Methods, devices, storage media, and computer equipment for evaluating the effectiveness of virtual items

Abstract

The virtual item effectiveness evaluation method, apparatus, storage medium, and computer equipment provided in this application, by responding to the activation command of the virtual item, acquire item identification information and current game scene data, and continuously acquire player operation data. By comprehensively considering the item identification information, current game scene data, player operation data, and a preset reference dataset, the predicted effectiveness result of the virtual item is dynamically evaluated. Compared to traditional single-dimensional judgment logic, this application can fully consider the characteristics of the virtual item itself, subtle differences in player operation behavior, and the specific needs of the current game scene, making the item effectiveness result more personalized and scene-adaptable.

Description

Technical Field

[0001] This application relates to the field of virtual interaction technology, and in particular to a method, apparatus, storage medium, and computer equipment for evaluating the effectiveness of virtual props. Background Technology

[0002] In simulation games, players often acquire various items through purchases, quests, and social gifts. As they progress through the game, they use these items in appropriate scenarios (battle scenes or quests corresponding to the items) to activate them, thereby obtaining corresponding game resources and creating engaging interactions. Since the backend system's calibration and judgment of player actions can influence the item's effectiveness and thus the game experience, managing the correlation between player actions and item effects is often crucial in games.

[0003] In existing games, the logic for determining the activation conditions of items is relatively rigid, usually relying solely on simple scene parameters or single-dimensional feature comparisons. This fails to adequately consider the player's individual circumstances, dynamic behavioral data during gameplay, and strategic choices. For example, in the scenario of using painting items, existing solutions only determine whether the player's drawing matches a preset image, without considering dynamic behavioral characteristics such as the player's drawing speed and stress, or differentiating the player's specific scenario (such as the different requirements of a silk shop design scenario versus a building construction scenario). This results in a lack of personalization and depth in the item's activation effects, failing to provide a better gaming experience for players. Summary of the Invention

[0004] The purpose of this application is to at least address one of the aforementioned technical deficiencies, particularly the fact that the existing technology uses a single dimension to determine the conditions for item activation, failing to fully consider the player's own situation, dynamic behavioral data during the game, and strategies, resulting in a lack of personalization and depth in the effects of items and an inability to provide players with a better gaming experience.

[0005] This application provides a method for evaluating the effectiveness of virtual items, the method comprising:

[0006] In response to the activation command of the virtual item, the system obtains the item identification information of the virtual item and the current game scene data, and continuously obtains player operation data;

[0007] The predicted effectiveness of the virtual item is dynamically evaluated based on the item identification information, the current game scene data, the player operation data, and the preset reference dataset.

[0008] The activation operation of the virtual item will be performed based on the predicted activation result.

[0009] Optionally, the step of dynamically evaluating the predicted effectiveness of the virtual item based on the item identification information, the current game scene data, the player operation data, and a preset reference dataset includes:

[0010] Invoke the multi-dimensional effective condition determination model;

[0011] The item identification information, the current game scene data, the player operation data, and the preset reference dataset are input into the multi-dimensional effectiveness condition determination model, and the predicted effectiveness result of the virtual item is generated through the multi-dimensional effectiveness condition determination model.

[0012] Optionally, the multi-dimensional effective condition determination model includes a basic scene matching layer, a player behavior analysis layer, an item characteristic adaptation layer, and a real-time interaction state layer;

[0013] The step of generating the predicted effectiveness result of the virtual item through the multi-dimensional effectiveness condition determination model includes:

[0014] The scene matching degree parameter is determined based on the basic scene matching layer and the current game scene data;

[0015] The operation standardization parameter is determined based on the player behavior analysis layer and the player operation data;

[0016] The prop ability parameters are determined based on the prop characteristic adaptation layer and the prop identification information;

[0017] The interaction adaptation parameters are determined based on the real-time interaction state layer and the current game scene data.

[0018] The predicted effectiveness result of the virtual prop is generated by combining at least two of the scene matching degree parameter, the operation standardization degree parameter, the prop ability parameter, and the interaction adaptation parameter.

[0019] Optionally, the current game scene data includes at least one of the following: the map environment where the player character is located, the type of task being performed, and the current team's character composition;

[0020] The step of determining the scene matching degree parameter based on the basic scene matching layer and the current game scene data includes:

[0021] The matching degree of the usage scenario of the virtual props is determined based on the map environment;

[0022] Determine the task type matching degree of the virtual item based on the task type;

[0023] The team configuration matching degree of the virtual props is determined based on the current team's role composition;

[0024] The usage scenario matching degree, the task type matching degree, and / or the team configuration matching degree are used as the scenario matching degree parameter.

[0025] Optionally, the player operation data includes at least one of the player's operation trajectory, operation speed, and operation pressure during the process of activating the item;

[0026] The step of determining the operation standardization parameter based on the player behavior analysis layer and the player operation data includes:

[0027] The trajectory matching parameters are determined based on the degree of matching between the operation trajectory and the preset reference dataset;

[0028] Determine the speed consistency parameter based on the stability of the operating speed;

[0029] The pressure uniformity parameter is determined based on the range of variation of the operating pressure;

[0030] The trajectory matching parameter, the speed consistency parameter, and / or the pressure uniformity parameter are used as the operation standardization parameter.

