value of each equipment solution, and perform screening processing based on the sorting result to obtain the first solution set.

[0113] Continue to refer to Figure 6 , in the time decay weighted model, there are a physical simulation module, a feature weighting module, and an output module connected in sequence.

[0114] As Figure 8 shown, the process of the time decay weighted model in the above artificial intelligence model for secondary screening of the first solution set from the dimensions of equipment performance and game environment to obtain the second solution set includes:

[0115] S801. The physical simulation module performs physical simulation on each candidate solution in the first solution set based on the game environment data to obtain the equipment performance data of each candidate solution in the current game environment, and inputs each equipment performance data into the feature weighting module.

[0116] Optionally, the physical simulation module can simulate the performance of the equipment in the current game environment. This includes: simulating physical characteristics such as the weight, speed, and acceleration of the equipment solution, simulating the influence of environmental factors such as weather and terrain on the equipment performance, and simulating the collision effect between the equipment solution and the environment, etc.

[0117] Through physical simulation, the equipment performance data of each candidate solution in the current game environment can be obtained. The equipment performance data includes at least one of the following: aggressiveness, defensiveness, movement speed, stability, durability.

[0118] S802. The feature weighting module determines the second weighted feature of each candidate solution according to the equipment performance data of each candidate solution in the current game environment, and inputs each second weighted feature into the output module.

[0119] Optionally, the feature weighting module can extract features from the equipment performance to obtain important equipment performance features, assign higher weights to the important equipment performances in the candidate solutions, and finally fuse different performance features in the candidate solutions to obtain the second weighted feature.

[0120] S803. The output module screens according to the second weighted feature of each candidate solution to obtain the second solution set.

[0121] Optionally, the output module may calculate the scores of each equipment plan in terms of equipment performance and environment dimensions based on the second weighted features, sort them based on the scores, and finally perform a secondary screening according to the sorting results to obtain the second plan set.

[0122] As Figure 9 shown, the process of the multi-modal cross-attention module and the decision-making module in the above artificial intelligence model for screening the second plan set from the player style dimension based on the player feature tags to obtain at least one target candidate plan includes:

[0123] S901. The multi-modal cross-attention module extracts multi-modal fusion features from the player feature tags and equipment performance data, and inputs the multi-modal fusion features into the decision-making module.

[0124] Optionally, the multi-modal cross-attention module may extract features from the player feature tags and equipment performance data respectively, perform cross-attention calculation on the extracted features to obtain the cross-attention weights of each modal feature, and then fuse each modal feature according to the cross-attention weights to obtain multi-modal fusion features.

[0125] S902. The decision-making module determines the target equipment corresponding to the optimization dimension indicated by the player feature tags, and assigns weight values to each equipment, where the weight value of the target equipment is greater than that of other equipment.

[0126] Among them, the target equipment corresponding to the optimization dimension indicated by the player feature tags may be equipment that can improve the optimization dimension indicated by the player feature tags. For example, if the optimization dimension indicated by the player feature tags is "prioritize improving the shooting frequency", the corresponding target equipment may be a multi-shot submachine gun. At this time, the weight value of the multi-shot submachine gun is greater than that of other equipment.

[0127] S903. The decision-making module determines the equipment plan scores of each candidate plan in the second plan set according to the weight values of each equipment and the multi-modal fusion features.

[0128] Optionally, the decision-making module may perform weighted calculation on the multi-modal fusion features based on the weight values of each equipment to obtain the equipment plan scores of each candidate plan.

[0129] Exemplarily, the product of the feature item of each equipment for calculating the multi-modal fusion feature and the weight value of this equipment may be calculated, and the sum of all products in the equipment plan is used as the equipment plan score.

[0130] In another example, the interaction between features may also be considered, and a polynomial function is used to calculate the equipment plan score according to the feature items of each equipment and the weight values of each equipment.

[0131] S904. The decision-making module uses the candidate solutions whose equipment solution scores meet the preset conditions as the target equipment solutions.

[0132] Optionally, the decision-making model can rank each candidate solution based on the equipment solution score, and use the top M solutions as the target equipment solutions.

[0133] In the embodiments of the present application, by comprehensively considering and screening equipment solutions from four aspects: equipment value dimension, equipment performance dimension, player style dimension, and environment dimension through an artificial intelligence model, it can adapt to complex and changeable game environments and further improve the accuracy of equipment solution recommendations.

