Information processing methods in games, computer-readable storage media and electronic devices

Abstract

This application discloses an information processing method, a computer-readable storage medium, and an electronic device for use in games. Applied to a server, the server communicates with a client. The method includes: responding to a trigger request received before the start of a game and determining game data; performing predictive processing on the game data to obtain a predicted game result; generating game prompt information based on the predicted result; and sending the game prompt information to the client. Thus, based on the equipment used by the participating players in the current game, the method predicts the impact of the current equipment configuration on the outcome and generates prompt information to feed back to the client, effectively ensuring the accuracy and relevance of the prediction results, thereby enhancing the reference value of the game prompt information.

Description

Technical Field

[0001] This application relates to the field of game technology, specifically to an information processing method, computer-readable storage medium, and electronic device in games. Background Technology

[0002] In online multiplayer games, players typically need to configure their virtual characters' equipment before a match begins. Most equipment recommendations are based on preset static rules or historical win rate statistics, such as preset official recommendations or sorting by usage frequency. However, this lack of consideration for the real-time state of the current match means that players struggle to understand the specific impact of their current equipment configuration on the match outcome, and they cannot receive targeted optimization tips. This leads to significant uncertainty in equipment selection, reducing decision-making efficiency and the overall gaming experience. Summary of the Invention

[0003] This application provides an information processing method, computer-readable storage medium, and electronic device for games. By responding to a trigger request before the start of a game, the method predicts the impact of the current equipment configuration on the outcome of the game based on the equipment used by each participating player and generates prompt information to be fed back to the client. This effectively ensures the accuracy and relevance of the prediction results, thereby improving the reference value of the game prompt information.

[0004] On one hand, this application provides an information processing method in a game, applied to a server, wherein the server is connected to a client. The method includes: responding to a trigger request received before the start of a game match and determining game match data, wherein the trigger request is generated by the client in response to a first equipment configuration operation and sent to the server, the first equipment configuration operation is used to configure the first game equipment of a virtual character, and the game match data includes at least the equipment attributes of the first game equipment of each virtual character participating in the game match; Based on the game data, the game results are predicted to obtain the predicted game results. Based on the predicted game results, game prompt information is generated, wherein the game prompt information is used to indicate the impact of the first equipment configuration operation on the game results, and / or to recommend the first game equipment used by the virtual character; The game notification information is sent to the client.

[0005] On the other hand, this application provides an information processing device for a game, applied to a server, wherein the server is communicatively connected to a client. The device includes: a first request-response module, used to respond to a trigger request received before the start of a game match and determine game match data, wherein the trigger request is generated by the client in response to a first equipment configuration operation and sent to the server, the first equipment configuration operation is used to configure the first game equipment of the virtual character, and the game match data includes at least the equipment attributes of the first game equipment of each virtual character participating in the game match; The prediction module is used to perform prediction processing on the game match data to obtain the predicted match result; The generation module is used to generate game prompt information based on the predicted game results, wherein the game prompt information is used to indicate the impact of the first equipment configuration operation on the game results, and / or to recommend the first game equipment used by the virtual character; The sending module is used to send the game prompt information to the client.

[0006] On the other hand, this application provides an information processing method for games, applied to a client. The client communicates with a server and provides a graphical user interface. The graphical user interface can display the game screen after the start of a game in the target game. The game screen includes a virtual character controlled by the client. The method includes: responding to a first equipment configuration operation before the start of a game, generating a trigger request and sending the trigger request to the server, so that the server responds to the received trigger request, determining game data, and performing prediction processing on the result of the game based on the game data to obtain a predicted game result, generating game prompt information based on the predicted game result, and sending the game prompt information to the client. The first equipment configuration operation is used to configure the first game equipment of the virtual character. The game data includes at least the equipment attributes of the first game equipment of each virtual character participating in the game. The game prompt information is used to indicate the impact of the first equipment configuration operation on the game result and / or to recommend the first game equipment used by the virtual character. The system receives the game prompt information sent by the server and updates the current display content of the graphical user interface based on the game prompt information.

[0007] On the other hand, this application embodiment provides an information processing device for a game, applied to a client. The client is communicatively connected to a server. The client provides a graphical user interface (GUI) that can display the game screen after the start of a game in the target game. The game screen includes a virtual character controlled by the client. The device includes: an operation response module, configured to respond to a first equipment configuration operation before the start of a game, generate a trigger request and send the trigger request to the server, so that the server responds to the received trigger request, determines game data, and performs prediction processing on the game result based on the game data to obtain a predicted game result, and generates game prompt information based on the predicted game result, and sends the game prompt information to the client. The first equipment configuration operation is used to configure the first game equipment of the virtual character. The game data includes at least the equipment attributes of the first game equipment of each virtual character participating in the game. The game prompt information is used to indicate the impact of the first equipment configuration operation on the game result and / or to recommend the first game equipment used by the virtual character. The information response module is used to respond to the game prompt information sent by the server and update the current display content control of the graphical user interface according to the game prompt information.

[0008] On the other hand, embodiments of this application provide a computer-readable storage medium storing a computer program adapted for loading by a processor to execute the information processing method in a game as described in any of the above embodiments.

[0009] On the other hand, embodiments of this application provide an electronic device, which includes a processor and a memory. The memory stores a computer program, and the processor executes the information processing method in the game as described in any of the above embodiments by calling the computer program stored in the memory.

[0010] On the other hand, embodiments of this application provide a computer program product, including computer instructions, which, when executed by a processor, implement the information processing method in a game as described in any of the above embodiments.

[0011] The information processing method, device, computer-readable storage medium, electronic device, and computer program product provided in this application embodiment, in this application embodiment, before the start of a game match, the server responds to the trigger request sent by the client, determines the game match data, and performs predictive processing on the game match result based on the game match data to obtain a predicted match result. Then, based on the predicted match result, it generates match prompt information that can indicate the impact of equipment configuration operations on the match result or recommend the virtual character to use the first game equipment. Finally, the match prompt information is sent to the client. This realizes the prediction of match result and equipment configuration-related prompts based on the player's actual equipment configuration operations before the game match, compared to relying on static rules or offline statistical data for player prediction. The system provides equipment recommendations but lacks a solution to provide feedback on match results based on players' real-time equipment configurations. This allows the game's equipment configuration-related prompts to align with players' pre-match actions. By relying on real data, including the attributes of each virtual character's equipment, to predict match results, the system effectively ensures the accuracy and relevance of the predictions. This provides reliable data support for the match prompts, enhancing their reference value. Furthermore, by transforming match result predictions into direct prompts on the impact of equipment configuration on victory or defeat, or equipment recommendations, players can intuitively understand the practical effects of equipment configuration without needing to analyze complex prediction data, thus reducing the cognitive load for players.

[0012] Furthermore, the server processes data, predicts results, and generates prompts in real time for players' equipment configuration operations, and sends them to the client. This achieves real-time linkage between players' equipment configuration operations and prompt feedback to a certain extent, allowing players to continuously receive targeted decision-making references during the pre-match equipment configuration process. This reduces the decision-making cost for players in pre-match equipment configuration to a certain extent, optimizes the player's operational experience in the game's equipment configuration process, and makes the game's equipment recommendation function meet the needs of actual matches, thereby enhancing the practical value of the equipment recommendation function. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a schematic diagram of an example game system provided in an embodiment of this application.

[0015] Figure 2This is a flowchart illustrating the information processing method in a game provided in an embodiment of this application.

[0016] Figure 3 This is a schematic diagram illustrating an application scenario of the information processing method in a game provided in an embodiment of this application.

[0017] Figure 4 This is a schematic diagram illustrating an application scenario of the information processing method in a game provided in an embodiment of this application.

[0018] Figure 5 This is a schematic diagram illustrating an application scenario of the information processing method in a game provided in an embodiment of this application.

[0019] Figure 6 This is a flowchart illustrating the information processing method in a game provided in an embodiment of this application.

[0020] Figure 7 This is a flowchart illustrating the information processing method in a game provided in an embodiment of this application.

[0021] Figure 8 This is a flowchart illustrating the information processing method in a game provided in an embodiment of this application.

[0022] Figure 9 This is a flowchart illustrating the information processing method in a game provided in an embodiment of this application.

[0023] Figure 10 This is a flowchart illustrating the information processing method in a game provided in an embodiment of this application.

[0024] Figure 11 This is a schematic diagram illustrating an application scenario of the information processing method in a game provided in an embodiment of this application.

[0025] Figure 12 This is a schematic diagram of the structure of the information processing device in the game provided in the embodiment of this application.

[0026] Figure 13 This is a schematic diagram of the structure of the information processing device in the game provided in the embodiment of this application.

[0027] Figure 14 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0028] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0029] This application provides an information processing method, an information processing device, a computer-readable storage medium, an electronic device, and a computer program product for games. Specifically, the information processing method for games in this application can be executed by an electronic device, which can be a terminal or a server. The terminal can be a smartphone, tablet, laptop, smart TV, wearable smart device, smart vehicle terminal, etc. The terminal can also include a client, which can be a game client, browser client, instant messaging client, or mini-program, etc. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0030] For example, when the information processing method in the game runs on a terminal device, the terminal device may include a display screen and a processor. The display screen is used to present the game screen and receive commands generated by the player interacting with it. The game screen may include a portion of a virtual game scene, which is a virtual world where virtual characters move. The processor is used to store the game application, run the game, generate game screens, respond to commands, and control the display of the game screen on the display screen. When the player interacts with the game screen through the display screen, the game screen can control the local content of the terminal device in response to the received operation commands. The terminal device can provide the graphical user interface to the player in various ways, such as rendering it on the terminal device's display screen or presenting the graphical user interface through holographic projection.

[0031] For example, when the information processing method in the game runs on a server, this method can be implemented and executed based on a cloud gaming system. A cloud gaming system refers to a gaming method based on cloud computing. A cloud gaming system includes servers and client devices. The main body running the game application and the main body presenting the game screen are separate. The storage and execution of the information processing method in the game are completed on the server. The presentation of the game screen is completed on the client, which is mainly used for receiving and sending game data and presenting the game screen. For example, the client can be a display device with data transmission capabilities close to the player, such as a mobile terminal, television, computer, PDA, personal digital assistant, head-mounted display device, etc. However, the terminal device for processing game data is the server in the cloud. When playing the game, the player operates the client to send commands to the server. The server controls the game to run according to the commands, encodes and compresses the game screen and other data, returns it to the client through the network, and finally, the client decodes and outputs the game screen.

[0032] It should be noted that in this embodiment, the executing entity of the information processing method in the game can be a terminal device or a server. The terminal device can be a local terminal device or a client device in the aforementioned cloud gaming. This embodiment does not limit the type of executing entity. It is understood that in the specific implementation of this application, user object data, context data, and other related data are involved. When this embodiment is applied to specific products or technologies, user permission or consent is required, and the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions.

[0033] For example, in conjunction with the above description, Figure 1 This application illustrates a game system 1000 for implementing an information processing method in a game, as provided in an embodiment of this application. The game system 1000 may include at least one terminal 1001, at least one server 1002, at least one database 1003, and a network. The user-held terminal 1001 can connect to different servers via the network. The terminal is any device with computing hardware capable of supporting and executing software applications corresponding to the game.

[0034] In the aforementioned game system 1000, terminal 1001 is used to install and run the game application. In some cases, the game application may not need to be pre-installed on terminal 1001, and players can directly access the game through a browser or other client. Players log in to the game application using their registered game account to control the virtual character corresponding to that account and participate in the game. When a player logs in to the game application, terminal 1001 sends a login request to server 1002. Server 1002 verifies the game account used by the player and determines the game mechanics corresponding to the game account based on the login request. If the verification is successful, a login success notification is returned to terminal 1001. During the player's participation in the game through the game application, terminal 1001 and server 1002 exchange data. Terminal 1001 sends various information to server 1002. Server 1002 determines the display data for terminal 1001 based on the stored game mechanics and the received information, and sends the display data back to terminal 1001 so that terminal 1001 can display the display data sent by server 1002 to the player.

[0035] In possible application scenarios, different terminals 1001 may be served by different servers 1002. Therefore, to distinguish the servers 1002 corresponding to different game terminals 1001, the embodiments of this application will use a first and a second approach for description. In fact, the servers 1002 corresponding to different game terminals 1001 can be the same server 1002. Therefore, without distinguishing between the first and second approaches, it can be understood that the terminals 1001 corresponding to virtual characters in the same game scene are served by the same server 1002. In addition, when the game system 1000 includes multiple terminals, multiple servers, and multiple networks, different terminals can connect to each other through different networks and different servers. The network can be a wireless network or a wired network, such as a wireless local area network (WLAN), local area network (LAN), cellular network, 2G network, 3G network, 4G network, 5G network, etc. In addition, different terminals can also use their own Bluetooth network or hotspot network to connect to other terminals or to servers, etc. In addition, the system 100 can include multiple databases, which are coupled to different servers, and can continuously store game-related information in the databases when different users are playing multi-user games online.

[0036] It should be noted that in this embodiment, multiple terminal devices are running the same virtual game. Therefore, data interaction between the multiple terminal devices can be achieved through the virtual game's server. Thus, sending data from terminal device 1 to terminal device 2 can be understood as: terminal device 1 sends data to the virtual game's server, and the server sends the data to terminal device 2. Receiving data from terminal device 2 can be understood as: terminal device 1 receives data sent by the virtual game's server, which is the data sent by terminal device 2 to the server. Alternatively, there may be no game server, and terminal device 1 directly sends game data to terminal device 2. It should be noted that... Figure 1 The game system diagram shown is merely an example. The game system 1000 described in this application embodiment is intended to more clearly illustrate the technical solutions of this application embodiment and does not constitute a limitation on the technical solutions provided in this application embodiment. As those skilled in the art will know, with the evolution of game systems and the emergence of new business scenarios, the technical solutions provided in this application embodiment are also applicable to similar technical problems.

[0037] It should be noted that the triggering operations mentioned in the subsequent detailed description of the information processing method in the game provided in the embodiments of this application can all be regarded as triggering operations performed by the player through a finger or by controlling a medium such as a mouse, keyboard, or stylus. The specific medium used can be determined according to the type of electronic device. For example, when the electronic device is a touch screen device such as a mobile phone, tablet computer, or game console, the player can operate on the touch screen using any suitable object or accessory such as a finger or stylus. When the terminal device is a non-touch screen terminal device such as a desktop computer or laptop computer, the player can operate using an external device such as a mouse or keyboard.

[0038] The technical solution of this application will be described in detail below through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0039] In this embodiment, a graphical user interface (GUI) is provided through a terminal device. The GUI includes at least a portion of a virtual scene and at least one virtual character. The virtual scene can be a game scene, which can be understood as a simulation of the real world within a game environment, a semi-simulated / semi-fictional virtual environment, or a purely fictional virtual environment. The game scene can be any of a two-dimensional, 2.5-dimensional, or three-dimensional virtual scene. A virtual scene typically includes multiple scene elements, which are the various elements required to constitute the virtual scene. For example, these may include, but are not limited to, at least one of the following: virtual character elements, virtual item elements, virtual building elements, virtual terrain elements, virtual vegetation elements, etc. Virtual terrain elements may include, but are not limited to, natural landforms such as land, ocean, lakes, and rivers. The virtual scene is a scene where the player controls a virtual character to complete game logic. It is understood that a virtual character is the game character controlled by the player in the game; that is, the player operates the virtual character to perform various game activities in the game scene, such as picking up items, engaging in combat, exploring, or solving puzzles. The virtual character can represent the player's image, and each virtual character can be implemented using a three-dimensional or two-dimensional virtual model; this embodiment does not specifically limit this. Virtual characters include, but are not limited to, at least one of virtual human, virtual animal, and virtual machine.

[0040] With the development of online gaming technology, competitive games have become one of the mainstream game genres. Competitive games include several sub-genres such as Multiplayer Online Battle Arena (MOBA), First-Person Shooter (FPS), and racing games. Before a match begins, players typically need to complete equipment configuration operations, including character selection, vehicle selection, passive buff configuration, and skill combinations. Equipment configuration not only directly affects the basic ability parameters of characters or vehicles but also influences the competitive situation and the final outcome of the match. Therefore, scientific and reasonable equipment combinations have become an important part of competitive games.

[0041] Some MOBA games require players to pre-select runes and skills, while some mobile MOBA games require players to pre-configure rune and equipment schemes. For example, in racing games, players need to comprehensively consider multiple attributes such as the car's positioning, proficiency, level, enhancement level, and chip combinations. There are compatibility relationships between vehicles and chips, and in multiplayer team gameplay, different team compositions create complex game dynamics and counter-relationships, escalating equipment selection from a single character's independent configuration to a multi-player, multi-vehicle, multi-chip, and multi-map combination optimization challenge. As game operation cycles lengthen, the player and equipment pools continue to expand, with equipment types, attribute dimensions, and combination possibilities growing exponentially. Ordinary players struggle to understand the attribute differences and compatibility logic of different equipment within a limited time, making it impossible to make optimal combination choices based on real-time information such as other players' equipment configurations and map characteristics. Even for high-level players, having multiple high-powered vehicles in the same location presents the decision-making pressure of refined vehicle selection and team composition strategy. Against this backdrop, data-driven equipment recommendations and win / loss prediction functions are gradually becoming important directions for game optimization. However, the equipment recommendations and win / loss predictions of related technical solutions are mostly limited to single-dimensional statistical recommendations, lacking real-time performance, sustainable learning capabilities, and user-friendly interactive experiences, making it difficult to meet the needs of complex game scenarios.