[0031] Optionally, the item identification information includes at least one of virtual item type and virtual item level;

[0032] The step of determining the item ability parameters based on the item characteristic adaptation layer and the item identification information includes:

[0033] Determine the basic functional parameters based on the type of virtual item;

[0034] The function unlock status parameters are determined based on the virtual item level;

[0035] The basic function parameters and / or the function unlock status parameters are used as the item ability parameters.

[0036] Optionally, determining the interaction adaptation parameters based on the real-time interaction state layer and the current game scene data includes:

[0037] When the current game scene data indicates team mode, the enhancement parameters of the item effect are determined based on the role classes of other members of the current team.

[0038] And / or, determine the conflict detection parameters for item effects based on the active items of other members of the current team.

[0039] Optionally, the method further includes:

[0040] Obtain the functional structural standards corresponding to the current game scene. The functional structural standards are used to define the structural elements that player input must satisfy.

[0041] Verify whether the player operation data meets the functional structural standards;

[0042] If the conditions are not met, the virtual item will be deemed invalid, and no evaluation of the predicted effectiveness will be performed.

[0043] Optionally, the process of determining the functional structural criteria includes:

[0044] When the current game scene is a clothing design scene, the functional structural standard includes at least the neckline structural element and the cuff structural element;

[0045] When the current game scene is a building construction scene, the functional structural standard includes at least door structural elements and window structural elements.

[0046] Optionally, the method further includes:

[0047] When the virtual props take effect, a working artwork is generated;

[0048] In response to the sharing instruction for the effective work, the effective work is published to the shared work pool;

[0049] The ranking information of the effective works is determined based on the interaction data of the effective works in the shared works pool;

[0050] Based on the ranking information, additional economic benefits will be provided to the players corresponding to the effective works.

[0051] Optionally, the step of performing the activation operation of the virtual item based on the prediction activation result includes:

[0052] When the probability of effectiveness in the predicted effectiveness result reaches or exceeds a preset threshold, the virtual item is determined to be effective, and an effectiveness effect is generated based on the expected effect parameters in the predicted effectiveness result.

[0053] If the probability of effectiveness in the predicted effectiveness result does not reach the preset threshold, it is determined that the virtual item has not taken effect, and a failure feedback message is generated.

[0054] Optionally, the preset threshold is dynamically adjusted based on the rarity of the virtual item and / or game balance requirements.

[0055] Optionally, the step of performing the activation operation of the virtual item based on the prediction activation result further includes:

[0056] Generate multimodal effect feedback information and present it to the player;

[0057] The multimodal effect feedback information includes text prompts, dynamic visual effects, and / or 3D preview effects.

[0058] Optionally, the multimodal effect feedback information further includes:

[0059] When the effect is determined to be effective, the dynamic process of the effect taking effect, the 3D preview effect of the effective work, and the estimated value information are displayed.

[0060] If the condition is determined to be ineffective, the specific reasons for the ineffectiveness and suggestions for improvement will be displayed.

[0061] Optionally, the method further includes:

[0062] Record complete data of this virtual item usage, including the activation command, the current game scene data, the player operation data, the predicted effect result, and the result of the effective operation;

[0063] The complete data is stored in the player behavior database.

[0064] Optionally, when the process of dynamically evaluating the predicted effectiveness of the virtual item calls a multi-dimensional effectiveness condition determination model for evaluation, the method further includes:

[0065] Based on the complete data of multiple players stored in the player behavior database, the parameters of the multi-dimensional effective condition determination model are optimized by machine learning algorithms. The parameters include the weights and evaluation thresholds of each dimension of data.

[0066] This application also provides a device for evaluating the effectiveness of virtual items, including:

[0067] The data acquisition module is used to respond to the activation command of the virtual item, acquire the item identification information of the virtual item and the current game scene data, and continuously acquire player operation data;

[0068] The effectiveness evaluation module is used to dynamically evaluate the predicted effectiveness result of the virtual item based on the item identification information, the current game scene data, the player operation data, and a preset reference dataset.

[0069] The item activation module is used to execute the activation operation of the virtual item based on the predicted activation result.

[0070] This application also provides a computer-readable storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the virtual item effectiveness evaluation method as described in any of the above embodiments.

[0071] This application also provides a computer device, including: one or more processors, and memory;

[0072] The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the virtual item effectiveness evaluation method as described in any of the above embodiments.