[0134] During the game operation, the price of equipment may fluctuate to a certain extent. The present application can also dynamically adjust the target equipment solution in response to the fluctuation of equipment prices. The specific process includes:

[0135] Respond to the change in equipment price data, adjust the target equipment solution, and obtain and push the adjusted equipment solution.

[0136] Optionally, if the fluctuation value of the equipment price data is greater than the preset fluctuation threshold, for example, the daily increase or decrease of the equipment price ≥ 15%, then an adjustment event for the target equipment solution can be triggered, and the adjusted equipment solution is obtained and re-pushed.

[0137] Among them, the adjustment event can be an event of replacing some equipment in the target equipment solution or an event of regenerating the target equipment solution.

[0138] Exemplarily, if the prices of multiple pieces of equipment in the target equipment solution have all fluctuated by more than the preset fluctuation threshold, then the entire target equipment solution can be replaced, that is, the above steps S101 - S104 are re-executed to obtain a new target equipment solution. If only a small part of the equipment prices in the target equipment solution have fluctuated by more than the preset fluctuation threshold, then only this part of the equipment can be replaced.

[0139] As Figure 10 shown, the process of adjusting the target equipment solution to obtain and push the adjusted equipment solution includes:

[0140] S1001. Determine the equipment to be adjusted in the target equipment solution and the replacement equipment for the equipment to be adjusted according to the change information of the equipment price data.

[0141] S1002. Replace the equipment to be adjusted in the target equipment solution with the replacement equipment to obtain the adjusted equipment solution.

[0142] S1003. Push the adjusted equipment solution.

[0143] Optionally, an equipment equivalent replacement library can be established in advance. The equipment equivalent replacement library includes multiple groups of equivalent equipment, and the functions of each group of equivalent equipment are equivalent. If the price change information of the equipment in the equipment plan is greater than the preset fluctuation threshold, the equipment can be regarded as an equipment to be adjusted, and a replacement equipment for the equipment to be adjusted can be found from the equipment equivalent replacement library.

[0144] Among them, the replacement equipment can be the equivalent equipment in the same group as the equipment to be adjusted in the equipment equivalent replacement library, and the price of the replacement equipment is lower than that of the equipment to be adjusted. The equivalent equipment can be a single equipment or a combination of multiple equipments. For example, the replacement equipment for the vertical grip can be a combination of a tactical grip + a muzzle compensator.

[0145] It should be noted that if there are multiple replacement equipments, the one with the lowest price and the best performance can be selected as the replacement equipment for the equipment to be adjusted.

[0146] Optionally, after replacing all the equipments to be adjusted, the finally obtained adjusted equipment plan can be recommended to the player in the form of an equipment code, and the player can update the equipment plan by copying the equipment code.

[0147] After the current game session ends, at least one target candidate plan is re-evaluated according to the execution result of the target equipment plan in the current game session, and the target equipment plan is determined.

[0148] Optionally, the target equipment plan can be re-evaluated after every X games are completed. If the hit rate of the player under the target equipment plan is continuously less than the preset threshold, the target candidate plans obtained in the above step S103 can be re-sorted, the priority of the currently used target equipment plan can be lowered, and the target equipment plan can be re-determined.

[0149] When the game version is updated, the method of this application can also make timely adjustments to the target equipment plan based on the version update information. The equipment equivalent replacement library in the above step S1001 can be forcibly refreshed after the version is updated Y times. At the same time, the basic parameters of the equipment can be adjusted according to the version update information, such as Figure 11 as shown, including:

[0150] S1101. If the game version is updated, obtain and parse the version update announcement to obtain the parameter adjustment information of at least one target equipment.

[0151] Optionally, the version update announcement can be parsed textually to obtain the equipment whose parameters have changed. For example, if the version description is "reduce the damage of 5.56mm bullets", the 5.56mm bullet can be regarded as a target equipment, and its parameter adjustment information can be determined as "damage value × 0.92".

[0152] S1102. Update the equipment data of each target equipment according to the parameter adjustment information.

[0153] Optionally, the equipment data of the target equipment can be replaced with the equipment data indicated by the parameter adjustment information, and a red warning sign can be additionally added for the equipment with reduced performance after the version update. Exemplarily, if the parameter adjustment information indicates that the equipment defense ability in the equipment data is weakened by 10%, the defense ability of the target equipment can be multiplied by 90% to obtain the updated equipment data.