[0042] Some competitive games use static rules or experience tables to recommend equipment. Multiple officially recommended builds or runes are preset for each hero or character, displayed directly after the player selects a character. Based on the player's current rank, frequently used characters, and other information, a pre-configured standard build is matched. For example, racing games use tags like "suitable map" and "suitable mode" in the vehicle or equipment interface to simply label specific equipment for player reference. This approach is simple in logic, requires no complex algorithms, and doesn't rely on real-time match information or large-scale data analysis; the recommendations are based on the designers' experience or limited statistical data. However, this approach results in fixed and inflexible recommendations. Relying on manual presets, it cannot adapt to dynamic changes such as game version updates and equipment attribute adjustments. Furthermore, it fails to consider real-time factors such as individual player differences, current room composition, and map characteristics, leading to insufficient personalization and targeting of the recommendations. Moreover, the recommendations are based solely on empirical data, unverified in actual gameplay, making their rationality and reliability questionable, and failing to provide players with effective references relevant to the current match scenario.

[0043] Some game products have introduced offline recommendation mechanisms based on historical data statistics. For example, for hero and equipment combinations, they calculate the overall win rate of hero and equipment combinations across the entire server and recommend them to players in order of win rate. For specific maps or modes, they analyze the overall performance of different character and equipment combinations and provide a list of recommended heroes or vehicles before players are matched. This type of solution uses offline log data analysis to extract simple statistical indicators such as usage rate and win rate of different character and equipment combinations, and then completes the filtering and sorting of recommended content. However, the recommendation data comes from offline statistics and lacks the ability to dynamically re-evaluate based on the current room lineup and the player's real-time status. It cannot reflect the particularity of real-time game scenarios, and the recommendation dimensions are single. The statistical indicators are limited to macro data such as overall win rate and usage rate, and cannot capture the adaptation differences of equipment combinations in different game scenarios, thus limiting the practicality of the recommendation results. In addition, some competition or battle platforms attempt to use machine learning to achieve win and loss prediction, using simple machine learning models such as logistic regression and decision trees. The input features include the player's historical win rate, rank, recent performance, the overall strength value of the selected character or lineup, and some map and mode information. This type of solution uses basic feature modeling to predict the outcome of game matches. However, the machine learning model structure is relatively simple, focusing on single-task predictions such as whether a team will win. It cannot perform refined modeling of the probability of an individual player achieving a certain ranking in the current match, and the input feature dimensions are limited, failing to fully integrate information such as equipment configuration details and player interaction relationships, thus limiting prediction accuracy. Furthermore, the machine learning model's response efficiency is insufficient, making it difficult to support low-latency multiple calls in scenarios where equipment is frequently adjusted before a match. When players switch equipment multiple times, server response timeouts and insufficient throughput issues can easily occur, impacting the player's gaming experience.

[0044] In addition, some games pre-set several fixed template combinations for different characters or vehicles. For example, they pre-set multiple rune pages and inscription pages and support one-click application, or they configure fixed chip schemes such as acceleration, balance, and control for a certain vehicle. These schemes reduce the difficulty of configuration for players by standardizing templates and reduce the manual operation cost for players. However, they lack sustainable learning capabilities, do not track and analyze the long-term equipment configuration behavior of real players, cannot absorb the latest configuration experience and understanding of high-level players in a timely manner, and their understanding of equipment attributes is limited to fixed dimensions designed by humans. They cannot abstract the multi-dimensional attributes of equipment such as chips into a unified attribute vector and learn preference weights, and can only make recommendations based on a single attribute or simplified tags. Furthermore, they cannot dynamically generate suitable schemes based on the player's own equipment pool resources. For complex scenarios involving multi-dimensional parameters such as attributes, rarity, level, and shape of chips, the recommendation effect is limited.

[0045] In summary, while the relevant technical solutions have made initial attempts in equipment recommendation and win / loss prediction, they generally suffer from problems such as insufficient real-time performance, low personalization, inability to depict the impact of multi-player interactions, lack of sustainable learning capabilities, fragmented system architecture, and poor user experience. As a result, they are unable to meet the needs of complex game scenarios such as multiple vehicles, multiple chips, multiple maps, and multi-player teams, and cannot provide players with high-precision, low-latency, and interactive equipment configuration assistance.

[0046] Please refer to the following for the above questions. Figure 2 , Figure 2 This is a flowchart illustrating a game information processing method provided in an embodiment of this application. It should be noted that the steps shown may be executed in a different logical order than those shown in the flowchart. The method is applied to a server, which communicates with the client. The method includes: Step 210: Responding to a trigger request received before the start of a game match, determining game match data, wherein the trigger request is generated by the client in response to a first equipment configuration operation and sent to the server, the first equipment configuration operation is used to configure the first game equipment of a virtual character, and the game match data includes at least the equipment attributes of the first game equipment of each virtual character participating in the game match; Step 220: Based on the game data, perform prediction processing on the game results to obtain the game result prediction; Step 230: Based on the prediction of the game results, generate game prompt information, wherein the game prompt information is used to indicate the impact of the first equipment configuration operation on the game results, and / or to recommend the first game equipment used by the virtual character; Step 240: Send a game notification message to the client.

[0047] Specifically, in competitive games, players often struggle to understand the actual impact of their pre-match equipment configuration on the outcome of the game. Furthermore, recommended equipment schemes often lack real-time feedback when players frequently adjust their equipment, leading to high decision-making costs and a poor gaming experience. To address these issues, this application provides an information processing method for games, applied to a server. The server and client establish a communication connection. Upon detecting a player's first equipment configuration action, the client generates a trigger request and sends it to the server. The server responds to the trigger request and determines game data including the equipment attributes of each virtual character. Based on this data, the server predicts the game outcome and generates game-related information that indicates the impact of equipment configuration on the outcome or recommends specific equipment. Finally, this information is sent to the client, providing players with real-time equipment configuration guidance.

[0048] In some embodiments, the server can be understood as a cluster of servers or a single server that provides services such as background data processing, logical operations, and model reasoning to the game, capable of receiving client requests, processing them accordingly, and then providing feedback. In some embodiments, the client can be understood as a terminal device used by the player to perform game operations, capable of displaying the game screen and receiving player operation commands, such as a mobile phone, computer, or game console.

[0049] In some embodiments, the target game can be understood as various games that apply the information processing method of the embodiments of this application, particularly competitive games with equipment configuration and match-based combat. Examples include multiplayer online competitive games, first-person shooter games, and racing games. In some embodiments, a game match can be understood as a complete game-based combat process in which players participate, including pre-match preparation, in-game combat, and match settlement stages.

[0050] In some embodiments, a trigger request can be understood as a digital information instruction sent by the client to the server to trigger background data processing.

[0051] In some embodiments, the first equipment configuration operation can be understood as the player's operation of selecting, setting, changing, and adjusting the equipment of a virtual character on the client's graphical user interface. In some embodiments, a virtual character can be understood as a game entity such as a virtual image, vehicle, or character controlled by the player on the client.

[0052] In some embodiments, the first game equipment can be understood as an in-game adjustable element configured for a virtual character, including vehicles, accessories, runes, inscriptions, and other items that can affect the performance of the virtual character.

[0053] In some embodiments, game match data can be understood as a set of various data representing game match-related characteristics. In some embodiments, equipment attributes can be understood as various performance characteristic indicators possessed by the first game equipment. For example, performance parameters, buff effects, etc.

[0054] In some embodiments, match result prediction can be understood as the server's predictive analysis of the final direction and outcome of a game match through data processing, model reasoning, and other methods. In some embodiments, match prompt information can be understood as various data contents generated by the server to convey information about the impact and recommendations related to equipment configuration to players.

[0055] To more clearly illustrate the information processing method in the game provided in the embodiments of this application, please refer to the following exemplary description: Before the start of a game, the player performs a first equipment configuration operation on the graphical user interface of the client. The first equipment configuration operation is used to configure suitable first game equipment for the virtual character controlled by the player. After the client detects the player's first equipment configuration operation in real time, it generates a corresponding trigger request and sends the trigger request to the server through the network.

[0056] After receiving a trigger request from the client, the server responds by determining the game data for the current match. This game data must include at least the equipment attributes of the first piece of equipment configured for each virtual character participating in the match, reflecting the overall equipment configuration of the current game. Based on this determined game data, the server uses preset algorithms and model reasoning to predict the final outcome of the match, deriving possible results and obtaining a match outcome prediction. This prediction reflects the impact of the current equipment configuration on the game.

[0057] After predicting the match outcome, the server generates match prompts based on this prediction and according to preset rules. These prompts can be used individually to inform the player about the specific impact of their first equipment configuration action on the match result, or they can be used individually to recommend suitable first equipment for the player's virtual character, or both simultaneously. Finally, the server sends the generated match prompts to the corresponding client via the network. Upon receiving the prompts, the client updates the current display on the graphical user interface in real time, presenting the impact of equipment configurations and equipment recommendations to the player, allowing them to obtain relevant reference information in real time.

[0058] Thus, in this embodiment, before the game begins, the server responds to the trigger request sent by the client, determines the game data, and performs predictive processing on the game result based on the game data to obtain a predicted result. Then, based on the predicted result, it generates game prompts that can indicate the impact of equipment configuration operations on the game result or recommend the virtual character to use the first game equipment. Finally, the game prompts are sent to the client. This achieves pre-game result prediction and equipment configuration-related prompts based on the player's actual equipment configuration operations. Compared to providing equipment recommendations based on static rules or offline statistical data, which cannot provide suggestions based on the player's real-time equipment configuration operations... The game result feedback scheme ensures that the game equipment configuration-related prompts align with players' actual pre-match operational scenarios. By relying on real data, including the equipment attributes of each virtual character in the game, to predict game results, the accuracy and relevance of the predictions are effectively guaranteed. This provides reliable data support for the game prompts, enhancing their reference value. Furthermore, by transforming game result predictions into direct prompts about the impact of equipment configuration on victory or defeat, or equipment recommendations, players can intuitively understand the practical effects of equipment configuration without needing to analyze complex prediction data, thus reducing the cognitive load for players.

[0059] Furthermore, the server processes data, predicts results, and generates prompts in real time for players' equipment configuration operations, and sends them to the client. This achieves real-time linkage between players' equipment configuration operations and prompt feedback to a certain extent, allowing players to continuously receive targeted decision-making references during the pre-match equipment configuration process. This reduces the decision-making cost for players in pre-match equipment configuration to a certain extent, optimizes the player's operational experience in the game's equipment configuration process, and makes the game's equipment recommendation function meet the needs of actual matches, thereby enhancing the practical value of the equipment recommendation function.

[0060] In some embodiments provided in this application, the game match data also includes at least one of game map information, game match type information, and character attributes of each virtual character participating in the game match.

[0061] Specifically, using only the equipment attributes of each virtual character's first in-game item as match data for prediction suffers from a lack of data dimension. It fails to comprehensively reflect the impact of the actual game context and the virtual character's own characteristics on the match outcome. Differences in game map scenarios directly affect the actual effectiveness of equipment; different match types have different combat rules and victory / defeat logic; and the differences in virtual character attributes determine the player's skill level and the character's basic combat power. Without the support of these two types of data, the accuracy and reference value of the prediction results are easily insufficient, leading to a lack of targeted match prompts and an inability to provide scientific guidance for players' equipment configurations that align with the actual needs of the match.

[0062] To address the aforementioned issues, some embodiments provided in this application supplement the game match data with at least one of the following: game map information, game match type information, and the character attributes of each virtual character participating in the game match. This enriches the dimensions of the match data, enabling the server to perform match result prediction processing based on more comprehensive match data, thereby improving the accuracy of match result prediction and making match prompts more targeted. In some embodiments, game map information can be understood as various feature data representing the game map used in the current game match in the target game. It can reflect the characteristics of the match scene, including scene features related to the match process such as the map's track type, terrain structure, size, special mechanisms, and resource distribution.

[0063] In some embodiments, game match type information can be understood as feature data that characterizes the competition mode of this game match, used to distinguish different match rules and competition formats, including relevant information such as the match participation format, competition level, and win / loss determination rules, such as different types of match identifiers and rule features such as single-player competition, multi-player team competition, casual match, and ranked match.

[0064] In some embodiments, character attributes can be understood as various data that characterize the inherent and developmental features of each virtual character participating in a game match, and can reflect the comprehensive ability of the virtual character, including the virtual character's level, rank, historical win rate, recent performance, proficiency, enhancement level, frequency of use, and other relevant features that can reflect the character's ability and the player's skill level.

[0065] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: When the server determines game match data in response to a trigger request, the determined game match data includes, in addition to the equipment attributes of the first game equipment of each virtual character participating in the current game match, at least one of the following: game map information, game match type information, and character attributes of each virtual character participating in the game match. Specifically, the game map information reflects the scene characteristics of the current game match; different game maps have different performance requirements for equipment. For example, in racing games, the length and terrain of different tracks affect the speed and handling performance of vehicles. The game match type information reflects the competition rules of the current game match; there are significant differences in equipment configuration strategies between solo and team matches. Character attributes reflect the overall strength of the virtual character; virtual characters with higher ranks and higher proficiency can better utilize the effects of equipment.

[0066] In some embodiments, the type of supplementary information can be flexibly selected based on the actual scenario and prediction needs of the target game. It only needs to meet the requirement of including at least one of the following: game map information, game match type information, and the character attributes of each virtual character participating in the game match. This improves the dimensionality of the game match data, enabling it to represent the basic characteristics of the match from multiple levels, such as equipment, scenario, match rules, and character abilities. Subsequently, the server uses the supplemented game match data as a basis to predict the outcome of the match using preset algorithms and model inference methods. It is understandable that game map information and match type information directly determine the scenario characteristics and strategy adaptation direction of the game match, and the character attributes of virtual characters can affect game performance. Incorporating this information into the game match data analysis ensures that the generated match prompts are consistent with the current match scenario and the character's own characteristics. This, to a certain extent, improves the relevance of the prompts to the player's pre-match equipment configuration guidance, thereby enhancing the practical value of the configuration suggestions received by the player.

[0067] Thus, in this embodiment, the server responds to the client's trigger request based on the player's first equipment configuration operation for a virtual character. It determines game match data including at least one of the following: equipment attributes of the first game equipment for each virtual character participating in the game, game map information, game match type information, and character attributes of each virtual character participating in the match. Based on this game match data, the server performs predictive processing to obtain a predicted match result. This enriches the dimensions of the game match data. Compared to relying solely on equipment attributes for match prediction, this method fully combines the actual game scenario, match type, and the characteristics of the virtual characters, effectively improving the accuracy and reliability of the predicted match result. Simultaneously, the fusion of multi-dimensional information ensures that the final output match prompts have reliable data support, enhancing their reference value and helping players make decisions more aligned with the actual match situation during the pre-match equipment configuration phase. This effectively reduces the player's decision-making cost and optimizes the player's gaming experience during the equipment configuration process.

[0068] In some embodiments provided in this application, the prediction of the game result includes a target prediction probability, which includes: a first prediction probability that each virtual character participating in the game will achieve a target ranking at the end of the game, and / or a second prediction probability that the game faction to which each virtual character belongs will achieve a target game result at the end of the game.

[0069] Specifically, some technical solutions predict game outcomes in a vague sense, offering only a general trend or a simple overall result, lacking refined quantitative probability analysis. Players cannot know the specific expected performance of their virtual character or the winning probability of their faction, resulting in limited predictive value and making it difficult for players to make accurate equipment adjustments based on these predictions. To address these issues, some embodiments in this application explicitly define game outcome prediction as including a target prediction probability. This target prediction probability includes both a first prediction probability of a single virtual character achieving a target ranking and / or a second prediction probability of the game faction to which a single virtual character belongs achieving a target game outcome. By using these two types of quantitative probability indicators, multi-dimensional prediction of game outcomes is achieved, providing players with more specific game outcome references.

[0070] In some embodiments, the target prediction probability can be understood as a numerical indicator used to quantify the probability of various game outcomes after the server predicts the game results based on game data. In some embodiments, the first prediction probability can be understood as the predicted probability value for each virtual character participating in the game that the virtual character can achieve the target ranking at the end of the game, used to characterize the expected performance of a single virtual character in the game.

[0071] In some embodiments, a ranking can be understood as a ranking position with actual competitive significance set according to the game rules, win / loss determination logic, and competitive characteristics of the target game. In some embodiments, a target ranking can be understood as the preset ranking position of the virtual character after the game ends, which is usually a high position in the game.

[0072] In some embodiments, a game faction can be understood as a group of different virtual characters in the target game that are divided according to the game rules. The division is based on the team composition of the players. In a single-player game, a single virtual character can be regarded as an independent faction.

[0073] In some embodiments, the second predicted probability can be understood as the predicted probability value for the game faction to which each virtual character participating in the game belongs, indicating the likelihood that the game faction will achieve the target game result at the end of the game, and is used to characterize the expected outcome of the team's game. In some embodiments, the target game result can be understood as the competition result that the game faction may obtain after the game ends, determined according to the game rules of the target game, which usually includes two categories: victory and defeat. Some games may add other result types such as draw according to the rules.

[0074] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: After determining the game match data, the server uses this game match data as a basis to perform professional prediction processing on the result of this game match, generating a refined match result prediction including a target prediction probability. This target prediction probability includes two types of probability indicators. The first is the first prediction probability that each virtual character participating in the game match will achieve the target ranking at the end of the game match. The target ranking is set according to the actual rules of the game, usually a high position in the match, reflecting the virtual character's performance. For example, in some racing games, the individual competition rules stipulate that players win by achieving a top-3 ranking at the end of the game. The game's win / loss determination rules are strongly correlated with the top three rankings, the target ranking is specifically set as the top three, and the first prediction probability is specifically the probability that each virtual character will achieve a top-three ranking in the match. The second is the second prediction probability that the game faction to which each virtual character belongs will achieve the target match result at the end of the game match. For single-player matches, a single virtual character can be considered as an independent faction for win / loss probability determination. For team-based competitive matches, the probability of winning for each game faction can be quantified.