[0073] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0074] The virtual item activation evaluation method, device, storage medium, and computer equipment provided in this application, by responding to the activation command of the virtual item, acquire item identification information and current game scene data, and continuously acquire player operation data. By comprehensively considering the item identification information, current game scene data, player operation data, and a preset reference dataset, the predicted activation result of the virtual item is dynamically evaluated. Compared to traditional single-dimensional judgment logic, this application can fully consider the characteristics of the virtual item itself, subtle differences in player operation behavior, and the specific needs of the current game scene, making the item activation result more personalized and scene-adaptable. For example, in a clothing design scenario, this application can not only determine whether the collar and cuff structures drawn by the player meet functional structural standards, but also combine behavioral data such as the smoothness and speed stability of the player's operation trajectory, as well as the specific requirements of the current task for clothing style (such as ancient style or modern style), to conduct a comprehensive evaluation, making the determination of item activation more comprehensive and accurate, and better reflecting the player's operational intentions and strategic choices. Furthermore, this application acquires current game scenario data (such as the different requirements of silk shop design scenarios and building construction scenarios) and uses it as an important dimension for evaluation, enabling the item effectiveness determination to produce differentiated results based on different scenarios. For example, in a clothing design scenario, the system focuses on evaluating whether the drawn graphic includes clothing structural elements such as collars and cuffs; in a building construction scenario, the system focuses on evaluating whether it includes architectural structural elements such as doors and windows. This scenario-differentiated determination mechanism makes item use more closely aligned with the game context, enhancing the strategic depth and realism of item use. In addition, the item identification information acquired in this application can include item type and item level, allowing the system to determine the function unlock status based on the item level, thereby dynamically invoking the corresponding determination conditions during the evaluation process. For example, when the item is in the initial stage, only basic graphic matching determination is enabled; when the item has been upgraded and unlocked advanced functions, corresponding preference data or advanced determination conditions can be additionally invoked. This dynamic adaptation mechanism allows the item's growth and functional expansion to be fully reflected in the effectiveness evaluation, thereby enriching the depth of item use. Attached Figure Description

[0075] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the 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.

[0076] Figure 1 This is a schematic diagram of the application architecture provided for an embodiment of this application;

[0077] Figure 2 A flowchart illustrating the method for evaluating the effectiveness of virtual items provided in this application embodiment;

[0078] Figure 3 The image shows the interface for players to receive tasks and activate items in an open-world simulation game, as provided in this application embodiment.

[0079] Figure 4 A model architecture diagram of the multi-dimensional effectiveness condition determination model provided in the embodiments of this application;

[0080] Figure 5 An interface illustration showing the unlockable function of the drawing props provided in this application embodiment;

[0081] Figure 6 A flowchart illustrating the process for verifying whether player operation data meets functional structural standards, provided in an embodiment of this application.

[0082] Figure 7 A display diagram of the clothing drawing ranking interface after clothing is drawn using drawing tools, provided as an embodiment of this application;

[0083] Figure 8 This is a schematic diagram of the dynamic visual effects display interface provided in the embodiments of this application;

[0084] Figure 9 This is a preview image of a 3D artwork provided in an embodiment of this application.

[0085] Figure 10 This application provides a diagram illustrating a game interface with a missing collar indicator as part of an embodiment of the invention.

[0086] Figure 11 A schematic diagram of the structure of a virtual prop effectiveness evaluation device provided in an embodiment of this application;

[0087] Figure 12 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0088] 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.

[0089] In simulation games, players often acquire various items through purchases, quests, and social gifts. As they progress through the game, they use these items in appropriate scenarios (battle scenes or quests corresponding to the items) to activate them, thereby obtaining corresponding game resources and creating engaging interactions. Since the backend system's calibration and judgment of player actions can influence the item's effectiveness and thus the game experience, managing the correlation between player actions and item effects is often crucial in games.

[0090] However, the existing logic for determining the effectiveness of items has significant flaws. Most rely solely on fixed, pre-set conditions for comparison, either verifying whether the scene meets the requirements or comparing the generated result to a pre-set template. If any one condition is not met, the item is deemed invalid. This rigid logic completely ignores the dynamic characteristics of player actions, fails to adapt to the differentiated needs of different scenarios, and fails to dynamically adjust based on the item's level and characteristics. This easily leads to situations where a player's correct strategy fails due to minor deviations, severely impacting player creativity and engagement.

[0091] For example, in the clothing design gameplay of open-world simulation games, when players use drawing tools to customize clothing designs, existing solutions only require that the outline of the player's drawing matches the preset draft to a fixed ratio. They completely disregard whether the clothing style required for the task is casual or formal, nor do they consider the smoothness and naturalness of the player's drawing process. Furthermore, they don't adjust the judgment criteria based on the level of the player's drawing tools—advanced drawing tools could clearly support more refined creations, yet the judgment logic is exactly the same as for basic tools. The final result neither meets the player's expectations nor reflects the value differences between different levels of tools, thus failing to provide a high-quality gaming experience. To address these problems in existing technologies, this application proposes a multi-dimensional dynamic effect evaluation scheme that can effectively improve the accuracy and adaptability of tool effect judgment, providing players with a more personalized and strategic gaming experience.