[0154] By constructing a complete process of "data collection - intelligent decision-making - dynamic optimization", this application can, through actual measurement, improve the optimization efficiency of the player's weapon configuration by 53.6%, control the response time of the plan generation within 800 ms, and achieve the timeliness of adapting to the version update within an average of 7 minutes and 42 seconds after the release of the update announcement.

[0155] Based on the same inventive concept, an apparatus for configuring game equipment corresponding to the method for configuring game equipment is also provided in the embodiments of this application. Since the principle of solving problems by the apparatus in the embodiments of this application is similar to that of the above method for configuring game equipment in the embodiments of this application, the implementation of the apparatus can refer to the implementation of the method, and the repeated parts will not be elaborated.

[0156] Figure 12 The structural schematic diagram of an apparatus for configuring game equipment provided by the embodiments of this application is shown. The apparatus includes: an acquisition module 1201 and a screening module 1202.

[0157] The acquisition module is used to acquire the real-time game data of the current game session and the initial equipment plan set. The real-time game data includes at least one of the following: game environment data, equipment data, and player operation data.

[0158] The screening module is used to screen the target equipment plan from the initial equipment plan set according to the real-time game data and update the current equipment according to the target equipment plan.

[0159] In a feasible implementation scheme, the screening module 1202 is used for:

[0160] Determine the screening conditions according to the real-time game data, and screen the candidate plan set that meets the screening conditions from the initial equipment plan set based on the improved genetic algorithm;

[0161] Screen the target equipment plan from the candidate plan set according to multiple screening dimensions. The multiple screening dimensions include: equipment performance, equipment value, game environment, and player style.

[0162] In a feasible implementation scheme, the screening module 1202 is used for:

[0163] Determine player characteristic labels based on player operation data;

[0164] Input the player characteristic labels, the candidate solution set, and the real-time game data into the artificial intelligence model;

[0165] The bidirectional LSTM model in the artificial intelligence model screens the candidate solution set from the dimension of equipment value to obtain the first solution set;

[0166] The time decay weighted model in the artificial intelligence model screens the first solution set from the dimensions of equipment performance and game environment based on the game environment data to obtain the second solution set;

[0167] The multi-modal cross-attention module and the decision-making module in the artificial intelligence model screen the second solution set from the dimension of player style based on the player characteristic labels to obtain the target equipment solution.

[0168] In a feasible implementation, the screening module 1202 is used for:

[0169] Extract key parameter data from the player operation data;

[0170] Determine player characteristic labels according to the key parameter data.

[0171] In a feasible implementation, the screening module 1202 is used for:

[0172] Score each key parameter data to obtain the parameter scores of each key parameter data;

[0173] According to each parameter score and the preset score interval, determine the target score interval of each parameter score, and use the characteristic label corresponding to the target score interval as the player characteristic label.

[0174] In a feasible implementation, the bidirectional LSTM model includes: a bidirectional LSTM module, a spatio-temporal attention weighting module, and an output layer;

[0175] The screening module 1202 is used for:

[0176] The bidirectional LSTM module extracts the dependency relationship of the equipment data in the candidate solution set to obtain the context information of the equipment data, and inputs the context information into the spatio-temporal attention weighting module. The context information is used to represent the semantic information of each equipment price data in the equipment data, and the dependency relationship between each equipment price data and other equipment price data in the equipment data;

[0177] The spatio-temporal attention weighting module extracts features from the context information to obtain the first weighted feature, and inputs the first weighted feature into the output layer;

[0178] The output layer determines the equipment value of each candidate solution in the candidate solution set based on the first weighted feature;

[0179] The output layer screens the candidate solution set according to the equipment value of each candidate solution to obtain the first solution set.

[0180] In a feasible implementation, the time decay weighted model includes: a physical simulation module, a feature weighting module, and an output module;

[0181] The screening module 1202 is used for:

[0182] The physical simulation module performs physical simulation on each candidate solution in the first solution set based on the game environment data to obtain the equipment performance data of each candidate solution in the current game environment, and inputs each equipment performance data into the feature weighting module;

[0183] The feature weighting module determines the second weighted feature of each candidate solution according to the equipment performance data of each candidate solution in the current game environment, and inputs each second weighted feature into the output module;

[0184] The output module screens according to the second weighted feature of each candidate solution to obtain the second solution set.