[0075] Table 1. Faction Victory / Defeat Determination Table

[0076] For example, in some racing games' team matches, the rules are that each team consists of 3 players who are randomly matched. The competition follows a strategy similar to the "Tian Ji Horse Racing" strategy, where the 1st, 2nd, and 3rd place finishers from each team are compared with the opposing team's players of the same rank. The team with the highest rank in a tie receives a flag, and a team wins if it has at least two flags at the end of the match. The team match's outcome is determined according to Table 1 above, which records the winning teams under different ranking combinations. If a team has two players in the top three, that team has approximately a 90% chance of winning. When faction A is ranked 1-2-3 and faction B is ranked 4-5-6, faction A wins; when faction A is ranked 1-2-x (where x represents other rankings) and faction B has three possible rankings, faction A wins; when faction A is ranked 1-3-x and faction B has three possible rankings, faction A wins; when faction A is ranked 2-3-4 and faction B is ranked 1-5-6, faction A wins; when faction A is ranked 2-3-5 and faction B is ranked 1-4-6, faction A wins; when faction A is ranked 2-3-6 and faction B is ranked 1-4-5, faction B wins. The server can flexibly choose to output the first prediction probability, the second prediction probability, or both of these probability indicators simultaneously, depending on the game type of the target game.

[0077] Thus, in this embodiment, after determining the game match data, the server performs prediction processing on the game match results based on the game match data to obtain a match result prediction including the target prediction probability. The target prediction probability includes the first prediction probability that each virtual character participating in the match will achieve the target ranking at the end of the match and / or the second prediction probability that the game faction to which each virtual character belongs will achieve the target match result. This upgrades the match result prediction from a single overall result judgment to a multi-dimensional probability quantitative analysis. Compared with the prediction method that gives a rough prediction of the outcome, it effectively realizes a refined quantitative expression of the match result. To a certain extent, it enables players to clearly know their personal match performance expectations under their own equipment configuration and the win / loss expectations of their faction, improving the accuracy of equipment configuration. It also takes into account the dual-dimensional probability prediction of personal target ranking and faction match result, which meets the needs of different match scenarios in competitive games, such as single-player competition and team confrontation. To a certain extent, it enables the generated match prompt information to be adapted to different gameplay, improving the pertinence of equipment configuration guidance for players. Furthermore, presenting prediction results in the form of probabilities improves the objectivity of judging the outcome of the game to a certain extent. This allows players to compare the actual effects of different equipment configurations based on specific probability values, effectively reducing the difficulty of players' pre-match equipment configuration decisions, improving the efficiency of players' decisions in the equipment configuration stage, and thus improving the rationality of players' equipment configuration choices to a certain extent, and optimizing the players' game experience in the equipment configuration stage.

[0078] In some embodiments provided in this application, the first equipment configuration operation is used to configure the first game equipment of a virtual character as the first target equipment. The game result prediction includes the target prediction probability of each virtual character when the first game equipment of the virtual character is the first target equipment and / or the target prediction probability of each virtual character when the first game equipment of the virtual character is a candidate equipment, wherein the candidate equipment is any equipment other than the first target equipment among all the equipment held by the virtual character.

[0079] Specifically, the match result prediction only outputs a single-dimensional probability based on the player's currently configured equipment, lacking a comparative prediction of other equipment held by the player. During the pre-match equipment adjustment phase, players cannot know how switching to other candidate equipment will change their personal ranking probability and their team's win / loss probability. This results in players lacking clear references when choosing among multiple equipment options, making it difficult to quickly determine the optimal equipment configuration. Based on the above problems, in some embodiments provided in this application, the server, when generating match result predictions, calculates the target prediction probability when the virtual character is configured with the first target equipment, and / or calculates the target prediction probability when configuring each candidate equipment, providing players with a probability comparison of different equipment configuration schemes so that players can understand the advantages and disadvantages of each configuration.

[0080] In some embodiments, the first target equipment can be understood as the first game equipment selected by the player for configuring the virtual character through a first equipment configuration operation. In some embodiments, candidate equipment can be understood as all other first game equipment available for the player to select and switch in the virtual character's equipment library, excluding the currently configured first target equipment.

[0081] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: In the pre-match preparation stage before the official start of the target game match, the player performs a first equipment configuration operation on the client's graphical user interface. The player's first equipment configuration operation is to configure the virtual character's first game equipment as the first target equipment. After the client detects the first equipment configuration operation, it generates a trigger request and sends it to the server. After responding to the received trigger request, the server determines the game match data for this game match and, based on the game match data, performs prediction processing for the match result of this game match.

[0082] In some embodiments, the server predicts a first predicted probability of each virtual character achieving the target ranking and / or a second predicted probability of each virtual character's game faction achieving the target match result, based on the first target equipment currently configured for the virtual character. In some embodiments, the server identifies any candidate equipment other than the first target equipment among all the first game equipment held by the virtual character, and for each candidate equipment, simulates a scenario where the virtual character is configured with that candidate equipment. Combining this with the current game match data, the server calculates the target predicted probability for each virtual character participating in the game match under that scenario. In some embodiments, the server can also simultaneously predict the target predicted probabilities in both of the above scenarios.

[0083] After calculating the predicted probabilities for all targets, the server integrates the prediction results for the primary target equipment with those for all candidate equipment. Following preset rules, it generates match-related prompts. These prompts can indicate the impact of the primary equipment configuration on the match outcome, recommend better primary equipment, or perform both functions simultaneously. Finally, the server sends the generated match-related prompts to the corresponding client. Upon receiving the prompts, the client presents a comparison of the predicted probabilities for different equipment configurations to the player, allowing them to compare the differences between the configurations.

[0084] Thus, in this embodiment, after determining the game match data, the server performs prediction processing on the game match results based on the game match data, obtaining the target predicted probability of each virtual character when the virtual character's equipment is the first target equipment and / or the target predicted probability of each virtual character when the virtual character's equipment is a candidate equipment. This realizes the prediction of the game results under different equipment choices. Compared with the scheme of predicting the results for a single equipment configuration, it allows players to directly obtain the expected game probability corresponding to all relevant equipment they hold, without having to repeatedly switch equipment and trigger predictions to obtain comparison information. This effectively reduces the number of times players have to repeatedly operate to trigger predictions, and to a certain extent reduces the comparison cost between different equipment for players, and improves the rationality of equipment configuration.

[0085] In some embodiments provided in this application, step 230 above includes: determining target data based on the target prediction probability of each virtual character, wherein the target data includes the result prediction probability of the target game faction to which the virtual character belongs obtaining the target game result at the end of the game, and / or the configuration guidance information of the virtual character's first game equipment; Based on the target data, generate game prompts.

[0086] Specifically, match result prediction includes a large amount of target prediction probability data. Directly displaying this data as a prompt to the player results in overly complex prompts, making it difficult for players to quickly extract reference information, reducing the practicality of the prompts, and affecting the efficiency of equipment adjustments. To address these issues, some embodiments provided in this application first extract the result prediction probability reflecting the overall win / loss prospects of the target game faction, and / or the first game equipment configuration guidance information for the virtual character, from the target prediction probabilities of all virtual characters. This forms target data focused on the needs of the current game match. Then, match prompts are displayed based on this target data, enabling players to quickly obtain reference information.

[0087] In some embodiments, target data can be understood as data extracted from target prediction probabilities that has direct reference value for player equipment selection. In some embodiments, outcome prediction probability can be understood as a numerical value representing the probability of the target game faction achieving the target match result, obtained by combining the target prediction probabilities of all virtual characters within the target game faction to which the virtual character belongs, and can be understood as a quantitative assessment of the overall match prospects of the faction.

[0088] In some embodiments, the target game faction can be understood as the game faction to which the virtual character belongs, or as a competitive unit in a team-based combat scenario. The outcome of the game for the target game faction is directly related to the virtual character's gameplay experience. In some embodiments, the configuration guidance information can be understood as guidance information generated based on target prediction probability analysis to guide players in choosing appropriate first game equipment, including equipment adaptation suggestions, team composition complementarity tips, and other practical content.

[0089] To more clearly illustrate the information processing method in the game provided in the embodiments of this application, please refer to the following exemplary description: After obtaining the target prediction probabilities corresponding to the first target equipment and candidate equipment, the server first extracts and analyzes data based on the target prediction probability of each virtual character to determine the target data.

[0090] Target data typically includes at least one of the following two aspects: first, the predicted probability of the target game faction to which the virtual character belongs achieving the target game outcome at the end of the match. This predicted probability is derived from the analysis of the target prediction probability and directly reflects the winning expectation of the target game faction under different equipment configurations. Second, the configuration guidance information for the virtual character's first game equipment. This configuration guidance information is generated by the server based on the comparison results of the target prediction probabilities and can provide players with clear equipment selection suggestions, such as recommending that players choose a certain candidate equipment or keep their current first target equipment. Target data can be flexibly selected according to the game scenario; it can include both the result prediction probability and configuration guidance information, or only one of them. After determining the target data, the server no longer directly uses the complex target prediction probability data. Instead, it generates game prompt information based on the extracted target data, ensuring the simplicity and relevance of the prompt information to a certain extent, and then sends the game prompt information to the client.

[0091] Thus, in this embodiment, after obtaining the target prediction probabilities corresponding to the first target equipment and candidate equipment, the server determines target data based on the target prediction probability of each virtual character, including the result prediction probability of the target game faction to which the virtual character belongs and / or the configuration guidance information of the virtual character's first game equipment. Based on this target data, the server generates game prompt information that can indicate the impact of the first equipment configuration operation on the game result and / or recommend the first game equipment used by the virtual character. This achieves the transformation from multi-dimensional prediction probabilities to prompt information. Compared to the solution of outputting raw probability data for players to interpret themselves, this effectively reduces the understanding cost of prediction results for players. Furthermore, by extracting the result prediction probability, the target prediction probability of a single virtual character is integrated into the overall win / loss probability at the faction level, allowing players to quickly grasp the game prospects of their faction without having to analyze and summarize scattered data themselves, thus reducing information filtering costs to some extent. Simultaneously, the configuration guidance information provides players with a clear direction for equipment adjustments, reducing the difficulty of equipment configuration decisions to some extent, thereby improving the decision-making efficiency of players in the pre-game equipment configuration stage. This allows players of different skill levels to quickly obtain equipment suggestions suitable for the current game, optimizing the player's gaming experience.

[0092] In some embodiments provided in this application, the outcome of a game match is determined by the ranking of each virtual character in each game faction. The step of determining the target data based on the target prediction probability of each virtual character includes: determining the average of the first prediction probabilities of each virtual character in each game faction as the outcome prediction probability of each game faction.

[0093] Specifically, in scenarios where the outcome of a game match is determined by the ranking of virtual characters within each game faction, there is a lack of a scientific and easily implemented method to derive the predicted probability of the game faction's outcome from the first predicted probability of the virtual characters. This results in low accuracy in calculating the probability of victory or defeat for each faction, or an overly complex calculation process that affects the server's processing efficiency and fails to meet the real-time requirements of pre-match equipment adjustments. Based on these issues, in some embodiments provided in this application, for scenarios where the outcome of a game match is determined by the ranking of virtual characters, the server directly determines the predicted probability of the outcome for that faction by the average of the first predicted probabilities of each virtual character within each game faction.

[0094] In some embodiments, the average first predicted probability of virtual characters within a faction can be understood as the sum of the first predicted probabilities of all virtual characters in the same game faction, divided by the total number of virtual characters in that game faction, so as to convert the individual ranking probability into the overall win or loss probability of the faction.

[0095] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: When the server determines the target data based on the target prediction probability, since the game result is jointly determined by the ranking of each virtual character in each game faction, the server collects the first prediction probability of all virtual characters in each game faction, and calculates the result prediction probability of the game faction by summing them and then dividing by the total number of virtual characters in the game faction, thereby realizing the standardized processing of converting the individual ranking expectation into the overall win / loss probability of the faction.

[0096] Thus, in this embodiment, when determining target data based on the target prediction probability of each virtual character, the average of the first prediction probability of each virtual character in each game faction is determined as the result prediction probability of each game faction. This achieves a quantitative calculation of the faction's winning probability. Compared to the scheme of judging the outcome of a faction based on the probability of individual characters or a simple sum of probabilities, this avoids the influence of the probability deviation of a single character on the overall judgment to a certain extent, making the result prediction probability more consistent with the actual game situation and improving the accuracy and credibility of the probability data. At the same time, the calculation method based on the average of the ranking probability of each character is consistent with the rule in the game that the outcome of a faction is determined by the ranking of all members. This makes the calculation logic of the result prediction probability consistent with the actual game outcome judgment logic, thereby improving the accuracy and credibility of the prediction results to a certain extent.

[0097] Please refer to the embodiments provided in this application as well. Figure 3 and Figure 4 , Figure 3 and Figure 4This is a schematic diagram illustrating an application scenario of the information processing method in a game provided in this application. The target data includes a first predicted probability when the first game equipment is the first target equipment, and a second predicted probability when the first game equipment is a candidate equipment. The step of generating game prompt information based on the target data includes: determining the probability level information corresponding to the first target equipment and the candidate equipment according to the relationship between the first predicted probability and the second predicted probability, wherein the probability level information is used to enable the client to display an identifier pattern corresponding to the probability level information. Based on the first game equipment of each virtual character in the target game faction, generate lineup prompt text, which is used to indicate the first game equipment that the virtual characters can use; The probability level information, lineup hint text, and the predicted probability of the first result are determined as the game prompt information.

[0098] Specifically, the in-game prompts only display a single probability value, forcing players to manually compare the probability differences between their primary target equipment and all candidate equipment, resulting in low decision-making efficiency. Furthermore, the lack of targeted suggestions based on the overall equipment configuration of the current faction makes it difficult for players to understand how their equipment choices fit the team composition, leading to insufficient practicality and usability of the in-game prompts and failing to fully realize the guiding value of probability prediction. To address these issues, some embodiments provided in this application compare the magnitude of the predicted probabilities of the first and second results, combine this with the faction's equipment configuration to generate team composition prompt text, and then integrate the probability information, team composition prompt text, and the predicted probability of the first result into the final in-game prompt information, improving the intuitiveness and relevance of the in-game prompts.

[0099] In some embodiments, the first outcome prediction probability can be understood as the probability value of the target game faction achieving the target match result when the first game equipment is the first target equipment, which can be understood as a quantitative assessment of the faction's overall match prospects. In some embodiments, the second outcome prediction probability can be understood as the probability value of the target game faction achieving the target match result when the first game equipment is a candidate equipment, which can be understood as a quantitative assessment of the faction's overall match prospects. In some embodiments, the probability level information can be understood as a judgment result obtained by the server by comparing the magnitude of the first outcome prediction probability and the second outcome prediction probability, used to characterize the degree of influence of different equipment on the win / loss probability of their respective factions.

[0100] In some embodiments, the identifier pattern can be understood as a visual graphic symbol displayed by the client based on probability information, used to quickly convey to players the relative probability of victory or defeat for different factions corresponding to different equipment, without requiring players to compare values ​​manually. In some embodiments, the faction suggestion text can be understood as targeted equipment selection suggestion text generated by the server based on the first game equipment configuration of all virtual characters within the target game faction.

[0101] To more clearly illustrate the information processing method in the game provided in the embodiments of this application, please also refer to... Figure 4 , Figure 5 The following is an illustrative explanation: The target data determined by the server includes the first predicted probability of the first game equipment being the first target equipment, and the second predicted probability of the first game equipment being a candidate equipment. The server compares the first predicted probability with the second predicted probability for each candidate equipment, and determines the probability information corresponding to the first target equipment and each candidate equipment based on the comparison results. This probability information reflects the difference in the winning probability of the target game faction under different equipment configurations. Furthermore, the server associates this probability information with an icon on the client's graphical user interface, allowing the client to visually display the probability levels through the icon. Subsequently, based on the first game equipment configuration of each virtual character in the target game faction and the comparison results of the probability information, the server generates a lineup suggestion text. This lineup suggestion text is in text form and directly indicates the first game equipment that the virtual character can use, providing players with clear equipment selection advice. Finally, the server integrates the determined probability information, the generated lineup suggestion text, and the first predicted probability to determine the final match prompt information, which is then sent to the client.

[0102] Thus, in this embodiment, the probability information corresponding to the first target equipment and candidate equipment is determined based on the relationship between the first result prediction probability and the second result prediction probability in the target data determined by the server. Based on the probability information, the client displays the corresponding identification pattern. At the same time, a lineup prompt text is generated based on the first game equipment of each virtual character in the target game faction. Finally, the probability information, the lineup prompt text, and the first result prediction probability are determined as the game prompt information and sent to the client. This achieves multi-dimensional integration and presentation of the game prompt information. Compared with the solution of providing a single recommendation list or raw probability data, it improves the readability and usability of the prompt information to a certain extent. At the same time, presenting the probability information in the form of identification patterns allows players to quickly judge the impact of different equipment choices on the faction's win rate without complex calculations, thereby reducing the information understanding cost to a certain extent. Furthermore, generating lineup prompts based on the equipment status of all virtual characters in the current faction can accurately indicate the direction for optimizing lineup configuration. To a certain extent, this avoids players falling into the misconception of judging solely on the strength of a single piece of equipment, guiding players to choose equipment from the perspective of team synergy. This effectively improves the rationality of lineup matching, thereby reducing the difficulty of player operation and decision-making to a certain extent, improving the efficiency and experience of pre-match equipment configuration, and enabling players of different skill levels to quickly make equipment choices that are suitable for the current game.

[0103] In some embodiments provided in this application, step 120 above includes: inputting game data into a pre-trained win / loss prediction model, so that the win / loss prediction model can perform prediction processing on the game results based on the game data to obtain a game result prediction.