[0092] It should be noted that the virtual items in this application can be used in various types of game scenarios, such as open-ended creative scenarios: clothing design, building construction, character graffiti, weapon customization, etc., and can also be used in combat scenarios, such as skill enhancement items, status recovery items, special summoning items, etc. This application does not specifically limit their application scenarios. Players can obtain them through various means such as purchasing from the store, completing tasks, event rewards, and social interaction. Different types and rarities of virtual items can provide players with different functional benefits, helping players complete specific operations and obtain corresponding game benefits in the game.

[0093] Furthermore, before describing the specific implementation process of this application, the application environment of this application will first be described. Please refer to [link / reference needed]. Figure 1 , Figure 1 This is a schematic diagram of the application architecture provided for an embodiment of this application; Figure 1 The application architecture includes server 110 and client 120. Server 110 can be of various types, such as a game server or application server. It is used to store game data, process player requests, and execute the core logic of evaluating the effectiveness of virtual items. Server 110 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing 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, CDN, and big data and artificial intelligence platforms. Client 120 can be a terminal device such as a smartphone, tablet, personal computer, or smartwatch. Players interact with the game system through the graphical user interface of client 120, such as triggering activation commands for virtual items and viewing the effectiveness results of virtual items. Server 110 and client 120 establish a communication connection through a network to achieve real-time data transmission and synchronization. For example, client 120 sends an item effectiveness evaluation request to server 110, and server 110 returns the effectiveness evaluation result.

[0094] It is understood that the above-mentioned method for evaluating the effectiveness of virtual items can run on personal mobile terminals, on server 110, or on third-party devices to provide evaluation of the effectiveness of virtual items. The specific method for evaluating the effectiveness of virtual items can be a program running on the above-mentioned devices, or a system component running on the above-mentioned devices, or a cloud service program. The specific operating mode depends on the actual scenario and is not limited here.

[0095] The specific implementation methods of this application will be described in detail below. For ease of understanding, the effectiveness evaluation method for virtual items proposed in this application will be explained in detail below with reference to the accompanying drawings. Please refer to the drawings. Figure 2 , Figure 2 The flowchart illustrating the method for evaluating the effectiveness of virtual items provided in this application embodiment can specifically include the following steps:

[0096] S110: In response to the activation command of virtual items, obtain the item identification information of virtual items and the current game scene data, and continuously obtain player operation data.

[0097] In this step, when a player selects a target virtual item in the game and clicks the "use" button to trigger the activation command, or when the activation command for a virtual item is automatically triggered when the player enters a battle or mission scene, the system can first read the unique identifier information corresponding to the virtual item. This information includes the specific type of the virtual item, its current level, unlocked functional permissions, and basic activation judgment requirements. For example, for painting creation tools, a level 1 basic painting tool only supports basic outline judgment, while a level 3 master painting tool unlocks additional judgment dimensions such as style matching and detail scoring. Different identifier information will directly correspond to different subsequent evaluation logic.

[0098] Simultaneously, the system can also synchronously access contextual data of the current game scene, including environmental parameters such as the player character's map environment, the type of task being performed, and the composition of the current team (if it is a team mode), as well as the player's real-time behavioral data in that scene (such as the painting operation trajectory, speed, and pressure when using painting tools). Specifically, environmental parameters can also include the player character's level, current map environment attributes (such as whether it is a combat area, whether it is inside a specific functional building such as a silk shop or a construction factory), and task type (such as main quest, side quest, or daily quest). Real-time behavioral data can also include the player's operational characteristics during the use of tools (such as the completeness of the painting trajectory, the matching degree with the system prompt graphic, the uniformity of the painting speed, and the range of changes in brush pressure), the composition of the player's current team's character classes and skill cooldown status (if applicable), and the player's historical interaction records in that scene (such as whether there have been similar tool usage behaviors recently, task completion status, etc.).

[0099] In addition, the system can also read operation data in real time from the input device operated by the player. For touch screen devices, operation data can include parameters such as touch point coordinates, touch pressure, swipe trajectory speed, and interval between adjacent touch points. For external drawing tablets, pressure sensitivity data can also be read. For keyboard and mouse operations, information such as mouse movement speed, click interval, and scroll wheel adjustment parameters can be read. This data can be continuously collected from the moment the player clicks to activate the item until the player confirms the completion of the operation, fully preserving all dynamic features of the player's operation process.

[0100] S120: Based on the item identification information, current game scene data, player operation data, and preset reference dataset, dynamically evaluate the predicted effectiveness of virtual items.

[0101] In this step, by obtaining the item identification information of the virtual item and the current game scene data through S110, and continuously obtaining the player operation data, this application can dynamically evaluate the predicted effect of the virtual item based on the item identification information, the current game scene data, the player operation data, and the preset reference dataset. In this way, the final effect operation can be executed based on the predicted effect.