[0185] In a feasible implementation, the screening module 1202 is used for:

[0186] The multi-modal cross-attention module performs feature fusion on the player feature label and the equipment performance data to obtain the multi-modal fusion feature, and inputs the multi-modal fusion feature into the decision-making module;

[0187] The decision-making module determines the target equipment corresponding to the player feature label and assigns weight values to each equipment, where the weight value of the target equipment is greater than that of other equipment;

[0188] The decision-making module determines the equipment plan score of each candidate solution in the second solution set according to the weight value of each equipment and the multi-modal fusion feature;

[0189] The decision-making module uses the candidate solutions whose equipment plan scores meet the preset conditions as the target equipment plans.

[0190] In a feasible implementation, the device further includes an adjustment module for:

[0191] In response to the change of the equipment price data, adjust the target equipment plan, and obtain and push the adjusted equipment plan.

[0192] In a feasible implementation, the adjustment module is further used for:

[0193] Determine the equipment to be adjusted and the replacement equipment for the equipment to be adjusted in the target equipment plan according to the change information of the equipment price data;

[0194] Replace the equipment to be adjusted in the target equipment plan with the replacement equipment to obtain the adjusted equipment plan;

[0195] Push the adjusted equipment plan.

[0196] In a feasible implementation, the adjustment module is further configured to:

[0197] After the current game session ends, re-evaluate at least one target candidate plan according to the execution result of the target equipment plan in the current game session, and determine the target equipment plan.

[0198] In a feasible implementation, the adjustment module is further configured to:

[0199] If the game version is updated, obtain and parse the version update announcement to obtain the parameter adjustment information of at least one target equipment;

[0200] Update the equipment data of each target equipment according to the parameter adjustment information.

[0201] By screening from a large number of equipment plans in combination with real-time game data, the embodiments of the present application can quickly determine the equipment plan that is most suitable for the current game environment and the player's operation style, improve the accuracy of the equipment plan recommendation, and also improve the efficiency of the player's selection of the equipment plan for complex and changeable game scenarios.

[0202] Figure 13 The schematic structural diagram of an electronic device provided by the embodiments of the present application is shown, including: a processor 1301, a storage medium 1302, and a bus 1303. The storage medium 1302 stores machine-readable instructions executable by the processor 1301. When the electronic device runs a configuration method of a game equipment as in the embodiment, the processor 1301 communicates with the storage medium 1302 through the bus 1303, and the processor 1301 executes the machine-readable instructions, the preamble part of the method item of the processor 1301, to perform the following steps:

[0203] Obtain the real-time game data of the current game session and the initial equipment plan set. The real-time game data includes at least one of the following: game environment data, equipment data, and player operation data;

[0204] Screen the target equipment plan from the initial equipment plan set according to the real-time game data, and update the current equipment according to the target equipment plan.

[0205] In a feasible implementation, when the processor 1301 executes to screen the target equipment plan from the initial equipment plan set according to the real-time game data, it is specifically used for:

[0206] Determine the screening conditions according to the real-time game data, and screen the candidate plan set that meets the screening conditions from the initial equipment plan set based on the improved genetic algorithm;

[0207] Screen the target equipment plan from the candidate plan set according to multiple screening dimensions, and the multiple screening dimensions include: equipment performance, equipment value, game environment, and player style.

[0208] In a feasible implementation, when the processor 1301 executes to screen the target equipment plan from the candidate plan set according to multiple screening dimensions, it is specifically used for:

[0209] Determine the player feature label according to the player operation data;

[0210] Input the player feature label, the candidate plan set, and the real-time game data into the artificial intelligence model;

[0211] The bidirectional LSTM model in the artificial intelligence model screens the candidate plan set from the equipment value dimension to obtain the first plan set;

[0212] The time decay weighted model in the artificial intelligence model screens the first plan set from the equipment performance dimension and the game environment dimension based on the game environment data to obtain the second plan set;

[0213] The multi-modal cross-attention module and the decision module in the artificial intelligence model screen the second plan set from the player style dimension based on the player feature label to obtain the target equipment plan.

[0214] In a feasible implementation, when the processor 1301 executes to determine the player feature label according to the player operation data, it is specifically used for:

[0215] Extract the key parameter data from the player operation data;

[0216] Determine the player feature label according to the key parameter data.

[0217] In a feasible implementation, when the processor 1301 executes to determine the player feature label according to the key parameter data, it is specifically used for:

[0218] Score each key parameter data to obtain the parameter score of each key parameter data;

[0219] Based on the scores of each parameter and the preset score range, determine the target score range of each parameter score, and use the feature label corresponding to the target score range as the player feature label.