[0104] Specifically, relying on manual statistics and simple data analysis to predict game match results is inefficient and struggles to identify the complex relationships between various factors in the game match data and the match outcome. This results in low accuracy, fails to meet the real-time requirements of frequent pre-match equipment adjustments in competitive games, and cannot provide players with reliable equipment configuration references. To address these issues, in some embodiments provided in this application, the server inputs the determined game match data into a pre-trained win / loss prediction model. This model automatically processes the game match result prediction, improving both efficiency and accuracy.

[0105] In some embodiments, the win / loss prediction model can be understood as a machine learning model or deep learning model based on game-related historical data and pre-trained. The win / loss prediction model takes game data as input, and through built-in algorithm logic and trained parameters, intelligently predicts the game results and outputs the game result prediction.

[0106] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: Before performing prediction processing, the server trains the win / loss prediction model based on a large amount of historical game data. This historical data includes past game data and corresponding actual game results. Through training, the win / loss prediction model learns the correlation between various factors in the game data and the game results, forming a pre-trained win / loss prediction model. Before the game starts, the server responds to the client's trigger request and determines the game data for this game. It then organizes the determined game data according to the input requirements of the win / loss prediction model and inputs it into the pre-trained model. After receiving the game data, the win / loss prediction model automatically analyzes and calculates the data based on the trained correlations, derives the game result for this game, completes the prediction processing, and outputs the corresponding game result prediction. The server then generates game prompt information based on the prediction result output by the win / loss prediction model.

[0107] Thus, in this embodiment, after the server responds to the client's trigger request and determines the game data for the current match, it inputs the game data into a pre-trained win / loss prediction model. This model then predicts the match outcome, thereby achieving professional and efficient match result prediction. Compared to prediction schemes relying on manual rules or simple statistical analysis, this improves the accuracy and reliability of the prediction process. Furthermore, the automated prediction process of the pre-trained win / loss prediction model can quickly process multi-dimensional game data, avoiding the tediousness and lag of manual analysis. Even when players frequently adjust their equipment before a match, the model can still output prediction results promptly, ensuring the real-time nature of the prompts. In addition, the standardized model prediction process ensures consistency of prediction results across different match scenarios, reducing the interference of subjective factors and providing stable data support for the generated match prompts. This helps players obtain more reliable equipment configuration references, thereby reducing the player's decision-making costs and optimizing the gaming experience.

[0108] In some embodiments provided in this application, the above-described step of inputting game data into a pre-trained win / loss prediction model so that the win / loss prediction model can predict the game result based on the game data and obtain the game result prediction includes: preprocessing the game data to obtain game processing data, wherein the preprocessing includes normalization processing and / or encoding processing. Game match processing data is input into a pre-trained win / loss prediction model, so that the win / loss prediction model can predict the game match results based on the game match processing data.

[0109] Specifically, raw game match data may include various types of features. Some data are numerical data such as equipment attribute values ​​and player levels, while other data are non-numerical data such as game modes and character roles. Furthermore, some data exhibit different dimensions and significant range differences. Directly inputting raw game match data into the win / loss prediction model makes it difficult for the model to accurately identify the relationships between features, affecting the accuracy and stability of the prediction results. To address these issues, in some embodiments provided in this application, the server preprocesses the game match data before inputting it into the win / loss prediction model. This preprocessing includes normalization and / or encoding. The processed game match data is then input into the win / loss prediction model for prediction, thereby improving the processing efficiency and prediction accuracy of the model to some extent.

[0110] In some embodiments, preprocessing can be understood as the normalization and standardization of the original data performed by the server before inputting the game data into the win / loss prediction model, in order to eliminate data noise, unify the data format, and make the data meet the input requirements of the win / loss prediction model, thereby improving the processing efficiency and prediction accuracy of the win / loss prediction model to a certain extent.

[0111] In some embodiments, normalization can be understood as a preprocessing method that uses a specific algorithm to map feature data of different dimensions and numerical ranges in game data to a unified numerical interval, thereby eliminating the influence of differences in data units and ensuring that each feature has equal weight in the training and prediction of the win / loss prediction model. In some embodiments, encoding can be understood as another preprocessing method, used to convert non-numerical features such as game modes and character positioning in game data into numerical data that the win / loss prediction model can recognize and calculate, ensuring that the win / loss prediction model can effectively parse and utilize various feature information.

[0112] In some embodiments, game match processing data can be understood as a standardized data set obtained after preprocessing, which meets the input format requirements of the win / loss prediction model and can be directly used for the prediction calculation of the win / loss prediction model.

[0113] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: After the server responds to the trigger request and determines the game match data, before inputting the game match data into the pre-trained win / loss prediction model, the original game match data is first preprocessed. Preprocessing mainly includes two methods: First, normalization processing. For numerical data with different dimensions and large range differences in the game match data, a normalization algorithm is used to convert the numerical data into values ​​within a uniform range to eliminate the differences in dimensions and ranges between data. Examples include the level of a virtual character, the attribute values ​​of equipment, the attack power of equipment, and the player's historical win rate. Second, encoding processing. For non-numerical data in the game match data, one-hot encoding, vector embedding, and other encoding methods are used to convert it into numerical codes that the model can recognize. Examples include the game match type (ranked or casual), the character role (damage dealer or support), and the game map type.

[0114] The server can preprocess the game data by selecting normalization and / or encoding, based on the actual type of the game data, to transform the raw game data into processed game data that meets the model's input requirements. For example, if the data includes both numerical and non-numerical features, the corresponding processing operations are performed according to the data type. Subsequently, the server inputs this processed game data into a pre-trained win / loss prediction model, which analyzes and calculates the optimized data to predict the game outcome and outputs the predicted result.

[0115] Thus, in this embodiment, before inputting game match data into the pre-trained win / loss prediction model, the server first preprocesses the game match data to obtain processed game match data. This processed data is then input into the pre-trained win / loss prediction model, which predicts the match outcome. This achieves adaptation between the game match data and the prediction model. Compared to directly inputting raw data into the model, this improves the accuracy and stability of the model's predictions to some extent. Furthermore, normalization eliminates the magnitude differences between data of different dimensions, avoiding model prediction bias caused by inconsistent data ranges such as equipment attributes and character parameters. Encoding transforms non-numerical information in the game match data into a standardized format recognizable by the model, broadening the model's data processing range and allowing classification information such as map type and match mode to effectively participate in prediction calculations. This enriches the analytical dimensions of the prediction model, reduces data noise interference with prediction results, improves model computational efficiency, and ensures real-time feedback requirements in pre-match equipment adjustment scenarios.

[0116] In some embodiments provided in this application, the game match data further includes game map information, game match type information, and the character attributes of each virtual character participating in the game match; the step of inputting the game match data into a pre-trained win / loss prediction model so that the win / loss prediction model can predict the game match result based on the game match data to obtain the predicted match result further includes: inputting the game map information, game match type information, virtual character's character attributes, and equipment attributes into the encoding sub-model of the win / loss prediction model so that the encoding sub-model can encode the received game map information, game match type information, character attributes, and equipment attributes to obtain the encoding information corresponding to the virtual character; The encoded information corresponding to each virtual character is input into the attention sub-model of the win / loss prediction model so that the attention sub-model can predict the influence between virtual characters in the game and obtain the influence information prediction result. The prediction results of the impact information are input into the prediction sub-model of the win / loss prediction model, so that the prediction sub-model can predict the game result based on the prediction results of the impact information and obtain the game result prediction.

[0117] Specifically, when processing multi-dimensional game data, the win / loss prediction model fails to effectively integrate different types of information and lacks modeling of the interaction relationships between virtual characters. This results in prediction results that cannot fully reflect the impact of teamwork and opponent interference in multi-player games, leading to insufficient prediction accuracy and making it difficult to meet the win / loss prediction needs of complex competitive games. Based on the above problems, in some embodiments provided in this application, the win / loss prediction model adopts a three-level sub-model architecture. The encoding sub-model uniformly encodes multi-dimensional game data, the attention sub-model models the interaction effects between virtual characters, and the prediction sub-model makes win / loss predictions based on the interaction effects, effectively improving the accuracy of the prediction.

[0118] In some embodiments, the encoding sub-model can be understood as a sub-module in the win / loss prediction model used to encode and transform multi-dimensional input data, capable of converting different types of game match data into unified format encoded information that the win / loss prediction model can process. For example, a preset encoding algorithm can be used to convert scattered map information, match type information, character attributes, equipment attributes, etc., into vector data in a unified format that can be parsed by the win / loss prediction model module. In some embodiments, the encoded information can be understood as standardized vector data obtained by the encoding sub-model after encoding multi-source game match data, which can reflect the characteristics of virtual characters and match scenes.

[0119] In some embodiments, the attention sub-model can be understood as a sub-module in the win / loss prediction model used to model the interaction between virtual characters in a game, capable of identifying the cooperation and restraint relationships between different characters. For example, it captures the interaction effects such as the cooperation effect between teammates and the restraint relationship between opponents, and outputs quantified influence information prediction results. In some embodiments, the attention sub-model can also be a graph neural network, with players as nodes and interactions between players as edges, similarly achieving the modeling of teammate cooperation and opponent interference, improving prediction accuracy through character relationship modeling. In some embodiments, the influence information prediction results can be understood as the result data output by the attention sub-model, quantifying the degree and direction of interaction between virtual characters in a game, including interaction features such as teammate cooperation contribution and opponent interference intensity.

[0120] In some embodiments, the prediction sub-model can be understood as the computational module of the win / loss prediction model. It takes the influence information prediction result output by the attention sub-model as input, and makes a final judgment on the win / loss trend of the game through the built-in inference algorithm, and outputs the game result prediction.

[0121] To more clearly illustrate the information processing method in the game provided in the embodiments of this application, please refer to... Figure 5 and the following exemplary description, Figure 5 This diagram illustrates an application scenario of the information processing method in a game provided in this application. Specifically, after determining the game match data, the server inputs game map information, game match type information, and the character attributes and equipment attributes of the virtual characters into the encoding sub-model of the win / loss prediction model. The encoding sub-model, through preset feature extraction and encoding algorithms, transforms multi-source data of different types and formats into a vector form of a unified dimension, generating encoded information corresponding to each virtual character, thus achieving effective fusion of multi-dimensional information. For example, one-hot encoding or vector embedding is used for map information and match type information, while character attributes and equipment attributes are normalized before encoding. Through unified encoding processing, various types of raw data are transformed into encoded information corresponding to virtual characters with a unified format, achieving the fusion of multi-source data.

[0122] Subsequently, the encoded information of all virtual characters is input into the attention sub-model of the win / loss prediction model. The attention sub-model calculates the correlation weights between the encoded information of different virtual characters to identify the interaction relationships between them, including teamwork, counter-relationships between opponents, and overall interference without distinguishing factions. It then outputs prediction results reflecting the impact of these interactions. For example, ... Figure 5As shown, taking Player 1 in a racing game as an example, the server first fuses the raw information such as Player 1 information, Car 1 information, track information, race information, and team member information through a neural network to obtain Player 1's information representation, i.e., the process of knowing oneself. Based on this, the information representations of Player 1, Player 2, Player 3, Player 4, Player 5, and Player 6 are input into the attention sub-model to further analyze three types of interactive influences, i.e., the process of knowing others: Interference from other players: This reflects the overall interference relationship without distinguishing factions, representing the combined impact of all opponents and teammates on the current player. Teammate cooperation: The collaborative support from teammates within the same game faction. Opponent interference: The restraint or suppression from opponents in the opposing faction. These interactions collectively constitute the predicted impact information, reflecting the complex game dynamics in multiplayer matches. Finally, the predicted impact information is input into the prediction sub-model of the win / loss prediction model. Based on the integrated interaction features and individual characteristics, the prediction sub-model uses built-in neural network classification, regression, and other inference algorithms to make a final prediction of the game's outcome. For example, the prediction sub-model can output the probability of each player entering the top three, i.e., Figure 5 Whether player 1 can enter the top 3 will provide a basis for subsequent equipment recommendations and win rate tips.

[0123] Thus, in this embodiment, after determining the game match data, the server inputs the game map information, game match type information, and the character attributes and equipment attributes of the virtual characters into the encoding sub-model of the pre-trained win / loss prediction model. The encoding sub-model encodes the above information to obtain the encoded information corresponding to the virtual characters. Then, the encoded information of each virtual character is input into the attention sub-model of the win / loss prediction model. The attention sub-model predicts the influence between virtual characters to obtain the influence information prediction result. Finally, the influence information prediction result is input into the prediction sub-model, which predicts the game match result to obtain the match result prediction. This achieves deep fusion of multi-source information and characterization of the interaction relationships between virtual characters, compared to relying on a single dimension. The prediction scheme, which either ignores or eliminates interactions between characters, improves the accuracy and relevance of game outcome predictions to some extent. Simultaneously, the encoding sub-model transforms multi-source information of different types and formats into unified encoded information, eliminating heterogeneity interference between various data types and improving the feature quality of the input model. The attention sub-model captures the complementary effects of teamwork and the restraining interference of opponents, enabling the prediction results to realistically reflect the complex game scenarios of multiplayer matches. The prediction sub-model makes the final prediction based on the encoded and integrated basic information and the interaction influence information between characters, improving the relevance and practicality of the prediction results to some extent. This, in turn, reduces the decision-making cost for players and enhances the competitive experience and playability of the game.

[0124] In some embodiments provided in this application, please refer to Figure 6 , Figure 6 This is a flowchart illustrating an information processing method in a game provided in this application embodiment. It should be noted that the steps shown may be executed in a different logical order than that shown in the flowchart. The first game equipment includes multiple components, each of which can be used with game virtual accessories. Each first game equipment has multiple sub-attributes as its equipment attributes. Each game virtual accessory is used to adjust at least one sub-attribute of the first game equipment. The method further includes: Step 250: Responding to a client's accessory recommendation request for a second game equipment received before the start of the game, determining a target accessory combination corresponding to the second game equipment, wherein the second game equipment is one of multiple first game equipment components, and the accessory recommendation request is generated by the client when it detects a preset trigger operation for the second game equipment and sent to the server. Step 260: Feedback the target component combination to the client.

[0125] Specifically, in competitive games, equipment recommendations focus on the selection of the primary game equipment itself, neglecting the compatibility and matching requirements between accessories and equipment. The performance of the primary game equipment often depends on the proper combination of accessories, and each primary game equipment includes multiple sub-attributes, corresponding to a wide variety of virtual game accessories, with different accessories having different effects on each sub-attribute. Players find it difficult to understand the compatibility logic between accessories and equipment within a limited time, and cannot quickly select the optimal combination from their own accessories, resulting in high decision-making costs for accessory combinations, reducing the rationality of accessory use and the gaming experience. Based on the above problems, in some embodiments provided in this application, the server responds to the client's accessory recommendation request for the second game equipment, combines the equipment attribute requirements and the player's actual accessories, determines the target accessory combination, and feeds it back to the client. The client then displays the recommendations through a graphical user interface, realizing equipment and accessory compatibility guidance.

[0126] In some embodiments, game virtual accessories can be understood as auxiliary equipment used in conjunction with the first game equipment to adjust at least one sub-attribute of the first game equipment. By combining different game virtual accessories, personalized optimization of equipment performance can be achieved to adapt to different game scenarios and player operation needs. Examples include chips, runes, and inscriptions. In some embodiments, sub-attributes can be understood as subdivided dimensions of the equipment attributes of the first game equipment. For example, the equipment's attack power can be subdivided into sub-attributes such as base attack power and critical attack power. Each sub-attribute collectively determines the overall performance of the equipment.

[0127] In some embodiments, the second game equipment can be understood as the specific piece of equipment among multiple first game equipment that triggers an accessory recommendation request by the player. In some embodiments, the accessory recommendation request can be understood as a request instruction sent by the client to the server to obtain a combination of accessories suitable for the second game equipment. In some embodiments, the preset triggering operation can be understood as a player operation pre-set by the client that can trigger the generation of the accessory recommendation request, such as clicking the accessory recommendation button, long-pressing the second game equipment icon, or other specific interactive behaviors.

[0128] In some embodiments, the target accessory combination can be understood as the optimal or most compatible combination of virtual game accessories selected by the server based on the attribute requirements of the second game equipment and the accessories held by the player.

[0129] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: Multiple first game equipment pieces are included, and each first game equipment piece has multiple sub-attributes. Game virtual accessories can adjust these sub-attributes. Before the start of a game match, players can perform two types of operations: First, they can configure the first equipment on their virtual character. The client generates a trigger request and sends it to the server. The server completes the determination of match data, result prediction, generation and sending of prompt information, and the client updates the interface. Second, they can execute a preset trigger operation on a specific second game equipment piece, i.e., one of the multiple first game equipment pieces. After detecting this preset trigger operation, the client generates an accessory recommendation request for the second game equipment and sends it to the server. The accessory recommendation request includes information such as the player identifier and the second game equipment identifier.

[0130] After receiving an accessory recommendation request, the server first determines the accessory compatibility direction based on the attribute requirements, sub-attribute characteristics, and potential game scenarios of the second game equipment. Then, it queries all virtual game accessories held by the virtual character, and, combined with the accessory usage rules of the second game equipment, filters out accessories that meet the criteria. Next, a preset algorithm is used to evaluate the combinations of these filtered accessories, analyzing the overall adjustment effect of different combinations on the sub-attributes of the second game equipment, and finally determining the optimal accessory combination. After determining the target accessory combination, the server sends it to the corresponding client. Upon receiving the information, the client displays the compatibility between the second game equipment and the target accessory combination through a graphical user interface, such as presenting accessory combination details in the form of a list or diagram, for easy viewing and selection by the player. In some embodiments, the client also supports players to replace the current accessory combination with a single click.