[0102] The reference dataset in this application refers to a set of feature matching standards bound to different prop types and game scenes. This set is stored in advance in the game server's database. Each type of prop corresponds to a basic reference template. Different scenes will adjust the feature weights in the template differently, and props of different levels will unlock additional reference feature dimensions. For example, for painting creation props, the preset reference dataset can include three parts: basic outline matching standards, operation smoothness reference thresholds, and style feature reference library. When the scene is a clothing design task, the matching weights of clothing-related features such as collar shape, cuff outline, and garment structure will be increased, and the weights of brushstroke style and color matching will also be adjusted according to the clothing style required by the task. If the scene is an architectural construction design task, the reference dataset will automatically call the features related to architectural structure, increase the matching weights of elements such as door, window, and wall outlines, and reduce the weight of clothing-related features. At the same time, when the prop level reaches the point where advanced creation functions can be unlocked, the reference dataset can also add two additional reference dimensions: detail refinement score and creative matching score, so that the evaluation criteria can adapt to the prop's growth progress.

[0103] During dynamic evaluation, this application first retrieves the basic evaluation framework for the corresponding item type and level from the reference dataset based on the item identification information. Then, it adjusts the weighting of each evaluation dimension in conjunction with the current game scene data. Next, it compares and scores the collected player operation data against the standards in the reference framework one by one. It should be noted that the scoring in this application does not only consider the matching degree of the final generated result like traditional solutions, but also incorporates the dynamic features of the operation process. For example, in the painting creation scenario, in addition to the score for the overlap between the final graphic and the reference outline, the smoothness score of the operation trajectory and the reasonableness score of the brush pressure change can also be added. If the player is painting ancient-style clothing, it will also compare whether the line style conforms to the characteristics of traditional meticulous brushwork or freehand brushwork. These scores are weighted according to the scene-adjusted weights to obtain the final comprehensive score, and then the prediction effectiveness is determined based on the score range. If the score is higher than the passing threshold, it can be considered effective. Based on the score, it can be divided into three different levels: normal effectiveness, excellent effectiveness, and perfect effectiveness. Different levels correspond to different game benefits. For example, normal effectiveness will grant basic task progress, excellent effectiveness will grant additional cloth material rewards, and perfect effectiveness will unlock exclusive clothing patterns. This tiered effectiveness result can better match the player's skill level and give the player more clear feedback on their progress.

[0104] Furthermore, this application can directly correlate the overall score with the probability of virtual items taking effect. Simultaneously, based on different ranges of the evaluation value, specific effect parameters after the item takes effect can be determined, such as the quality level (common, excellent, rare) after successfully designing clothing using painting tools, the base price range when selling, and whether special attributes are added (such as increasing character charm). For items with interactive promotional functions, the potential popularity of the work among players can also be assessed, serving as a preliminary basis for subsequent rankings and economic gains.

[0105] S130: Execute the activation operation of virtual items based on the predicted activation result.

[0106] In this step, after obtaining the predicted effectiveness result through S120, this application can execute the corresponding virtual item activation operation based on the predicted effectiveness result and present matching game feedback to the player. For example, if it is determined to be invalid, the player can be presented with the prompt "The creation does not meet the requirements, please adjust and try again," and the player's current creation content will be retained for easy modification, without directly clearing all content, giving the player more friendly adjustment space; if it is determined to be effective at different levels, different levels of rewards and task progress will be given accordingly. For example, in the ancient style clothing design main quest of the silk shop, perfect effectiveness can not only directly complete the main quest, but also unlock the exclusive side quest of "Royal Weaving", giving players more gameplay extensions and fully mobilizing the player's creative enthusiasm.

[0107] In a specific example, such as Figure 3 As shown, Figure 3 The image shows the interface for players to receive tasks and activate items in an open-world simulation game, as provided in this application embodiment. Figure 3 In this open-world simulation game, players unlock the silk shop management gameplay and receive the main quest to "design a birthday dress for the city lord's wife." Players click to use a painting tool in their inventory that has been upgraded to level three, triggering the tool activation command. The system first obtains the tool's identification information: type is creative - painting tool, level three, style matching and detail scoring functions are unlocked, and the basic judgment requirement is an outline overlap of ≥60%. Next, it obtains the current game scene data: the current scene is the silk shop's creation table, the task type is a main quest, and the task requirement is an ancient-style dress. The system then continuously collects player operation data. The player uses a touchscreen phone to draw, and the system collects the coordinates, pressure, and speed of each swipe, completely recording the player's entire operation process of drawing the neckline, cuffs, body, and hem.

[0108] During the evaluation phase, the system can retrieve the evaluation framework of the Level 3 painting props from the preset reference dataset, and then adjust the weight of each dimension in combination with the ancient style clothing design scene: outline feature weight accounts for 30%, of which the weight of the outline of clothing structure such as collar, cuffs, and body accounts for 80% of the outline score, and the overall outline matching weight accounts for 20%; operation dynamic feature weight accounts for 20%, of which trajectory smoothness accounts for 60% and speed uniformity accounts for 40%; style matching weight accounts for 25%, of which ancient style line features account for 70% and the overall pattern conformity of the dress accounts for 30%; detail score weight accounts for 25%, this dimension, unlocked by the Level 3 props, mainly evaluates the matching degree of pattern details. Subsequently, the system compares the collected player operation data with the reference standard one by one and scores them, and finally weights them to obtain a comprehensive score of 92 points, which belongs to the perfect effectiveness level. Finally, the system executed the corresponding actions for a perfect outcome: directly completing the main quest, unlocking the "Royal Weaving" side quest, rewarding ten sets of top-quality Yun Brocade materials, and granting the exclusive title of "Master Craftsman's Work." The entire process not only matched the player's skill level but also provided sufficient positive feedback, greatly enhancing the player's gaming experience.