[0220] In a feasible implementation, the bidirectional LSTM model includes: a bidirectional LSTM module, a spatio-temporal attention weighting module, and an output layer;

[0221] In a feasible implementation, when the processor 1301 executes the bidirectional LSTM model in the artificial intelligence model to screen the candidate solution set from the equipment value dimension and obtain the first solution set, it specifically is used for:

[0222] The bidirectional LSTM module extracts the dependency relationship of the equipment data in the candidate solution set to obtain the context information of the equipment data, and inputs the context information into the spatio-temporal attention weighting module. The context information is used to represent the semantic information of each equipment price data in the equipment data, and the dependency relationship between each equipment price data and other equipment price data in the equipment data;

[0223] The spatio-temporal attention weighting module extracts features from the context information to obtain the first weighted feature, and inputs the first weighted feature into the output layer;

[0224] The output layer determines the equipment value of each candidate solution in the candidate solution set according to the first weighted feature;

[0225] The output layer screens the candidate solution set according to the equipment value of each candidate solution to obtain the first solution set.

[0226] In a feasible implementation, the time decay weighting model includes: a physical simulation module, a feature weighting module, and an output module;

[0227] In a feasible implementation, when the processor 1301 executes the time decay weighting model in the artificial intelligence model to screen the first solution set from the equipment performance dimension and the game environment dimension based on the game environment data and obtain the second solution set, it specifically is used for:

[0228] The physical simulation module performs physical simulation on each candidate solution in the first solution set based on the game environment data to obtain the equipment performance data of each candidate solution in the current game environment, and inputs each equipment performance data into the feature weighting module;

[0229] The feature weighting module determines the second weighted feature of each candidate solution according to the equipment performance data of each candidate solution in the current game environment, and inputs each second weighted feature into the output module;

[0230] The output module screens according to the second weighted feature of each candidate solution to obtain the second solution set.

[0231] In a feasible implementation, when the processor 1301 executes the filtering of the second set of solutions from the player style dimension based on the player feature tags by the multimodal cross-attention module and the decision-making module in the artificial intelligence model to obtain the target equipment solution, it is specifically used for:

[0232] The multimodal cross-attention module performs feature fusion on the player feature tags and the equipment performance data to obtain multimodal fusion features, and inputs the multimodal fusion features into the decision-making module;

[0233] The decision-making module determines the target equipment corresponding to the player feature tags, and assigns weight values to each piece of equipment. Among them, the weight value of the target equipment is greater than that of other pieces of equipment;

[0234] The decision-making module determines the equipment solution scores of each candidate solution in the second set of solutions according to the weight values of each piece of equipment and the multimodal fusion features;

[0235] The decision-making module uses the candidate solutions whose equipment solution scores meet the preset conditions as the target equipment solutions.

[0236] In a feasible implementation, the processor 1301 is also used to execute:

[0237] In response to the change of the equipment price data, adjust the target equipment solution, and obtain and push the adjusted target equipment solution.

[0238] In a feasible implementation, when the processor 1301 executes the adjustment of the target equipment solution to obtain and push the adjusted target equipment solution, it is specifically used for:

[0239] According to the change information of the equipment price data, determine the equipment to be adjusted in the target equipment solution and the replacement equipment for the equipment to be adjusted;

[0240] Replace the equipment to be adjusted in the target equipment solution with the replacement equipment to obtain the adjusted target equipment solution;

[0241] Push the adjusted target equipment solution.

[0242] In a feasible implementation, the processor 1301 is also used to execute:

[0243] After the current game session ends, based on the execution result of the target equipment solution in the current game session, re-filter the initial set of equipment solutions to obtain the updated target equipment solution.

[0244] In a feasible implementation, the processor 1301 is also used to execute:

[0245] If the game version is updated, obtain and parse the version update announcement to obtain parameter adjustment information for at least one target piece of equipment;

[0246] According to the parameter adjustment information, update the equipment data of each target piece of equipment. By screening from a large number of equipment solutions in combination with real-time game data in the embodiments of the present application, the most suitable equipment solution for the current game environment and the player's operation style can be quickly determined, improving the accuracy of equipment solution recommendation, and also improving the efficiency of the player's selection of equipment solutions for complex and changeable game scenarios.

[0247] The embodiments of the present application also provide a computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, the processor executes the following steps:

[0248] Obtain the real-time game data of the current game session and the initial equipment solution set. The real-time game data includes at least one of the following: game environment data, equipment data, and player operation data;

[0249] Screen the target equipment solution from the initial equipment solution set according to the real-time game data, and update the current equipment according to the target equipment solution.