[0131] Thus, in this embodiment, before the game begins, the server can respond to the client's accessory recommendation request for the second game equipment received before the game starts, determine the target accessory combination corresponding to the second game equipment, and feed back the target accessory combination to the client. This achieves the adaptation recommendation of game equipment and virtual accessories. Compared with providing preset accessory templates or general accessory recommendation schemes that do not distinguish specific equipment, this improves the targeting and practicality of accessory recommendations to a certain extent. At the same time, the server generates target accessory combinations for specific second game equipment, which makes the sub-attribute combinations of the second game equipment more in line with the positioning and game needs of the second game equipment. This fully utilizes the potential capabilities of the second game equipment to a certain extent, avoids the imbalance of the second game equipment attributes caused by players blindly matching accessories, improves the player's performance in the game using the second game equipment, and thus optimizes the accessory matching experience in the game to a certain extent, enhancing the game's fun. In addition, the client can initiate a recommendation request through a preset trigger operation, and the server quickly feeds back the target accessory combination. Players do not need to study the adaptation relationship between different accessories and equipment sub-attributes themselves, which reduces the learning threshold and decision-making cost of accessory matching to a certain extent.

[0132] In some embodiments provided in this application, step 250 above includes: responding to an accessory recommendation request and determining equipment attribute adjustment bias information of the second game equipment, wherein the equipment attribute adjustment bias information is used to determine the adjustment bias of each sub-attribute of the second game equipment; From the virtual game accessories held by the virtual character, select the combination of virtual game accessories that meets the accessory usage constraint information of the second game equipment to obtain multiple candidate accessory combinations. The accessory usage constraint information is used to determine the virtual game accessories that can be used with the second game equipment. The target accessory combination includes at least one virtual game accessory. Based on the equipment attribute adjustment bias information and the sub-attribute adjustment amount of each virtual accessory in the candidate accessory combination, determine the accessory combination score of the candidate accessory combination; The candidate accessory combination with the highest accessory combination score is identified as the target accessory combination.

[0133] Specifically, the accessory recommendation scheme lacks clear screening criteria and a scientific evaluation system when determining target accessory combinations. This results in recommended preset combinations that fail to generate optimal combinations based on equipment attribute adjustment needs and the player's actual accessory situation, leading to insufficient practicality and relevance of the recommendation results. To address these issues, in some embodiments provided in this application, after the server responds to the accessory recommendation request, it first determines the equipment attribute adjustment bias information of the second game equipment, then filters candidate accessory combinations that meet the constraints, calculates a combination score based on the bias information and sub-attribute adjustment amounts, and finally selects the combination with the highest score as the target accessory combination.

[0134] In some embodiments, equipment attribute adjustment bias information can be understood as information determined by the server based on the positioning of the second game equipment, the needs of the game scenario, and the direction of performance optimization. This information is used to clarify the priority adjustment direction and optimization focus of each sub-attribute of the second game equipment, providing a basis for accessory combination selection. In some embodiments, accessory usage constraint information can be understood as the accessory adaptation rules inherent to the second game equipment. This information is used to limit the range of virtual game accessories that can be used with the second game equipment, including accessories compatibility requirements, installation quantity limits, level adaptation conditions, and other constraints.

[0135] In some embodiments, a candidate accessory combination can be understood as a feasible accessory combination selected from the game virtual accessories held by the virtual character that meets the usage constraint information of the second game equipment accessory.

[0136] In some embodiments, the sub-attribute adjustment amount can be understood as the adjustment value of a single virtual game accessory to a specific sub-attribute of the second game equipment, which can be used to measure the degree of influence of the accessory on the equipment performance. In some embodiments, the accessory combination score can be understood as a quantitative score obtained by the server after comprehensively evaluating the sub-attribute adjustment amounts of all accessories in the candidate accessory combination based on the equipment attribute adjustment bias information, which can be used to judge the suitability of the candidate accessory combination.

[0137] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: After receiving a request for accessory recommendations for a second game equipment, the server responds to the accessory recommendation request and, in conjunction with the second game equipment's positioning (racing, interference, assistance, etc.), as well as the game scenario (map, mode, etc.) and the accessory matching preferences of top players, determines the equipment attribute adjustment bias information for the second game equipment. This equipment attribute adjustment bias information clarifies the adjustment direction and priority of each sub-attribute. For example, racing equipment may prioritize the improvement of speed and acceleration sub-attributes.

[0138] Subsequently, the system queries the player's currently held virtual game accessories data. Based on the accessory usage constraints of the second game equipment, virtual game accessories that meet compatibility requirements, installation quantity limits, and level suitability conditions are selected from the player's accessory pool. Multiple feasible candidate accessory combinations are generated using a combination algorithm, with each candidate accessory combination containing at least one virtual game accessory. Next, the sub-attribute adjustment amount of each virtual game accessory in each candidate accessory combination is obtained—that is, the specific adjustment value of each sub-attribute of the second game equipment by the accessory. Combined with the equipment attribute adjustment bias information, a preset scoring algorithm, such as weighted summation, is used to calculate the accessory combination score for each candidate accessory combination. The scoring algorithm must reflect the degree of matching between different sub-attribute adjustment amounts and adjustment biases. Then, the accessory combination scores of all candidate accessory combinations are sorted, and at least one candidate accessory combination with the highest score is selected as the target accessory combination corresponding to the second game equipment. In some embodiments, if multiple optimal combinations with the same score exist, all can be included in the target accessory combination range.

[0139] After determining the target accessory combination, the server sends this information to the corresponding client. Upon receiving the target accessory combination information, the client displays the compatibility between the second game equipment and the target accessory combination through a graphical user interface, allowing players to quickly view and select the appropriate equipment.

[0140] Thus, in this embodiment, the server responds to the client's accessory recommendation request for the second game equipment received before the start of the game match, determines the equipment attribute adjustment bias information of the second game equipment. This equipment attribute adjustment bias information is used to determine the adjustment bias of each sub-attribute of the second game equipment. Then, it selects game virtual accessory combinations that meet the accessory usage constraints of the second game equipment from the game virtual accessories held by the virtual character, obtaining multiple candidate accessory combinations. Subsequently, based on the equipment attribute adjustment bias information and the sub-attribute adjustment amount of each game virtual accessory in the candidate combinations, it determines the accessory combination score of each candidate combination. Finally, it determines at least one candidate combination with the highest accessory combination score as the target accessory combination and feeds it back to the client, so that the client can recommend the second game equipment to be used with the target accessory combination through the graphical user interface. This achieves the refinement and optimization of game accessory recommendations, compared to the lack of clear attributes. The component recommendation scheme, which is geared towards performance adjustments and does not screen available components or provide quantitative evaluation, improves the rationality, usability, and adaptability of component recommendations to a certain extent. Meanwhile, the determination of equipment attribute adjustment bias information provides clear guidance for component recommendations, ensuring that recommended component combinations match the attribute optimization needs of the second game equipment. This effectively avoids a disconnect between component combinations and equipment positioning, and enhances the targeting of equipment sub-attribute adjustments. The introduction of component usage constraints effectively filters out candidate component combinations that meet equipment usage requirements, to a certain extent eliminating the recommendation of unusable or incompatible components, reducing the time cost for players to try ineffective combinations, and improving the practicality of the recommendations. Furthermore, by quantitatively evaluating candidate combinations through component combination scores, the system organically combines equipment attribute adjustment bias with component sub-attribute adjustment amounts, avoiding subjective recommendation bias to a certain extent and improving the credibility of the recommendation results.

[0141] In addition, selecting the candidate combination with the highest score as the target accessory combination ensures, to a certain extent, the compatibility of the recommended scheme with the needs of adjusting equipment attributes in the second game, optimizes equipment performance, and thus reduces the decision-making difficulty and learning threshold of accessory matching to a certain extent, thereby optimizing the player's gaming experience.

[0142] In some embodiments provided in this application, the method further includes: sending accessory combination scores to the client so that when the client recommends the second game equipment to be used with the target accessory combination, the accessory combination scores of the target accessory combination are displayed simultaneously.

[0143] Specifically, only the specific composition of the accessory combination is returned, without providing quantitative evidence of its suitability. When receiving recommendations, players cannot intuitively understand the specific advantages of the combination. If multiple target accessory combinations exist, it is also difficult to compare the differences in suitability between different recommended combinations, leading players to question the credibility of the recommendations and affecting decision-making efficiency. Based on the above problems, in some embodiments provided in this application, the server sends the corresponding accessory combination score along with the target accessory combination to the client. The client simultaneously displays this accessory combination score when recommending the use of the combination, providing players with a decision-making reference.

[0144] In some embodiments, the accessory combination score can be understood as a value obtained by the server combining the equipment attribute adjustment bias information of the second game equipment and comprehensively quantifying the adjustment amount of the sub-attributes of all virtual game accessories in the candidate accessory combination. It can be used to measure the compatibility between the accessory combination and the second game equipment. The higher the score, the stronger the compatibility and the better the improvement effect on equipment performance.

[0145] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: After determining the target accessory combination and its corresponding score, the server sends the specific composition information of the target accessory combination to the client and simultaneously sends the corresponding accessory combination score to the client. Upon receiving the above information, the client integrates and displays it through a graphical user interface. For example, in the details list of the target accessory combination, the accessory combination score of the target accessory combination is clearly marked, or in the comparison list of multiple recommended combinations, the score of each combination is displayed separately, allowing players to view it intuitively. In some embodiments, if the server recommends multiple target accessory combinations, the client can display each combination and its corresponding score in descending order of score, facilitating player comparison and selection.

[0146] Thus, in this embodiment, the server sends the corresponding accessory combination score along with the target accessory combination to the client. This allows the client to simultaneously display the accessory combination score when recommending a second game item to be used with the target accessory combination. This achieves quantitative transparency in accessory recommendations. Compared to solutions that only provide accessory combination recommendations without quantitative basis, this improves the credibility of the recommendation results and the efficiency of player decision-making. Furthermore, using the accessory combination score as a quantitative indicator directly reflects the degree of fit between the target accessory combination and the attribute adjustment needs of the second game item, avoiding questioning of the recommendation results and enhancing player trust in the recommendation function. The score display provides players with a clear decision-making reference, allowing them to quickly determine the suitability priority of different combinations without needing to analyze complex attribute relationships themselves. This reduces decision-making costs and, consequently, increases player acceptance and willingness to use the accessory recommendation function, optimizing the interactive experience of accessory matching.

[0147] In some embodiments provided in this application, the method further includes: if the number of selected game virtual accessory combinations is one, determining the selected game virtual accessory combination as the target accessory combination.

[0148] Specifically, when only one virtual game accessory combination meets the constraints and is selected from the player's accessory pool, performing the step of calculating the accessory combination score would result in unnecessary redundant calculations, reducing server-side processing efficiency, hindering the rapid feedback of recommendation results, and impacting the player's user experience. Based on the above issues, in some embodiments provided in this application, after the server obtains candidate accessory combinations through screening, it first determines whether the number of candidate accessory combinations is one. If it is one, the candidate accessory combination is directly designated as the target accessory combination, skipping the score calculation and sorting steps. If there are multiple combinations, the target combination is determined according to the original process. By simplifying the process through scenario-based approaches, the service response efficiency in special scenarios is effectively improved.

[0149] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: After the server obtains candidate accessory combinations through screening, it first determines the number of game virtual accessory combinations obtained through screening. If only one game virtual accessory combination that meets the usage constraint information of the second game equipment accessory is obtained after screening, the server skips the step of calculating the accessory combination score and directly determines the unique game virtual accessory combination as the target accessory combination corresponding to the second game equipment, without the need for score sorting and screening. If multiple candidate accessory combinations are obtained after screening, the server continues to calculate the accessory combination score of each combination and selects the combination with the highest score as the target accessory combination. After determining the target accessory combination, the server sends the specific composition information of the combination to the client, and the client displays and recommends it.

[0150] Thus, in this embodiment, after the server selects candidate accessory combinations, if the number of selected game virtual accessory combinations is one, it directly determines the combination as the target accessory combination and sends it back to the client. This allows the client to recommend a second game equipment to be used with the target accessory combination through the graphical user interface. This simplifies and improves the accessory recommendation process in special scenarios. Compared to a solution that requires full score calculation and sorting, this improves the recommendation response speed to a certain extent, reduces server resource consumption, and shortens the time players wait for feedback.

[0151] In some embodiments provided in this application, the step of selecting game virtual accessory combinations that satisfy the accessory usage constraint information of the second game equipment from the game virtual accessories held by the virtual character to obtain multiple candidate accessory combinations includes: selecting game virtual accessory combinations that satisfy the accessory usage constraint information of the second game equipment from the game virtual accessories held by the virtual character based on a predetermined accessory selection strategy to obtain multiple candidate accessory combinations.

[0152] Specifically, the lack of a unified selection strategy when screening candidate accessory combinations leads to inefficiency and potential omission of valid combinations, affecting the comprehensiveness and usability of the recommendations. To address these issues, in some embodiments provided in this application, the server pre-determines accessory selection strategies adapted to different scenarios. When screening candidate accessory combinations, the server uses these pre-determined strategies to select combinations from the player's virtual game accessories that conform to the second game equipment accessory usage constraint information, resulting in multiple candidate accessory combinations. This standardization strategy improves the efficiency and comprehensiveness of candidate accessory combination generation.

[0153] In some embodiments, the accessory selection strategy can be understood as a standardized rule system pre-defined by the server for selecting accessory combinations that meet the constraints from the virtual game accessories held by the virtual character. In some embodiments, the accessory selection strategy can be flexibly set according to the game scenario, the number of accessories, performance requirements, etc. Common types include an exhaustive strategy that traverses all feasible combinations when the number of accessories is small, a heuristic greedy strategy that prioritizes accessories that contribute more to the adjustment of sub-attributes and quickly generates high-quality combinations, and a dynamic programming strategy that finds near-optimal groups under multiple constraints, so as to balance the selection efficiency and combination quality.

[0154] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: The server pre-determines and stores accessory selection strategies adapted to different scenarios based on game design requirements, accessory combination complexity, computing performance requirements, etc. During the pre-match preparation stage before the official start of the target game, the player performs a preset trigger operation on the second game equipment, and the client generates an accessory recommendation request and sends it to the server.

[0155] After receiving an accessory recommendation request, the server first responds to the request, determines the bias information for adjusting equipment attributes based on the positioning of the second game equipment and the needs of the game scenario, and clarifies the optimization direction of sub-attributes. Then, it queries all game virtual accessories held by the virtual character, retrieves the pre-determined accessory selection strategy, and filters game virtual accessory combinations that meet the usage constraints of the second game equipment accessories such as compatibility, number of installations, and level adaptation. Multiple candidate accessory combinations are generated. For example, when the number of accessories held by the player is small, an exhaustive strategy is used to traverse all feasible combinations. When the number of accessories is large, a heuristic greedy strategy is used to prioritize accessory combinations that contribute highly to the adjustment of sub-attributes. When a near-optimal combination needs to be pursued and performance allows, a dynamic programming strategy is used for filtering.

[0156] After the filtering is complete, the server calculates the accessory combination score for each candidate combination based on the equipment attribute adjustment bias information and the sub-attribute adjustment amount of the virtual game accessories in each candidate combination, using a preset algorithm. The combinations are then sorted by score, and at least one with the highest score is selected as the target accessory combination. Finally, the server sends the specific composition information of the target accessory combination to the client, which then displays the recommendations intuitively through a graphical user interface, making it easier for players to choose.

[0157] Thus, in this embodiment, the server responds to the client's accessory recommendation request for the second game equipment received before the start of the game, determines the equipment attribute adjustment bias information of the second game equipment, selects game virtual accessory combinations that meet the accessory usage constraints of the second game equipment from the virtual accessories held by the virtual character, based on a pre-determined accessory selection strategy, and obtains multiple candidate accessory combinations. Then, based on the equipment attribute adjustment bias information and the sub-attribute adjustment amount of each candidate combination, the accessory combination score is determined, and at least one candidate combination with the highest score is determined as the target accessory combination and fed back to the client, so that the client can recommend the second game equipment to be used with the target accessory combination through the graphical user interface. This realizes the standardization and efficiency of the candidate accessory combination selection process. Compared with the scheme of blindly enumerating all possible combinations without a clear selection strategy, it improves the selection efficiency to a certain extent and reduces the computational burden of the server. Meanwhile, the pre-determined accessory selection strategy provides clear guidance for the screening of candidate combinations, enabling targeted screening of accessory combinations that meet the usage requirements of the second game equipment and are suitable for the player's actual holdings. This avoids the generation of invalid combinations and unnecessary score calculations to a certain extent, shortens the processing time of the recommendation process, and the strategic selection method can also ensure the consistency and relevance of the screening results. To a certain extent, it ensures that the generated candidate accessory combinations are all based on the attribute adjustment requirements of the second game equipment, improving the rationality and accuracy of the target accessory combination recommendations.

[0158] In some embodiments provided in this application, the equipment attribute adjustment bias information includes the bias weight of each sub-attribute of the second game equipment. The step of determining the accessory combination score of the candidate accessory combination based on the equipment attribute adjustment bias information and the sub-attribute adjustment amount of each game virtual accessory in the candidate accessory combination includes: summing the sub-attribute adjustment vector of each game virtual accessory in the candidate accessory combination to obtain the adjustment vector corresponding to the candidate accessory combination, wherein the sub-attribute adjustment vector is used to indicate the adjustment amount of the game virtual accessory for each sub-attribute; Vector calculations are performed on the bias weights and the adjustment vectors corresponding to the candidate component combinations to obtain the component combination score of the candidate component combinations.