[0109] In the above embodiments, by responding to the activation command of virtual props, prop identification information and current game scene data are obtained, and player operation data is continuously acquired. The predicted effectiveness of virtual props is dynamically evaluated by comprehensively considering prop identification information, current game scene data, player operation data, and a preset reference dataset. Compared to traditional single-dimensional judgment logic, this application can fully consider the characteristics of the virtual props themselves, subtle differences in player operation behavior, and the specific needs of the current game scene, making the prop effectiveness result more personalized and scene-adaptable. For example, in a clothing design scenario, this application can not only determine whether the collar and cuff structures drawn by the player meet functional structural standards, but also combine the player's operational trajectory smoothness, speed stability, and other behavioral data, as well as the specific requirements of the current task for clothing style (such as ancient style or modern style), to conduct a comprehensive evaluation, making the prop effectiveness determination more comprehensive and accurate, and better reflecting the player's operational intentions and strategic choices. Furthermore, by acquiring current game scene data (such as the different requirements of silk shop design scenarios and building construction scenarios) and using it as an important dimension for evaluation, this application enables the prop effectiveness determination to produce differentiated results based on different scenarios. For example, in a clothing design scenario, the system focuses on evaluating whether the drawn graphic includes clothing structural elements such as collars and cuffs; in a building construction scenario, the system focuses on evaluating whether it includes architectural structural elements such as doors and windows. This scenario-differentiated judgment mechanism makes the use of props more closely aligned with the game context, enhancing the strategic depth and realism of prop usage. Furthermore, the prop identification information obtained in this application can include prop type and prop level, allowing the system to determine the function unlock status based on the prop level, thereby dynamically invoking corresponding judgment conditions during the evaluation process. For example, when the prop is in its initial stage, only basic graphic matching judgment is enabled; when the prop has been upgraded and unlocked with advanced functions, corresponding preference data or advanced judgment conditions can be additionally invoked. This dynamic adaptation mechanism allows the growth and functional expansion of props to be fully reflected in the effective evaluation, thus enriching the depth of prop usage.

[0110] In one embodiment, S120, dynamically evaluating the predicted effectiveness of the virtual item based on the item identification information, the current game scene data, the player operation data, and a preset reference dataset, may include:

[0111] S121: Call the multi-dimensional effective condition determination model.

[0112] S122: Input the item identification information, the current game scene data, the player operation data, and the preset reference dataset into the multi-dimensional effectiveness condition determination model, and generate the predicted effectiveness result of the virtual item through the multi-dimensional effectiveness condition determination model.

[0113] In this embodiment, when dynamically evaluating the predicted effectiveness of virtual props, a pre-configured multi-dimensional effectiveness condition judgment model can be called first. This model is a deep learning model trained based on a large amount of historical prop usage data. It can automatically assign weights and fuse features for different input features. Compared with manually set fixed weighting rules, it can cope with more varied scenarios and operational features, and the output results have higher accuracy.

[0114] Understandably, during the model training phase, this application can collect hundreds of thousands of usage data points for different types and levels of virtual items in various game scenarios. Each data point includes complete item identification information, scene data, player operation data, and effective result tiers or specific effect parameters annotated by professional planners and experienced players. Next, this application can divide this data into training and testing sets according to a preset ratio, inputting them into the model for iterative training until the model's accuracy on the testing set reaches a preset requirement, at which point training can stop and the model can be deployed for practical use. In actual use, this application only needs to directly input all the collected parameters into the trained model, and the model will automatically output the predicted effective result for the corresponding tier or the predicted effective result for the corresponding probability of effectiveness based on feature matching results. This eliminates the need for manual adjustment of weights, significantly improving system computational efficiency and reducing player waiting time.

[0115] In another embodiment, to further improve the accuracy of the evaluation and avoid misjudging the player's operational intentions, after continuously acquiring player operation data, the collected raw player operation data can be denoised to remove invalid operation data generated by accidental touches. For example, extra sliding trajectories generated by players accidentally touching the screen during the drawing process, or accidental touch landing points generated by clicking on blank areas, do not affect the player's final creative result. If included in the evaluation, they would lower the scores of dimensions such as operation smoothness. Therefore, in the preprocessing stage, these invalid data can be identified and removed by using sliding trajectory length thresholds and click dwell time thresholds, retaining only the valid operation data generated by the player's conscious creation, thereby making the subsequent evaluation results more consistent with the player's actual operation level.

[0116] In one embodiment, such as Figure 4 As shown, Figure 4 This is a model architecture diagram of the multi-dimensional effectiveness condition determination model provided in the embodiments of this application; the multi-dimensional effectiveness condition determination model may include a basic scene matching layer, a player behavior analysis layer, an item characteristic adaptation layer, and a real-time interaction state layer.