[0250] In a feasible implementation, when the processor executes screening the target equipment solution from the initial equipment solution set according to the real-time game data, it is specifically used for:

[0251] Determine the screening conditions according to the real-time game data, and screen the candidate solution set that meets the screening conditions from the initial equipment solution set based on the improved genetic algorithm;

[0252] Screen the target equipment solution from the candidate solution set according to multiple screening dimensions. The multiple screening dimensions include: equipment performance, equipment value, game environment, and player style.

[0253] In a feasible implementation, when the processor executes screening the target equipment solution from the candidate solution set according to multiple screening dimensions, it is specifically used for:

[0254] Determine the player characteristic label according to the player operation data;

[0255] Input the player characteristic label, the candidate solution set, and the real-time game data into the artificial intelligence model;

[0256] The bidirectional LSTM model in the artificial intelligence model screens the candidate solution set from the equipment value dimension to obtain the first solution set;

[0257] The time decay weighted model in the artificial intelligence model screens the first set of solutions from the equipment performance dimension and the game environment dimension based on the game environment data to obtain the second set of solutions;

[0258] The multi-modal cross-attention module and the decision-making module in the artificial intelligence model screen the second set of solutions from the player style dimension based on the player feature tags to obtain the target equipment solution.

[0259] In a feasible implementation, when the processor executes to determine the player feature tags according to the player operation data, it is specifically used for:

[0260] Extract key parameter data from the player operation data;

[0261] Determine the player feature tags according to the key parameter data.

[0262] In a feasible implementation, when the processor executes to determine the player feature tags according to the key parameter data, it is specifically used for:

[0263] Score each key parameter data to obtain the parameter scores of each key parameter data;

[0264] According to each parameter score and the preset score interval, determine the target score interval of each parameter score, and use the feature tag corresponding to the target score interval as the player feature tag.

[0265] In a feasible implementation, the bidirectional LSTM model includes: a bidirectional LSTM module, a spatio-temporal attention weighting module, and an output layer;

[0266] In a feasible implementation, when the processor executes to screen the candidate solution set from the equipment value dimension by the bidirectional LSTM model in the artificial intelligence model to obtain the first solution set, it is specifically used for:

[0267] The bidirectional LSTM module extracts the dependency relationship of the equipment data in the candidate solution set to obtain the context information of the equipment data, and inputs the context information into the spatio-temporal attention weighting module. The context information is used to represent the semantic information of each equipment price data in the equipment data, and the dependency relationship between each equipment price data and other equipment price data in the equipment data;

[0268] The spatio-temporal attention weighting module extracts features from the context information to obtain the first weighted feature, and inputs the first weighted feature into the output layer;

[0269] The output layer determines the equipment value of each candidate solution in the candidate solution set according to the first weighted feature;

[0270] The output layer screens the candidate solution set according to the equipment value of each candidate solution to obtain the first solution set.

[0271] In a feasible implementation, the time decay weighted model includes: a physical simulation module, a feature weighting module, and an output module;

[0272] In a feasible implementation, when the processor executes the time decay weighted model in the artificial intelligence model to screen the first solution set from the equipment performance dimension and the game environment dimension based on the game environment data to obtain the second solution set, it is specifically used for:

[0273] The physical simulation module performs physical simulation on each candidate solution in the first solution set based on the game environment data to obtain the equipment performance data of each candidate solution in the current game environment, and inputs each equipment performance data into the feature weighting module;

[0274] The feature weighting module determines the second weighted feature of each candidate solution according to the equipment performance data of each candidate solution in the current game environment, and inputs each second weighted feature into the output module;

[0275] The output module screens according to the second weighted feature of each candidate solution to obtain the second solution set.

[0276] In a feasible implementation, when the processor executes the multi-modal cross-attention module and the decision module in the artificial intelligence model to screen the second solution set from the player style dimension based on the player feature label to obtain the target equipment solution, it is specifically used for:

[0277] The multi-modal cross-attention module performs feature fusion on the player feature label and the equipment performance data to obtain the multi-modal fusion feature, and inputs the multi-modal fusion feature into the decision module;

[0278] The decision module determines the target equipment corresponding to the player feature label, and assigns weight values to each equipment, where the weight value of the target equipment is greater than that of other equipment;

[0279] The decision module determines the equipment solution score of each candidate solution in the second solution set according to the weight value of each equipment and the multi-modal fusion feature;

[0280] The decision module takes the candidate solutions whose equipment solution scores meet the preset conditions as the target equipment solutions.