[0159] Specifically, the lack of a clear calculation method for determining the accessory combination score based on equipment attribute adjustment bias information and sub-attribute adjustment amounts leads to a lack of scientific rigor and uniformity in score calculation. This may result in scoring results that fail to accurately reflect the compatibility between candidate accessory combinations and the second game equipment, affecting the accuracy of target accessory combination selection. To address these issues, in some embodiments provided in this application, the server first sums the sub-attribute adjustment vectors of each virtual game accessory in the candidate accessory combination to obtain an adjustment vector reflecting the overall adjustment effect of the combination. Then, through vector calculation, this adjustment vector is fused with a bias weight representing optimization priority to obtain a quantified accessory combination score.

[0160] In some embodiments, the sub-attribute adjustment vector can be understood as vector data used to quantify the adjustment effect of game virtual accessories on each sub-attribute of the second game equipment. Each dimension of the vector corresponds to the adjustment value of a sub-attribute, and the adjustment value directly indicates the amount of adjustment of the accessory on the corresponding sub-attribute. In some embodiments, the adjustment vector can be understood as a comprehensive vector obtained by summing the sub-attribute adjustment vectors of all game virtual accessories in a single candidate accessory combination according to their corresponding dimensions, which centrally reflects the overall adjustment effect of the accessory combination on all sub-attributes of the second game equipment.

[0161] In some embodiments, bias weight can be understood as a quantitative representation of the bias information of equipment attribute adjustment. It is a vector data that corresponds one-to-one with the sub-attribute dimension. The value of each dimension represents the optimization priority and importance of the corresponding sub-attribute. The higher the weight, the more critical the impact of the sub-attribute on the improvement of equipment performance.

[0162] In some embodiments, vector computation can be understood as the mathematical operation performed on the bias weight vector and the adjustment vector, and the information of the two vectors is fused through preset algorithms such as dot product operation and weighted accumulation, and finally a single quantified value is output as the accessory combination score.

[0163] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: The equipment attribute adjustment bias information includes the bias weight of each sub-attribute of the second game equipment. This bias weight reflects the adjustment priority of each sub-attribute. For example, the bias weight of speed is higher than that of control. When the server calculates the accessory combination score of a candidate accessory combination, it first obtains the sub-attribute adjustment vector of each virtual game accessory in the candidate accessory combination. Each dimension of the vector corresponds to a sub-attribute of the second game equipment, and the vector value is the adjustment amount of the accessory for that sub-attribute, for example, +5 or -2.

[0164] Subsequently, the adjustment vectors of the sub-attributes of all virtual game accessories in the candidate accessory combination are summed element-wise. This sums the adjustment amounts corresponding to the same sub-attribute to obtain the adjustment vector corresponding to the candidate accessory combination. This adjustment vector reflects the overall adjustment effect of the combination on each sub-attribute of the second game equipment. Next, a preset vector calculation algorithm is invoked to perform vector calculations with the bias weight vector in the equipment attribute adjustment bias information. For example, a dot product operation is used to multiply the values ​​of each dimension of the adjustment vector by the corresponding sub-attribute bias weight and then sum them to obtain a single quantified value, namely the accessory combination score of the candidate accessory combination. Other vector calculation methods, such as weighted summation, can also be used according to actual needs. After the scores of all candidate combinations are calculated, the server sorts them from highest to lowest score, selects at least one candidate combination with the highest score as the target accessory combination, and sends the specific composition information of the target accessory combination to the client. After receiving the information, the client intuitively displays the matching relationship between the second game equipment and the target accessory combination through a graphical user interface, making it convenient for players to choose.

[0165] Thus, in this embodiment, the server sums the sub-attribute adjustment vectors of each virtual game accessory in the candidate accessory combination to obtain the adjustment vector corresponding to the candidate accessory combination. Then, it performs vector calculation on the bias weight and the adjustment vector to obtain the accessory combination score of the candidate accessory combination. Finally, it determines at least one candidate combination with the highest accessory combination score as the target accessory combination and feeds it back to the client, so that the client can recommend a second game equipment to be used with the target accessory combination through the graphical user interface. This achieves the accuracy and rationality of accessory combination score calculation. Compared with the scheme that does not distinguish the priority of sub-attributes, simply superimposes accessory attributes, or subjectively evaluates the suitability of the combination, it improves the scientificity and suitability of score calculation to a certain extent. Simultaneously, by summing the sub-attribute adjustment vectors of each virtual accessory in the game within each candidate accessory combination, the overall adjustment effect of the combination on the sub-attributes of the second game equipment can be integrated. This avoids neglecting the overall attribute synergy effect of the combination by viewing the adjustment effect of a single accessory in isolation. This allows the adjustment vector to truly reflect the actual attribute changes of the combination. The biased weights define the adjustment priority and demand level of the second game equipment for different sub-attributes. By organically combining this priority with the overall adjustment effect of the combination through vector calculation, the degree of fit between the candidate combination and the equipment attribute adjustment needs can be quantified. This makes the accessory combination score directly correspond to the strength of adaptability, thereby improving the physical meaning and reference value of the score results to a certain extent. This also avoids the bias caused by subjective judgment to a certain extent, ensuring that the recommended target accessory combination meets the attribute requirements of the second game equipment, reducing decision-making costs, and optimizing the player's equipment configuration experience.

[0166] In some embodiments provided in this application, the method further includes: acquiring first game performance data, second game performance data, first accessory combination, and second accessory combination for each first game equipment, wherein the first game performance data is used to indicate the performance of the first game equipment in a game when the accessory combination used with the first game equipment is the first accessory combination, and the second game performance data is used to indicate the performance of the first game equipment in a game after the accessory combination used with the first game equipment is changed from the first accessory combination to the second accessory combination; The sub-attribute adjustment vectors of each virtual game accessory in the first accessory combination are summed to obtain the first adjustment vector corresponding to the first accessory combination; The sub-attribute adjustment vectors of each virtual game accessory in the second accessory combination are summed to obtain the second adjustment vector corresponding to the second accessory combination. The first game performance data, second game performance data, first adjustment vector, and second adjustment vector of each first game equipment are input into the pre-trained weight prediction model. The weight prediction model then predicts the bias weight of each sub-attribute of each first game equipment based on the first game performance data, second game performance data, first adjustment vector, and second adjustment vector of each first game equipment, thus obtaining the bias weight of each sub-attribute of each first game equipment.

[0167] Specifically, the equipment attribute adjustment bias information includes bias weights, and the accessory combination score is obtained through vector calculation. In some technical solutions, the bias weights rely on manual presets or simple statistics, which makes it impossible for the bias weights to accurately reflect the actual impact of different sub-attributes on the equipment's performance in the game. This, in turn, affects the accuracy of the accessory combination score calculation and the recommendation effect of the target accessory combination. Based on the above problems, in some embodiments provided in this application, by collecting the performance data of the first game equipment under different accessory combinations, calculating the adjustment vector of the corresponding accessory combination, and inputting the performance data and adjustment vectors into a pre-trained weight prediction model, the model automatically learns and outputs the bias weights of each sub-attribute of each first game equipment. This data-driven approach improves the accuracy and dynamic adaptation capability of the bias weights.

[0168] In some embodiments, the first match performance data can be understood as the performance indicators of the equipment recorded in a match when the first game equipment is paired with the first accessory combination. This data can be used to measure the suitability of the accessory combination, including indicators reflecting the equipment's actual combat performance such as win rate, output efficiency, survival time, and key skill hit rate. In some embodiments, the second match performance data can be understood as the performance indicators recorded in a match after the accessory combination of the first game equipment is changed from the first accessory combination to the second accessory combination. This data is used to compare with the first match performance data to evaluate the impact of the accessory combination change on the equipment's performance.

[0169] In some embodiments, the first accessory combination can be understood as the combination of virtual game accessories that the first gaming equipment is equipped with at a certain stage, and can be understood as the baseline combination before the accessory change, providing a reference for performance comparison. In some embodiments, the second accessory combination can be understood as the combination of virtual game accessories that the first gaming equipment replaces after the first accessory combination, and can be understood as the comparison combination after the accessory change, used to observe the performance changes brought about by the change.

[0170] In some embodiments, the first adjustment vector can be understood as a comprehensive vector obtained by summing the sub-attribute adjustment vectors of all game virtual accessories in the first accessory combination according to their corresponding dimensions, which centrally reflects the overall adjustment effect of the first accessory combination on each sub-attribute of the first game equipment. In some embodiments, the second adjustment vector can be understood as a comprehensive vector obtained by summing the sub-attribute adjustment vectors of all game virtual accessories in the second accessory combination according to their corresponding dimensions, which centrally reflects the overall adjustment effect of the second accessory combination on each sub-attribute of the first game equipment.

[0171] In some embodiments, the weight prediction model can be understood as a pre-trained machine learning model or deep learning model that learns the correlation between the performance differences before and after the change of accessory combination and the change of adjustment vector, and automatically predicts the bias weight of each sub-attribute of the first game equipment, thereby achieving dynamic optimization of the bias weight.

[0172] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: The server pre-trains a weight prediction model, which has learned the correlation between accessory combination adjustments and equipment performance changes through large-scale historical data. The server first collects match data of all players using each first game equipment, extracts match records with accessory combination changes from the match data, obtains the corresponding first accessory combination, second accessory combination, and first match performance data and second match performance data when these two accessory combinations are used. The first match performance data corresponds to the usage scenario of the first accessory combination, and the second match performance data corresponds to the scenario after the accessory combination is changed to the second accessory combination.

[0173] Subsequently, the adjustment vectors of the sub-attributes of each virtual game accessory in the first accessory combination are summed to obtain the first adjustment vector. Similarly, the second adjustment vector is calculated. These two vectors reflect the differences in the adjustment of the sub-attributes of the first game equipment by the two accessory combinations. Next, the first and second match performance data, the first adjustment vector, and the second adjustment vector corresponding to each piece of the first game equipment are input into the weight prediction model. The model analyzes the correlation between the differences in performance data and the adjustment vectors, automatically learning the contribution of each sub-attribute to the improvement of equipment performance. It then outputs the bias weight of each sub-attribute of the first game equipment and stores these weights in the server database, updating incrementally periodically with new match data. When a player initiates an accessory recommendation request for the second game equipment, the server reads the latest bias weight of the equipment from the database, filters candidate accessory combinations, calculates the accessory combination score, determines the target accessory combination, and sends the result back to the client.

[0174] Thus, in this embodiment, the server first obtains the first game performance data, the second game performance data, the first accessory combination, and the second accessory combination for each first game equipment. It then sums the adjustment vectors of the sub-attributes of each virtual game accessory in the first accessory combination to obtain the first adjustment vector, and sums the adjustment vectors of the sub-attributes of each virtual game accessory in the second accessory combination to obtain the second adjustment vector. The server then inputs the first game performance data, the second game performance data, the first adjustment vector, and the second adjustment vector for each first game equipment into a pre-trained weight prediction model. This allows the weight prediction model to predict the bias weights of each sub-attribute of each first game equipment based on these data, thereby obtaining the bias weights of each sub-attribute of each first game equipment. This achieves intelligent, data-driven generation of bias weights, which, compared to schemes relying on manual settings, static rules, or single statistical data to determine attribute weights, improves the accuracy, fit, and dynamic adaptability of the bias weights to a certain extent. Meanwhile, the performance data of the first and second matches directly reflect the impact of changes in accessory combinations on the actual performance of the first game equipment. To a certain extent, this provides a reliable data foundation for learning biased weights. Furthermore, by summing the adjustment vectors of the sub-attributes of the first and second accessory combinations respectively, the corresponding adjustment vectors can be obtained. This allows the model to quantify the comprehensive adjustment effect of different accessory combinations on the equipment sub-attributes, enabling the model to clearly capture the relationship between attribute changes and performance improvement. To a certain extent, this provides a clear quantitative input for weight prediction.

[0175] Furthermore, the weighted prediction model can automatically discover the contribution of each sub-attribute of equipment to performance without continuous manual intervention and adjustment, which reduces system maintenance costs to a certain extent. It can also capture the impact of dynamic changes such as game version updates and equipment attribute adjustments, effectively achieving dynamic optimization of bias weights. This, in turn, improves the personalization and targeting of accessory recommendations to a certain extent, reduces the decision-making cost of equipment configuration, and optimizes the player's gaming experience.

[0176] In some embodiments provided in this application, the steps of obtaining the first game performance data and the second game performance data of each first game equipment include: obtaining the first game performance data, the second game performance data, the first accessory combination, and the second accessory combination of the target virtual character using the first game equipment, wherein the target virtual character is a preset number of virtual characters ranked first in the character ranking data corresponding to the first game equipment.

[0177] Specifically, directly using match data from all virtual characters using game equipment may include a large amount of invalid data from virtual characters with average or poor performance. This would cause the trained bias weights to fail to accurately reflect the optimal accessory combination logic of the first game equipment, making it difficult to convey the combination experience of high-level players, affecting the rationality and practical value of target accessory combination recommendations, and failing to meet the needs of ordinary players for high-quality equipment combination guidance. Based on the above problems, in some embodiments provided in this application, when collecting match performance data, the server only selects the relevant data of the top-ranked target virtual characters in the character ranking data corresponding to the first game equipment, excluding noisy data from virtual characters with average performance. By focusing on high-quality data, the accuracy of the bias weights is improved, thereby optimizing the accessory recommendation effect.

[0178] In some embodiments, a target virtual character can be understood as a top-performing virtual character selected from all virtual characters using a specific first-level game equipment, based on character ranking data. The target virtual character's in-game performance and equipment combinations have high reference value. For example, a virtual character controlled by a top player. In some embodiments, character ranking data can be understood as a dataset formed by the server after quantifying and ranking virtual characters' in-game performance indicators using a specific first-level game equipment. The ranking dimensions include indicators that reflect the virtual character's actual combat ability using the equipment, such as win rate, average ranking, and performance data.

[0179] In some embodiments, the preset number can be understood as a pre-set standard by the server for filtering target virtual characters, in order to lock in a group of high-performing virtual characters. The preset number can be flexibly set according to the scale of the game players, data quality requirements, etc., for example, the top 5%, the top 10, or the top 100.

[0180] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: When collecting game match data, the server first sorts all virtual characters corresponding to each first game equipment. The sorting is based on the virtual character's match performance data when using the equipment, such as win rate, average ranking, total score, etc. Subsequently, the server selects a predetermined number of virtual characters ranked at the top of the sorting results as target virtual characters. The match performance of these virtual characters is among the top in the entire server, making the decision on changing accessory combinations more valuable. The server only collects the first match performance data, second match performance data, first accessory combination, and second accessory combination of these target virtual characters using the first game equipment, generates adjustment vectors, and inputs them into the weight prediction model to train and obtain the bias weights of each sub-attribute.

[0181] When a player initiates a component recommendation request for a second piece of game equipment, the server reads the bias weight of the second piece of game equipment obtained from the target virtual character data from the database, filters candidate component combinations, calculates component combination scores, determines the target component combination, and feeds it back to the client.

[0182] Thus, in this embodiment, when acquiring relevant data for training the weight prediction model, the system acquires the first and second match performance data, first accessory combinations, and second accessory combinations of the top-ranked virtual characters in the character ranking data corresponding to the first game equipment. Then, it sums the sub-attribute adjustment vectors of the first and second accessory combinations to obtain the first and second adjustment vectors, respectively. This data is then input into the weight prediction model to obtain the bias weights of each sub-attribute of the first game equipment. This optimizes the data source for bias weight learning. Compared to using data from ordinary players or random players for model training, this improves the quality and effectiveness of the training data, effectively avoiding data noise interference caused by ineffective trial and error or unreasonable combinations by ordinary players. This, to a certain extent, improves the effectiveness and practical adaptability of the bias weights, thereby enhancing the accuracy and practicality of accessory recommendations.

[0183] In some embodiments provided in this application, the method further includes: determining the target win rate when the second game equipment is used in combination with the target accessory, and sending the target win rate to the client so that the client can synchronously display the target win rate when recommending the use of the second game equipment in combination with the target accessory.

[0184] Specifically, some technical solutions only recommend target accessory combinations for the second game equipment to the client. Players cannot know the expected win rate of this combination and can only judge indirectly through the accessory combination score. This lack of direct reference for victory or defeat leads to insufficient trust in the recommendations and high decision-making costs. To address these issues, in some embodiments provided in this application, after determining the target accessory combination, the server additionally calculates the target win rate when the target accessory combination is used with the second game equipment. The server then sends the target accessory combination and the target win rate together to the client. When the client recommends the accessory combination in the graphical user interface, it simultaneously displays the corresponding target win rate, providing players with sufficient decision-making reference.

[0185] In some embodiments, the target win rate can be understood as a numerical value calculated by the server based on multi-dimensional data, representing the probability of winning in the current or similar game scenario when the second game equipment and the target accessory combination are used together. It can be used to measure the practical adaptability and competitiveness of the second game equipment and accessory combination. In some embodiments, the target win rate is calculated based on factors affecting the game outcome, such as the historical win rate of the target accessory combination, the lineup configuration of the current game, the characteristics of the game map, and the rules of the game type, and can directly reflect the practical value of the accessory combination.

[0186] To more clearly illustrate the information processing method in the game provided in this application embodiment, please refer to the following exemplary description: After determining the target component combination, the server calculates the target win rate of the combination in the current game scenario based on the historical match data of the second game equipment and the target component combination, the scene information of the current game, and the game's balance rules, using a preset algorithm or by calling a win / loss prediction model. For example, if model prediction is used, the win / loss prediction model is called, and the current game scenario information, such as the map, game mode, and team composition, as well as the attribute data of the second game equipment and the target component combination, are input to predict the target win rate. Subsequently, the server sends the specific composition information of the target component combination and the calculated target win rate to the client. After receiving the information, the client integrates and displays it through a graphical user interface. For example, the details list of the target component combination clearly indicates: "Recommended combination win rate 68%", allowing players to intuitively view the actual winning probability of the combination.