[0117] In S122, generating the predicted activation result of the virtual item through the multi-dimensional activation condition determination model may include:

[0118] S1221: Determine the scene matching degree parameter based on the basic scene matching layer and the current game scene data.

[0119] S1222: Determine the operation standardization parameter based on the player behavior analysis layer and the player operation data.

[0120] S1223: Determine the item capability parameters based on the item characteristic adaptation layer and the item identification information.

[0121] S1224: Determine the interaction adaptation parameters based on the real-time interaction state layer and the current game scene data.

[0122] S1225: Generate the predicted effectiveness result of the virtual prop by combining at least two of the scene matching degree parameter, the operation standardization degree parameter, the prop ability parameter and the interaction adaptation parameter.

[0123] In this embodiment, as Figure 4 As shown, the multi-dimensional effective condition determination model can include an input data layer, a first processing layer, a parameter output layer, a comprehensive determination layer, and a result output layer. The first processing layer can be further subdivided into a basic scene matching layer, a player behavior analysis layer, an item characteristic adaptation layer, and a real-time interaction state layer. Therefore, when generating prediction results through the multi-dimensional effective condition determination model, the four layered structures will be responsible for feature extraction and calculation of different dimensions, avoiding mutual interference between different types of features and further improving the accuracy of the results.

[0124] First is the basic scene matching layer. This layer is mainly responsible for extracting the core requirements of the current game scene, such as whether it is a clothing design task or an architectural design task, whether the required style is ancient or modern, and whether the task difficulty is a beginner task or an advanced task. Then, combined with the scene standards corresponding to the preset reference dataset, the matching degree parameter between the current player's creative result and the scene requirements is calculated. This parameter will serve as the basic weight for subsequent calculations, determining the proportion of other dimension parameters.

[0125] Next is the player behavior analysis layer. This layer is specifically designed to process the collected player operation data, extracting dynamic behavioral features such as the continuity of operation trajectory, the uniformity of operation speed, and the accuracy of operation landing point. Combined with the operation standards of the corresponding scenario in the preset reference dataset, the operation standardization parameter is calculated. This is a dimension that traditional evaluation schemes often overlook. It can better reflect the actual level of player operation, rather than just looking at the static matching degree of the final result.

[0126] Next is the item characteristic adaptation layer. This layer can calculate the corresponding item ability parameters based on the item type, item level, and unlocked functions in the item identification information. If it is a low-level basic item, only the basic evaluation dimensions are used. If it is an upgraded advanced item, the unlocked additional evaluation dimensions are used to match the evaluation standards with the item's own ability range. This avoids the problem of low-level items requiring advanced evaluation standards or high-level items being evaluated with low standards, which would lead to distorted results.

[0127] Finally, there is the real-time interaction state layer. This layer is mainly for game scenarios with real-time interaction attributes, such as when players use items in multiplayer team creation tasks or when players use items in timed events. This layer will extract the current real-time interaction state, such as the degree of cooperation between teammates in team tasks, or the impact of the remaining operation time on the completion rate in timed events, and calculate the corresponding interaction adaptation parameters. If it is a single-player non-timed scene, this parameter will be set to the standard value by default and will not have an additional impact on the final result.

[0128] After the four layers complete the calculations and obtain four parameters, the model can perform weighted fusion of the four parameters according to the pre-trained fusion rules, and finally output the prediction result corresponding to the level. Alternatively, it can select at least two parameters from the four parameters according to preset selection rules for comprehensive calculation, which can also generate a prediction result that meets the needs of the scenario. This layered processing mode enables more accurate feature extraction from different dimensions, avoids calculation errors caused by feature cross-interference, and makes the final result more in line with the actual game scenario and player operation. Furthermore, it can also adapt to the evaluation needs of different types of virtual props. Whether it is a creation prop, a construction prop, or an interactive prop, it can be adapted by adjusting the parameters of each layer, without the need to build evaluation logic separately for each prop, which greatly reduces the maintenance cost of game development.

[0129] The aforementioned preset selection rules extract core parameters matching the current virtual item type from four parameters. For example, for creative virtual items, operation standardization and scene matching parameters are prioritized for comprehensive calculation; for functional enhancement items, item ability and interaction adaptation parameters are prioritized; and for multiplayer interactive items, interaction adaptation and scene matching parameters are prioritized. This ensures that the calculation logic always revolves around the core function of the item, avoiding irrelevant parameters from interfering with the final effectiveness determination result. Furthermore, adjustments can be made based on the type of the current game scenario. For instance, in limited-time event scenarios, the weight of the interaction adaptation parameter can be automatically doubled and included in the comprehensive calculation to adapt to the special requirements of real-time scenarios; in single-player regular task scenarios, the initial weights of the four parameters are retained by default to maintain the stability of the evaluation logic. This set of rules ensures the adaptability of evaluations under different scenarios and items without consuming excessive computing resources due to full-parameter, full-dimensional calculations. It achieves a balance between evaluation accuracy and computational efficiency, further reducing player waiting time and improving the smoothness of the game experience.