[0281] In a feasible implementation, the processor is also used to execute:

[0282] In response to the change of the equipment price data, adjust the target equipment solution, and obtain and push the adjusted target equipment solution.

[0283] In a feasible implementation, when the processor executes the adjustment of the target equipment plan and obtains and pushes the adjusted target equipment plan, it is specifically used for:

[0284] Determine the equipment to be adjusted in the target equipment plan and the replacement equipment for the equipment to be adjusted according to the change information of the equipment price data;

[0285] Replace the equipment to be adjusted in the target equipment plan with the replacement equipment to obtain the adjusted target equipment plan;

[0286] Push the adjusted target equipment plan.

[0287] In a feasible implementation, the processor is also used to execute:

[0288] After the current game session ends, based on the execution result of the target equipment plan in the current game session, re-screen the initial equipment plan set based on the execution result to obtain the updated target equipment plan.

[0289] In a feasible implementation, the processor is also used to execute:

[0290] If the game version is updated, obtain and parse the version update announcement to obtain the parameter adjustment information of at least one target equipment;

[0291] Update the equipment data of each target equipment according to the parameter adjustment information.

[0292] By screening from a large number of equipment plans in combination with real-time game data in the embodiments of the present application, the most suitable equipment plan for the current game environment and the player's operation style can be quickly determined, improving the accuracy of equipment plan recommendation, and also improving the efficiency of players choosing equipment plans for complex and changeable game scenarios.

[0293] In the embodiments of the present application, when the computer program is run by the processor, it can also execute other machine-readable instructions to execute the methods described in other parts of the embodiments. For the specific method steps and principles of execution, refer to the description of the embodiments, and will not be elaborated in detail here.

[0294] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For another 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 coupling or direct coupling or communication connection between each other can be through some communication interfaces. The indirect coupling or communication connection of the devices or units can be in electrical, mechanical or other forms.

[0295] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place or 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.

[0296] In addition, each functional unit in the embodiments provided in the present application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0297] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium 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 described in various embodiments of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0298] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. In addition, the terms "first", "second", "third", etc. are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0299] Finally, it should be noted that the above-described embodiments are only specific implementation manners of the present application, used to illustrate the technical solutions of the present application, rather than limiting it. The protection scope of the present application is not limited thereto. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that any person skilled in the art within the technical scope disclosed by the present application can still modify the technical solutions recorded in the foregoing embodiments, or can easily think of changes, or perform equivalent replacements on some of the technical features; and these modifications, changes or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application. All should be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described.

Claims

1. A configuration method for a game equipment, characterized in that Including: Obtain the real-time game data of the current game session and the set of initial equipment plans. The real-time game data includes at least one of the following: game environment data, equipment data, and player operation data; Filter the target equipment plan from the set of initial equipment plans according to the real-time game data, and update the current equipment according to the target equipment plan.

2. The method according to claim 1, wherein The filtering of the target equipment plan from the set of initial equipment plans according to the real-time game data includes: Determine the screening conditions according to the real-time game data, and filter the candidate plan set that meets the screening conditions from the set of initial equipment plans based on the improved genetic algorithm; Filter the target equipment plan from the candidate plan set according to multiple screening dimensions. The multiple screening dimensions include: equipment performance, equipment value, game environment, and player style.

3. The method according to claim 2, characterized in that The filtering of the target equipment plan from the candidate plan set according to multiple screening dimensions includes: Determine the player characteristic label according to the player operation data; Input the player characteristic label, the candidate plan set, and the real-time game data into the artificial intelligence model; The bidirectional LSTM model in the artificial intelligence model filters the candidate plan set from the equipment value dimension to obtain the first plan set; The time decay weighted model in the artificial intelligence model filters the first plan set from the equipment performance dimension and the game environment dimension based on the game environment data to obtain the second plan set; The multi-modal cross-attention module and the decision module in the artificial intelligence model filter the second plan set from the player style dimension based on the player characteristic label to obtain the target equipment plan.

4. The method according to claim 3, characterized in that, The determination of the player characteristic label according to the player operation data includes: Extract the key parameter data from the player operation data; Determine the player characteristic label according to the key parameter data.