[0187] Thus, in this embodiment, after determining the target accessory combination, the server further determines the target win rate when the second game equipment is used in conjunction with the target accessory combination, and sends this target win rate to the client. This allows the client to simultaneously display the target win rate when recommending the second game equipment to be used in conjunction with the target accessory combination. This achieves a binding between accessory recommendations and win rate expectations. Compared to a solution that only recommends accessory combinations, this enhances the persuasiveness of the recommendation function and the scientific nature of player decision-making to a certain extent. Simultaneously, the target win rate intuitively quantifies the expected practical effect of using the second game equipment in conjunction with the target accessory combination, allowing players to clearly understand the probability of winning or losing in a match. This avoids blindly choosing recommended solutions to a certain extent, increases player trust in the recommendation results, reduces decision-making difficulty and trial-and-error costs, and thus optimizes the player's game interaction experience to a certain extent, improving the player's operational efficiency and satisfaction in the equipment configuration process.

[0188] In some embodiments provided in this application, please refer to Figure 7 , Figure 7 This is a flowchart illustrating an information processing method in a game provided in this application embodiment. It should be noted that the steps shown may be executed in a different logical order than those shown in the flowchart. This method is applied to a client, which communicates with a server. The client provides a graphical user interface (GUI) that displays the game screen after the start of a match in the target game. The game screen includes a controlled virtual character controlled by the client. The method includes: Step 270: Responding to a first equipment configuration operation before the start of the game, generating a trigger request and sending it to the server, so that the server responds to the received trigger request, determines game data, and performs prediction processing on the game result based on the game data to obtain a predicted game result, and generates a game prompt message based on the predicted game result, and sends the game prompt message to the client. The first equipment configuration operation is used to configure the first game equipment of the virtual character. The game data includes at least the equipment attributes of the first game equipment of each virtual character participating in the game. The game prompt message is used to indicate the impact of the first equipment configuration operation on the game result and / or to recommend the first game equipment used by the virtual character. Step 280: Respond to the game prompt information sent by the receiving server, and update the current display content of the graphical user interface according to the game prompt information.

[0189] Specifically, the embodiments of this application are basically the same as the embodiments corresponding to steps 210 to 240 above, the only difference being the implementing entity. Steps 210 to 240 use the server as the implementing entity to implement the complete flow of the information processing method in the game. Steps 270 and 280 use the client as the implementing entity to implement the corresponding flow of the information processing method in the game. The implementation flow corresponding to the client is consistent with the implementation flow of the server. For the specific implementation flow, please refer to the specific implementation flow of the server above, which will not be repeated here.

[0190] In some embodiments provided in this application, please refer to Figure 8 , Figure 8 This is a flowchart illustrating an information processing method in a game provided in an embodiment of this application. It should be noted that the steps shown may be executed in a logical order different from that shown in the flowchart. Multiple first game pieces are included, each capable of being used with virtual game accessories. Each first game piece has multiple sub-attributes as its equipment attributes. Each virtual game accessory is used to adjust at least one sub-attribute of the first game piece. The method further includes: Step 290: Responding to a preset trigger operation for a second game piece, generating an accessory recommendation request for the second game piece, and sending the accessory recommendation request to the server, so that the server responds to the received accessory recommendation request, determines a target accessory combination corresponding to the second game piece, and feeds back the target accessory combination to the client, wherein the second game piece is one of multiple first game pieces. Step 300: Respond to the target accessory combination sent by the receiving server, and recommend the second game equipment to be used in conjunction with the target accessory combination through the graphical user interface.

[0191] Specifically, the embodiments of this application are basically the same as the embodiments corresponding to steps 250 to 260 above, the only difference being the implementing entity. Steps 250 to 260 use the server as the implementing entity to implement the complete process of the information processing method in the game. Steps 290 and 300 use the client as the implementing entity to implement the corresponding process of the information processing method in the game. The implementation process corresponding to the client is consistent with the implementation process of the server. For the specific implementation process, please refer to the specific implementation process of the server above, which will not be repeated here.

[0192] To more clearly illustrate the information processing method in the game provided in the embodiments of this application, please also refer to... Figure 9 , Figure 10 and Figure 11The following exemplary description illustrates that: In this application embodiment, a server-client communication connection architecture is adopted. The server includes a game server, a win rate prediction artificial intelligence (AI) service module, a chip recommendation AI offline learning module, and a database. The client provides a graphical user interface and displays the game preparation interface during the pre-match equipment adjustment stage, i.e., the equipment adjustment stage.

[0193] like Figure 9 As shown, the win rate prediction AI is responsible for receiving requests, model prediction, post-processing, returning prediction results, verifying results, and updating win rates. The server is responsible for responding to player car adjustments by organizing data, storing the preference weights of different vehicles for chip attributes, obtaining chip attribute weights, and recommending the highest-scoring combinations from the chip pool based on these weights. The game client also involves storing preference weights, obtaining weights, and recommending combinations during device adjustments. In addition, the server continuously updates chip preference weights through periodic offline tasks of learning chips.

[0194] When a player configures their first piece of equipment during the pre-match equipment adjustment phase, the client detects this change event, generates a trigger request, and sends it to the server. The server responds to the trigger request, determines the game data, which includes at least the equipment attributes of the first piece of equipment for each virtual character participating in the match, and may also include game map information, match type information, and the character attributes of each virtual character. Subsequently, the server organizes the above game data into a compact data structure and calls the win rate prediction AI for prediction processing. For example... Figure 9 As shown, a pre-trained win / loss prediction model is first used, with game map information, match type information, character attributes, and equipment attributes jointly input into the encoding sub-model. The win / loss prediction model includes an encoding sub-model, an attention sub-model, and a prediction sub-model. The encoding sub-model uses methods such as one-hot encoding, vector embedding, and normalization to transform multi-source data into encoded information of a unified dimension, with each virtual character corresponding to one set of encoded information.

[0195] Next, the encoded information of all virtual characters is input into the attention sub-model. Taking players 1 to 6 as an example, the attention sub-model fuses information through a neural network to generate an information representation for each player and further analyzes three types of interactive influences: interference from other players (overall interference), teammate cooperation, and opponent interference. These interactive influences together constitute the influence information prediction result. Finally, the influence information prediction result is input into the prediction sub-model, which outputs the probability that each player can enter the top 3, i.e., the first prediction probability. For team matches, the server also calculates the second prediction probability of each game faction achieving the target match result based on the first prediction probability of each virtual character and the team match rules. For individual matches, each virtual character is considered an independent faction, and the first prediction probability is directly used as the winning probability.

[0196] After obtaining the predicted match result, the server generates match prompts. These prompts include: a comparison of the win rate change after replacing the current vehicle with a candidate vehicle (i.e., comparing the predicted probability of the first result corresponding to the first target equipment with the predicted probability of the second result corresponding to the candidate equipment), generating an upward or downward label, the overall win rate value of the current lineup, and lineup prompt text. The server then returns this information to the client, which updates the graphical user interface based on the prompts, displaying upward or downward indicators next to the candidate vehicles and showing the win rate value and suggested text in a designated area, thus providing a real-time, interactive win rate prediction assistant. Furthermore, in actual deployment, to support the real-time prediction needs of multiple players frequently switching vehicles during the pre-match equipment adjustment phase, the win rate prediction AI service module provides a batch prediction interface. This interface collects all equipment adjustment requests from all players in the same room, the character and equipment attributes of all players in the room, map information, match type, and other global match information all at once, packages them into a compact data structure, and batches them into the pre-trained win rate prediction model for calculation. By using batch prediction, the number of communications between the client and the server is effectively reduced, network overhead is lowered, the speed advantage of batch processing of neural network models is fully utilized, and inference efficiency is improved by using parallel computing.

[0197] Furthermore, in some embodiments, the preference weights of each car for chip attributes can be automatically learned from the behavior of top players. For example... Figure 10As shown, the process starts with the input vehicle list and scans the nth vehicle sequentially, determining whether all vehicles have been scanned. For each vehicle, the server identifies the top players using that vehicle, i.e., players who rank in the top percentage on that vehicle. Subsequently, the server reconstructs the chip combination replacement events that occurred on that vehicle based on the time sequence. For example, at time0, combination 0 changes to combination 1; at time1, combination 1 changes to combination 2; at time2, combination 2 changes to combination 3; at time3, combination 3 changes back to combination 2, and so on.

[0198] The server transforms each combination replacement event into a chip attribute vector change. First, each chip is converted into a vector containing 33-dimensional attributes. The attribute vectors of all chips in a combination are summed element-wise to obtain the overall attribute vector for that combination. In some embodiments, the number of dimensions in the chip attribute vector can be increased or decreased according to the actual game design; as long as the accessory combination score is calculated by the dot product of the attribute vector and the preference weights, it is considered an equivalent implementation. Then, the difference between the attribute vectors of the changed combination and the original combination is calculated to form the attribute change vector; for example, the attribute vector of combination 1 minus the attribute vector of combination 0. Simultaneously, based on the difference in the player's win rate or ranking on the vehicle before and after the change, each change is labeled as a positive or negative change. Next, the server uses the attribute change vector as input and the change label as output to train a binary classification or regression model, i.e., a weight prediction model. The model parameters reflect the contribution of different attribute changes to performance improvement, i.e., the vehicle's preference weights for chip attributes. After training, the model parameters are saved as the preference weights for that vehicle. The above steps are repeated until all vehicles have been scanned, ending the learning process. The server stores the latest preference weights for each vehicle in the database and updates them periodically.

[0199] Before a game begins, when a player performs a pre-triggered action such as clicking the "Recommend Chip" button on a second piece of in-game equipment, the client generates an equipment recommendation request and sends it to the server. The server responds to the request by first determining the vehicle's equipment attribute adjustment bias information, specifically by retrieving the vehicle's latest 33-dimensional attribute preference weights from the database. Then, the server queries all in-game virtual equipment (chips) currently held by the virtual character, considering chip usage constraints such as position, shape, and quantity, and selects multiple candidate equipment combinations that meet the criteria. If there is only one candidate combination, it is directly determined as the final target equipment combination. Otherwise, for each candidate combination, the attribute vectors of all chips included in that combination are summed to obtain a combination adjustment vector. This adjustment vector is then multiplied or weighted by the preference weight vector to obtain the equipment combination score. The server selects at least one candidate combination with the highest score as the target equipment combination. In some embodiments, the server can also call a win rate prediction AI to estimate the win rate corresponding to this combination, replacing or assisting the score calculation based on preference weights.

[0200] Finally, the server returns the target component combination and its corresponding score and / or win rate to the client. The client then receives the recommended solution and displays it through a graphical user interface.

[0201] like Figure 11 As shown, the graphical user interface presents a chip selection area, including my own selections and a recommended selection area. The recommended selection area, titled "Recommended by Experts" and "Personal Speedruns," displays specific chip combinations with win rates such as 77.25% and 22.25%. Each combination has a "Replace My" button next to it, along with helpful text encouraging players to easily copy expert picks. When a player clicks the "Replace My" button, the client instantly replaces the current chip combination with the recommended one and updates the local cache. This achieves personalized, low-operational-cost chip combination assistance based on learning from top player behavior.

[0202] In some embodiments, to ensure high concurrency and low latency service quality in actual deployment, a multi-layered server architecture can be adopted. For example, a gateway server is configured to receive client requests, perform authentication, request parsing, traffic distribution, and load balancing, forwarding legitimate requests to the corresponding game logic server. The game logic server is configured to maintain the current room's player lineup, equipment configuration, map, mode, and other status information, call the win rate prediction AI service and chip recommendation service, and cache some static data to reduce database access. An AI inference server cluster is configured to deploy the win rate prediction model and provide high-performance model inference services. This AI inference server cluster can scale horizontally according to the request volume and communicate with the game logic server through internal Remote Procedure Call (RPC) or message queues. A background data analysis and training cluster is configured to handle offline tasks, processing the periodic learning of chip preference weights, iterative training and updating of the win / loss prediction model, and cleaning and feature engineering of log data. The servers interact with each other through internal RPC or message queues. The overall topology can be elastically scaled according to business scale and operational needs, ensuring high availability and horizontal scalability of electronic devices.

[0203] In addition to the above embodiments, some embodiments can also push some of the logic generated by the chip solution down to the client side, and only distribute the preference weights on the server side. The client uses local algorithms such as greedy search and dynamic programming to generate recommendation schemes, so as to further reduce server load and reduce network transmission.

[0204] All of the above technical solutions can be combined in any way to form optional embodiments of this application, and will not be described in detail here.

[0205] To facilitate better implementation of the information processing method in games according to the embodiments of this application, the embodiments of this application also provide an information processing apparatus for games. Please refer to... Figure 12 , Figure 12 This is a schematic diagram of the structure of an information processing device in a game provided in an embodiment of this application. The information processing device 400 in the game is applied to a server, which is communicatively connected to a client. The device includes: a first request-response module 410, used to respond to a trigger request received before the start of a game match and determine game match data. The trigger request is generated by the client in response to a first equipment configuration operation and sent to the server. The first equipment configuration operation is used to configure the first game equipment of a virtual character. The game match data includes at least the equipment attributes of the first game equipment of each virtual character participating in the game match. The prediction module 420 is used to predict the outcome of the game based on the game data, and obtain the game outcome prediction. The generation module 430 is used to generate game prompt information based on the prediction of the game results. The game prompt information is used to indicate the impact of the first equipment configuration operation on the game results and / or to recommend the first game equipment used by the virtual character. The sending module 440 is used to send game prompts to the client.

[0206] In some embodiments, game match data may also include at least one of game map information, game match type information, and character attributes of each virtual character participating in the game match.

[0207] In some embodiments, the prediction of the game outcome includes a target prediction probability, which includes: a first prediction probability that each virtual character participating in the game will achieve a target ranking at the end of the game, and / or a second prediction probability that the game faction to which each virtual character belongs will achieve a target game outcome at the end of the game.

[0208] In some embodiments, the first equipment configuration operation is used to configure the first game equipment of a virtual character as the first target equipment. The game result prediction includes the target prediction probability of each virtual character when the first game equipment of the virtual character is the first target equipment and / or the target prediction probability of each virtual character when the first game equipment of the virtual character is a candidate equipment, wherein the candidate equipment is any equipment other than the first target equipment among all the equipment held by the virtual character.

[0209] In some embodiments, the generation module 430 can also be used to determine target data based on the target prediction probability of each virtual character, and generate game prompt information based on the target data, wherein the target data includes the result prediction probability of the target game faction to which the virtual character belongs to achieve the target game result at the end of the game, and / or the configuration guidance information of the virtual character's first game equipment.

[0210] In some embodiments, the outcome of a game match is determined by the ranking of each virtual character in each game faction, and the generation module 430 can also be used to determine the average of the first predicted probabilities of each virtual character in each game faction as the outcome predicted probability of each game faction.

[0211] In some embodiments, the target data includes a first result prediction probability when the first game equipment is the first target equipment, and a second result prediction probability when the first game equipment is a candidate equipment. The generation module 430 can also be used to determine the probability information corresponding to the first target equipment and the candidate equipment respectively based on the relationship between the first result prediction probability and the second result prediction probability, and generate lineup prompt text based on the first game equipment of each virtual character in the target game faction. Finally, the probability information, the lineup prompt text and the first result prediction probability are determined as the game prompt information. The probability information is used to make the client display the corresponding logo pattern, and the lineup prompt text is used to prompt the first game equipment that the virtual character can use.

[0212] In some embodiments, the prediction module 420 can also be used to input game data into a pre-trained win / loss prediction model, so that the win / loss prediction model can perform prediction processing on the game results based on the game data to obtain the game result prediction.

[0213] In some embodiments, the game match data further includes game map information, game match type information, and the character attributes of each virtual character participating in the game match; the prediction module 420 can also be used to input the game map information, game match type information, character attributes, and equipment attributes of the virtual characters into the encoding sub-model of the win / loss prediction model, so that the encoding sub-model encodes the received game map information, game match type information, character attributes, and equipment attributes to obtain the encoding information corresponding to the virtual characters, inputs the encoding information corresponding to each virtual character into the attention sub-model of the win / loss prediction model, so that the attention sub-model predicts the influence between virtual characters in the game match and obtains the influence information prediction result, and inputs the influence information prediction result into the prediction sub-model of the win / loss prediction model, so that the prediction sub-model performs prediction processing on the game match result based on the influence information prediction result and obtains the game result prediction.

[0214] In some embodiments, the first game equipment includes multiple components, each of which can be used with game virtual accessories. Each first game equipment has multiple sub-attributes as its equipment attributes. Each game virtual accessory is used to adjust at least one sub-attribute of the first game equipment. The first request-response module 410 can also be used to respond to a client's accessory recommendation request for a second game equipment received before the start of a game match, and to determine a target accessory combination corresponding to the second game equipment. The second game equipment is one of multiple first game equipment components. The accessory recommendation request is generated by the client when it detects a preset trigger operation for the second game equipment and sent to the server. The sending module 440 can also be used to provide feedback on the target accessory combination to the client.

[0215] In some embodiments, the first request response module 410 can also be used to respond to accessory recommendation requests, determine the equipment attribute adjustment bias information of the second game equipment, and select game virtual accessory combinations that meet the accessory usage constraint information of the second game equipment from the game virtual accessories held by the virtual character, to obtain multiple candidate accessory combinations. Finally, based on the equipment attribute adjustment bias information and the sub-attribute adjustment amount of each game virtual accessory in the candidate accessory combinations, the accessory combination score of the candidate accessory combinations is determined, and at least one candidate accessory combination with the highest accessory combination score is determined as the target accessory combination. The equipment attribute adjustment bias information is used to determine the adjustment bias of each sub-attribute of the second game equipment, the accessory usage constraint information is used to determine the game virtual accessories that can be used with the second game equipment, and the target accessory combination includes at least one game virtual accessory.

[0216] In some embodiments, the sending module 440 can also be used to send accessory combination scores to the client so that when the client recommends the second game equipment to be used with the target accessory combination, the accessory combination scores of the target accessory combination are displayed synchronously.

[0217] In some embodiments, the first request response module 410 may also be used to determine the selected game virtual accessory combination as the target accessory combination if the number of selected game virtual accessory combinations is one.

[0218] In some embodiments, the first request response module 410 can also be used to select game virtual accessory combinations that meet the accessory usage constraint information of the second game equipment from the game virtual accessories held by the virtual character based on a predetermined accessory selection strategy, thereby obtaining multiple candidate accessory combinations.

[0219] In some embodiments, the equipment attribute adjustment bias information includes the bias weight of each sub-attribute of the second game equipment. The first request response module 410 can also be used to sum the sub-attribute adjustment vectors of each game virtual accessory in the candidate accessory combination to obtain the adjustment vector corresponding to the candidate accessory combination, and to perform vector calculation on the bias weights and the adjustment vectors corresponding to the candidate accessory combination to obtain the accessory combination score of the candidate accessory combination. The sub-attribute adjustment vector is used to indicate the adjustment amount of the game virtual accessory for each sub-attribute.

[0220] In some embodiments, the prediction module 420 can also be used to acquire first game performance data, second game performance data, first accessory combination, and second accessory combination for each first game equipment, then sum the sub-attribute adjustment vectors of each virtual game accessory in the first accessory combination to obtain the first adjustment vector corresponding to the first accessory combination, and sum the sub-attribute adjustment vectors of each virtual game accessory in the second accessory combination to obtain the second adjustment vector corresponding to the second accessory combination. Finally, the first game performance data, second game performance data, first adjustment vector, and second adjustment vector of each first game equipment are input into the pre-trained weight prediction module. In the model, the weight prediction model predicts the bias weight of each sub-attribute of each first game equipment based on the first game performance data, the second game performance data, the first adjustment vector, and the second adjustment vector of each first game equipment, thereby obtaining the bias weight of each sub-attribute of each first game equipment. The first game performance data is used to indicate the performance of the first game equipment in the game when the accessory combination used with the first game equipment is the first accessory combination, and the second game performance data is used to indicate the performance of the first game equipment in the game after the accessory combination used with the first game equipment is changed from the first accessory combination to the second accessory combination.

[0221] In some embodiments, the prediction module 420 can also be used to obtain the first game performance data, the second game performance data, the first accessory combination, and the second accessory combination of the target virtual character using the first game equipment, wherein the target virtual character is a preset number of virtual characters ranked at the top in the character ranking data corresponding to the first game equipment.

[0222] In some embodiments, the first request response module 410 can also be used to determine the target win rate when the second game equipment and the target accessory are used together, and send the target win rate to the client so that the client can synchronously display the target win rate when recommending the use of the second game equipment and the target accessory together.

[0223] To facilitate better implementation of the information processing method in games according to the embodiments of this application, the embodiments of this application also provide an information processing apparatus for games. Please refer to... Figure 13 , Figure 13This is a schematic diagram of the structure of the information processing device in the game provided in this application embodiment. The information processing device 500 in the game is applied to the client, which is connected to the server. The client provides a graphical user interface, which can display the game screen after the start of the game. The game screen includes a virtual character controlled by the client. The device includes: an operation response module 510, which is used to respond to the first equipment configuration operation before the start of the game, generate a trigger request and send the trigger request to the server so that the server responds to the received trigger request, determine the game data, and perform prediction processing on the game result based on the game data to obtain the game result prediction, and generate game prompt information based on the game result prediction, and send the game prompt information to the client. The first equipment configuration operation is used to configure the first game equipment of the virtual character. The game data includes at least the equipment attributes of the first game equipment of each virtual character participating in the game. The game prompt information is used to indicate the impact of the first equipment configuration operation on the game result and / or to recommend the first game equipment used by the virtual character. The information response module 520 is used to respond to game prompts sent by the receiving server and update the current display content of the graphical user interface according to the game prompts.

[0224] In some embodiments, the first game equipment includes multiple components, each of which can be used with game virtual accessories. Each first game equipment has multiple sub-attributes as its equipment attributes. Each game virtual accessory is used to adjust at least one sub-attribute of the first game equipment. The operation response module 510 can also be used to respond to a preset trigger operation for the second game equipment, generate an accessory recommendation request for the second game equipment, and send the accessory recommendation request to the server. This allows the server to respond to the received accessory recommendation request, determine the target accessory combination corresponding to the second game equipment, and provide feedback on the target accessory combination to the client. The second game equipment is one of multiple first game equipment components. The information response module 520 can also be used to respond to receiving the target accessory combination sent by the server and recommend the second game equipment to be used with the target accessory combination through a graphical user interface.

[0225] Each unit in the information processing device 400 or 500 of the aforementioned game can be implemented entirely or partially through software, hardware, or a combination thereof. These units can be embedded in or independent of the processor in the electronic device in hardware form, or stored in the memory of the electronic device in software form, so that the processor can call and execute the operations corresponding to each unit. The information processing device 400 or 500 in the game can be integrated into a terminal or server with storage and a processor, thus possessing computing capabilities; alternatively, the information processing device 400 or 500 in the game can serve as such a terminal or server.

[0226] Optionally, this application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments. Figure 14 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may be a terminal or a server. Figure 14 As shown, the electronic device 600 includes a processor 601 with one or more processing cores, a memory 602 with one or more computer-readable storage media, and a computer program stored in the memory 602 and executable on the processor. The processor 601 and the memory 602 are electrically connected. Those skilled in the art will understand that the electronic device structure shown in the figures does not constitute a limitation on the electronic device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0227] The processor 601 is the control center of the electronic device 600. It connects various parts of the electronic device 600 through various interfaces and lines. By running or loading software programs and / or modules stored in the memory 602, and calling data stored in the memory 602, it executes various functions of the electronic device 600 and processes data, thereby performing overall processing of the electronic device 600.

[0228] In this embodiment of the application, the processor 601 in the electronic device 600 loads the instructions corresponding to the process of one or more computer programs into the memory 602 according to the following steps, and the processor 601 runs the computer programs stored in the memory 602 to realize various functions: responding to the trigger request received before the start of the game, determining the game data, wherein the trigger request is generated by the client in response to the first equipment configuration operation and sent to the server, the first equipment configuration operation is used to configure the first game equipment of the virtual character, and the game data includes at least the equipment attributes of the first game equipment of each virtual character participating in the game; Based on game data, predict the outcome of the game to obtain the predicted outcome. Based on the prediction of the game results, game prompt information is generated. The game prompt information is used to indicate the impact of the first equipment configuration operation on the game results, and / or to recommend the first game equipment to be used by the virtual character. Send match notification information to the client.

[0229] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0230] Optional, such as Figure 14As shown, the electronic device 600 also includes: a display screen 603, a radio frequency circuit 604, an audio circuit 605, an input unit 606, and a power supply 607. The processor 601 is electrically connected to the display screen 603, the radio frequency circuit 604, the audio circuit 605, the input unit 606, and the power supply 607. Those skilled in the art will understand that... Figure 14 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0231] The display screen 603 can be used to display a graphical user interface (GUI) and receive operation commands generated by the user interacting with the GUI. The display screen 603 may include a display panel and a touch panel. The display panel can be used to display information input by the user or information provided to the user, as well as various graphical user interfaces of the electronic device. These graphical user interfaces can be composed of graphics, text, icons, video, and any combination thereof. The touch panel can be used to collect touch operations performed by the user on or near it (such as operations performed by the user using a finger, stylus, or any suitable object or accessory on or near the touch panel), generate corresponding operation commands, and execute the corresponding program. Optionally, the touch panel may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch location and the signal generated by the touch operation, and transmits the signal to the touch controller. The touch controller receives touch information from the touch detection device, converts it into touch point coordinates, sends it to the processor 601, and can receive and execute commands from the processor 601. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it transmits the information to the processor 601 to determine the type of touch event. Subsequently, the processor 601 provides corresponding visual output on the display panel according to the type of touch event. In this embodiment, the touch panel and the display panel can be integrated into the display screen 603 to achieve input and output functions. However, in some embodiments, the touch panel and the display screen 603 can be implemented as two independent components to achieve input and output functions. That is, the display screen 603 can also be used as part of the input unit 606 to achieve input functions.

[0232] The radio frequency (RF) circuit 604 can be used to transmit and receive RF signals to establish wireless communication with network devices or other electronic devices, and to send and receive signals with network devices or other electronic devices. The audio circuit 605 can provide an audio interface between the user and electronic devices via a speaker or microphone. The audio circuit 605 can convert received audio data into electrical signals and transmit them to the speaker, where the speaker converts them into sound signals for output. Conversely, the microphone converts collected sound signals into electrical signals, which are received by the audio circuit 605, converted back into audio data, and then processed by the processor 601 before being transmitted via the RF circuit 604 to, for example, another electronic device, or output to the memory 602 for further processing. The audio circuit 605 may also include an earphone jack to provide communication between peripheral headphones and electronic devices.

[0233] Input unit 606 can be used to receive input numerical or character information or object feature information (such as fingerprints, iris scans, facial information, etc.), and generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control. Power supply 607 is used to power the various components of electronic device 600. Optionally, power supply 607 can be logically connected to processor 601 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. Power supply 607 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. Although Figure 14 As not shown in the diagram, the electronic device 600 may also include a camera, sensor, wireless fidelity module, Bluetooth module, etc., which will not be described in detail here.

[0234] This application also provides a computer-readable storage medium for storing a computer program. This computer-readable storage medium can be applied to an electronic device, and the computer program causes the electronic device to execute the corresponding flow in the information processing method of the game in the embodiments of this application; for the sake of brevity, further details are omitted here.

[0235] This application also provides a computer program product including computer instructions stored in a computer-readable storage medium. The processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the corresponding flow in the information processing method of the game in the embodiments of this application. For the sake of brevity, further details are omitted here.

[0236] This application also provides a computer program comprising computer instructions stored in a computer-readable storage medium. The processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform the corresponding flow in the information processing method of the game described in this application; for brevity, this will not be elaborated further.

[0237] It should be understood that the processor in this application may be an integrated circuit chip with signal processing capabilities. In implementation, the steps of the above method embodiments can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor described above can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0238] It is understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DR RAM). It should be noted that the memory used in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0239] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0240] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0241] In the embodiments of this application, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0242] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0243] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, the functional units in this application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0244] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the 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 to cause an electronic device (which may be a personal computer or a server) to execute all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0245] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An information processing method in a game, characterized in that, Applied to the server side, where the server communicates with the client, the method includes: In response to a trigger request received before the start of a game match, game match data is determined, wherein the trigger request is generated by the client in response to a first equipment configuration operation and sent to the server, the first equipment configuration operation is used to configure the first game equipment of the virtual character, and the game match data includes at least the equipment attributes of the first game equipment of each virtual character participating in the game match; Based on the game data, the game results are predicted to obtain the predicted game results. Based on the predicted game results, game prompt information is generated, wherein the game prompt information is used to indicate the impact of the first equipment configuration operation on the game results, and / or to recommend the first game equipment used by the virtual character; The game notification information is sent to the client.

2. The method according to claim 1, characterized in that, The prediction of the game result includes a target prediction probability, which includes: a first prediction probability that each virtual character participating in the game will achieve a target ranking at the end of the game, and / or a second prediction probability that the game faction to which each virtual character belongs will achieve a target game result at the end of the game.

3. The method according to claim 2, characterized in that, The first equipment configuration operation is used to configure the first game equipment of the virtual character as the first target equipment. The game result prediction includes the target prediction probability of each virtual character when the first game equipment of the virtual character is the first target equipment and / or the target prediction probability of each virtual character when the first game equipment of the virtual character is a candidate equipment. The candidate equipment is any equipment other than the first target equipment among all the equipment held by the virtual character.

4. The method according to claim 3, characterized in that, The step of predicting and generating game prompt information based on the game results includes: Based on the target prediction probability of each virtual character, target data is determined, wherein the target data includes the result prediction probability of the target game faction to which the virtual character belongs obtaining the target game result at the end of the game, and / or the configuration guidance information of the first game equipment of the virtual character; Based on the target data, the game prompt information is generated.

5. The method according to claim 1, characterized in that, The step of predicting the game result based on the game data to obtain a game result prediction includes: The game data is input into a pre-trained win / loss prediction model, so that the win / loss prediction model performs the prediction process on the game results based on the game data, and obtains the game result prediction.

6. The method according to claim 5, characterized in that, The game match data also includes game map information, game match type information, and the character attributes of each virtual character participating in the game match. The step of inputting the game data into a pre-trained win / loss prediction model, so that the win / loss prediction model performs the prediction process on the game results based on the game data, to obtain a game result prediction, includes: The game map information, the game match type information, the character attributes of the virtual character, and the equipment attributes are all input into the encoding sub-model of the win / loss prediction model, so that the encoding sub-model encodes the received game map information, game match type information, character attributes, and equipment attributes to obtain the encoding information corresponding to the virtual character; The encoded information corresponding to each virtual character is input into the attention sub-model of the win / loss prediction model, so that the attention sub-model predicts the influence between the virtual characters in the game and obtains the influence information prediction result. The prediction results of the influence information are input into the prediction sub-model of the win / loss prediction model, so that the prediction sub-model performs the prediction processing on the game result of the game based on the prediction results of the influence information, and obtains the game result prediction.

7. The method according to claim 1, characterized in that, The first game equipment includes multiple components, each of which can be used in conjunction with a virtual game accessory. Each first game equipment has multiple sub-attributes as its equipment attributes. Each virtual game accessory is used to adjust at least one of the sub-attributes of the first game equipment. The method further includes: In response to receiving a component recommendation request from the client for a second game equipment before the start of a game match, the system determines a target component combination corresponding to the second game equipment, wherein the second game equipment is one of multiple first game equipment. The component recommendation request is generated by the client when it detects a preset trigger operation for the second game equipment and is sent to the server. The target accessory combination is fed back to the client.

8. The method according to claim 7, characterized in that, The response to receiving the client's accessory recommendation request for the second game equipment before the start of the game, and determining the target accessory combination corresponding to the second game equipment, includes: In response to the accessory recommendation request, determine the equipment attribute adjustment bias information of the second game equipment, wherein the equipment attribute adjustment bias information is used to determine the adjustment bias of each of the sub-attributes of the second game equipment; From the game virtual accessories held by the virtual character, select game virtual accessory combinations that satisfy the accessory usage constraint information of the second game equipment to obtain multiple candidate accessory combinations. The accessory usage constraint information is used to determine the game virtual accessories that can be used with the second game equipment. The target accessory combination includes at least one game virtual accessory. Based on the equipment attribute adjustment bias information and the sub-attribute adjustment amount of each virtual accessory in the candidate accessory combination, the accessory combination score of the candidate accessory combination is determined; The candidate accessory combination with the highest accessory combination score is determined as the target accessory combination.

9. The method according to claim 8, characterized in that, The method further includes: The accessory combination score is sent to the client so that when the client recommends using the second game equipment with the target accessory combination, the accessory combination score of the target accessory combination is displayed simultaneously.

10. The method according to claim 8, characterized in that, The equipment attribute adjustment bias information includes the bias weight of each sub-attribute of the second game equipment. The step of determining the accessory combination score of the candidate accessory combination based on the equipment attribute adjustment bias information and the sub-attribute adjustment amount of each virtual game accessory in the candidate accessory combination includes: The adjustment vectors of the sub-attributes of each virtual game accessory in the candidate accessory combination are summed to obtain the adjustment vector corresponding to the candidate accessory combination, wherein the sub-attribute adjustment vector is used to indicate the adjustment amount of the virtual game accessory for each sub-attribute; Vector calculation is performed on the bias weight and the adjustment vector corresponding to the candidate component combination to obtain the component combination score of the candidate component combination.

11. The method according to claim 10, characterized in that, The method further includes: Acquire first game performance data, second game performance data, first accessory combination and second accessory combination for each first game equipment, wherein the first game performance data is used to indicate the performance of the first game equipment in the game when the accessory combination used with the first game equipment is the first accessory combination, and the second game performance data is used to indicate the performance of the first game equipment in the game after the accessory combination used with the first game equipment is changed from the first accessory combination to the second accessory combination. The sub-attribute adjustment vectors of each virtual game accessory in the first accessory combination are summed to obtain the first adjustment vector corresponding to the first accessory combination. The sub-attribute adjustment vectors of each virtual game accessory in the second accessory combination are summed to obtain the second adjustment vector corresponding to the second accessory combination; The first game performance data, the second game performance data, the first adjustment vector, and the second adjustment vector of each first game equipment are input into a pre-trained weight prediction model, so that the weight prediction model predicts the bias weight of each sub-attribute of each first game equipment based on the first game performance data, the second game performance data, the first adjustment vector, and the second adjustment vector of each first game equipment, thereby obtaining the bias weight of each sub-attribute of each first game equipment.

12. The method according to claim 7, characterized in that, The method further includes: The target win rate is determined when the second game equipment is used in combination with the target accessory, and the target win rate is sent to the client so that the client can recommend the use of the second game equipment in combination with the target accessory and simultaneously display the target win rate.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted for loading by a processor to execute the information processing method in a game according to any one of claims 1-12.

14. An electronic device, characterized in that, The electronic device includes a processor and a memory, the memory storing a computer program, and the processor executing the information processing method in the game according to any one of claims 1-12 by calling the computer program stored in the memory.

Images