5. The method according to claim 4, wherein The determination of the player characteristic label according to the key parameter data includes: Score each of the key parameter data to obtain the parameter scores of each of the key parameter data; According to each of the parameter scores and the preset score interval, determine the target score interval of each of the parameter scores, and use the characteristic label corresponding to the target score interval as the player characteristic label.

6. The method according to claim 3, characterized in that, The bidirectional LSTM model includes: a bidirectional LSTM module, a spatio-temporal attention weighting module, and an output layer; The process of filtering the candidate plan set from the equipment value dimension by the bidirectional LSTM model in the artificial intelligence model to obtain the first plan set includes: The bidirectional LSTM module extracts the dependency relationship of the equipment data in the candidate plan set to obtain the context information of the equipment data, and inputs the context information into the spatio-temporal attention weighting module. The context information is used to represent the semantic information of each equipment price data in the equipment data, and the dependency relationship between each equipment price data and other equipment price data in the equipment data; The spatio-temporal attention weighting module extracts features from the context information to obtain a first weighted feature, and inputs the first weighted feature into the output layer; The output layer determines the equipment value of each candidate solution in the candidate solution set according to the first weighted feature; The output layer performs a screening process on the candidate solution set according to the equipment value of each candidate solution to obtain a first solution set.

7. The method according to claim 3, wherein The time decay weighting model includes: a physical simulation module, a feature weighting module, and an output module; The process of the time decay weighting model in the artificial intelligence model screening the first solution set from the equipment performance dimension and the game environment dimension based on the game environment data to obtain a second solution set includes: The physical simulation module performs physical simulation on each candidate solution in the first solution set based on the game environment data to obtain the equipment performance data of each candidate solution in the current game environment, and inputs each piece of equipment performance data into the feature weighting module; The feature weighting module determines the second weighted feature of each candidate solution according to the equipment performance data of each candidate solution in the current game environment, and inputs each second weighted feature into the output module; The output module performs screening according to the second weighted feature of each candidate solution to obtain a second solution set.

8. The method according to claim 3, characterized in that The process of the multi-modal cross-attention module and the decision-making module in the artificial intelligence model screening the second solution set from the player style dimension based on the player feature label to obtain the target equipment solution includes: The multi-modal cross-attention module performs feature fusion on the player feature label and the equipment performance data to obtain a multi-modal fusion feature, and inputs the multi-modal fusion feature into the decision-making module; The decision-making module determines the target equipment corresponding to the player feature label, and assigns weight values to each piece of equipment, where the weight value of the target equipment is greater than that of other equipment; The decision-making module determines the equipment solution score of each candidate solution in the second solution set according to the weight value of each piece of equipment and the multi-modal fusion feature; The decision-making module uses the candidate solutions whose equipment solution scores meet the preset conditions as the target equipment solutions.

9. The method according to claim 1, characterized in that, The method further includes: In response to the change of equipment price data, adjust the target equipment solution, and obtain and push the adjusted target equipment solution.

10. The method according to claim 9, wherein The adjusting the target equipment solution to obtain and push the adjusted target equipment solution includes: According to the change information of the equipment price data, determine the equipment to be adjusted in the target equipment solution and the replacement equipment for the equipment to be adjusted; Replace the equipment to be adjusted in the target equipment solution with the replacement equipment to obtain the adjusted target equipment solution; Push the adjusted target equipment solution.

11. The method according to claim 1, wherein The method further includes: After the current game session ends, based on the execution result of the target equipment plan in the current game session, the initial equipment plan set is re-screened based on the execution result to obtain an updated target equipment plan.

12. The method according to claim 1, wherein The method further includes: If the game version is updated, obtain and parse the version update announcement to obtain parameter adjustment information for at least one target equipment; Update the equipment data of each target equipment according to the parameter adjustment information.

13. A configuration device for a game equipment, characterized in that, It includes: An acquisition module, configured to acquire real-time game data of the current game session and an initial equipment plan set, where the real-time game data includes at least one of the following: game environment data, equipment data, and player operation data; A screening module, configured to screen a target equipment plan from the initial equipment plan set according to the real-time game data, and update the current equipment according to the target equipment plan.

14. An electronic device, characterized in that, It includes: A processor, a storage medium, and a bus. The storage medium stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to perform the steps of a method for configuring a game equipment as described in any one of claims 1 to 12.

15. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is run by the processor, it performs the steps of a method for configuring a game equipment as described in any one of claims 1 to 12.

Images