[0130] In one specific implementation, after the model is constructed, the validity determination is no longer limited to a single scenario parameter, but comprehensively considers the following factors:

[0131] First, verify the basic scene matching degree, that is, check whether the current game scene (such as the silk shop design scene) meets the basic usage scene requirements of the virtual prop. Second, analyze player behavior characteristics. For painting props, focus on analyzing the similarity between the player's painting operation trajectory and the system prompt graphic, the stability of the drawing speed, and dynamic behavioral data such as the drawing pressure at key feature points, to judge the standardization of the player's operation and the accuracy of the expression of intent. Third, combine the prop level and function unlock status. For example, if the painting prop is at the initial stage, only the basic graphic matching judgment in the template-assisted mode is used; if the painting prop has been upgraded and the "draw according to the needs of a confidante" function has been unlocked, then the confidante's preference data (such as color preferences, style tendencies) needs to be additionally called as one of the judgment conditions. Finally, evaluate the real-time interaction status. When used in team mode, it is necessary to consider whether the character classes of other members of the current team can benefit from the drawn prop, or whether there is a conflict with the prop effects of other players. Through this multi-dimensional model, the system can more accurately capture the player's actual operation intention and complex game situations.

[0132] Furthermore, this application can comprehensively evaluate the parsed current game scene data based on a multi-dimensional effectiveness condition determination model to generate item effectiveness probability and expected effect parameters. Specifically, the model can quantify the data in each dimension. For example, the painting matching degree is converted into a matching score of 0-100, and the compatibility between character composition and item effect (such as the correlation between the proportion of group buff items and healing characters) is converted into weight coefficients. Through preset algorithms (such as weighted summation, neural network models, etc.), these quantified data are calculated to obtain a comprehensive evaluation value, which directly corresponds to the item's effectiveness probability.

[0133] Furthermore, this application can also determine the specific effect parameters of the props after they take effect based on different ranges of the comprehensive evaluation value, such as the quality level (common, excellent, rare) after successfully designing clothes using painting props, the base price range when selling, and whether special attributes are added (such as increasing the character's charm value). For props with interactive promotion functions, the potential popularity of the work among players can also be evaluated as a preliminary basis for subsequent ranking and economic gains.

[0134] This multi-dimensional, layered evaluation mechanism, compared to the traditional method of judging effectiveness based solely on static graphic matching, can effectively reduce the probability of misjudgment. It avoids situations where players' actions meet the requirements but fail to be judged due to slight deviations in the graphics, and also prevents actions that do not meet the requirements from being judged as effective due to a coincidental match in the final graphics. This greatly improves the player's gaming experience and makes the feedback on item usage more reasonable and controllable.

[0135] In one embodiment, the current game scene data includes at least one of the following: the map environment where the player character is located, the type of mission being performed, and the current team's character composition.

[0136] In step S1221, determining the scene matching degree parameter based on the basic scene matching layer and the current game scene data may include:

[0137] S211: Determine the usage scenario matching degree of the virtual props based on the map environment.

[0138] S212: Determine the task type matching degree of the virtual prop based on the task type.

[0139] S213: Determine the team configuration matching degree of the virtual props based on the current team's role composition.

[0140] S214: Use the usage scenario matching degree, the task type matching degree, and / or the team configuration matching degree as the scenario matching degree parameter.

[0141] In this embodiment, the current game scene data may include at least one of the following: the map environment where the player character is located, the type of task being performed, and the character composition of the current team. Therefore, when determining the scene matching parameters based on the basic scene matching layer and the current game scene data, this application can calculate the corresponding matching degree for different categories of scene data separately, and then integrate multiple matching degrees to obtain the final scene matching degree parameter.

[0142] The matching degree calculation corresponding to the map environment mainly determines whether the virtual item is allowed to be used in the current map. Some special items have usage range restrictions, such as production items that can only be used in designated production maps, and festival items that can only trigger special effects in event-specific maps. If the map where the current player is located meets the preset usage range requirements of the item, the usage scenario matching degree will be assigned a full score. If it does not meet the requirements, the parameter will be assigned a zero score, and the item will be directly determined to be ineffective. There is no need to perform subsequent dimension calculations, which can also save unnecessary computing resources.

[0143] The calculation of task type matching degree is to determine whether the current task meets the trigger requirements of the item. For example, the creative item mentioned in this application only supports creative tasks such as clothing design and architectural drawing. If the player forcibly uses it in a combat mission, the task type matching degree will give a very low score, and the final predicted effect will be judged as failure, which is in line with the item design logic of the game itself.

[0144] Team configuration matching mainly refers to functional items in multi-player team scenarios, such as group buff items. If the current team mainly consists of melee DPS characters, then buff items that increase melee damage will have a higher team configuration matching degree. If it is a buff item that increases ranged spell damage, the matching degree will be relatively lower. This parameter can also more accurately reflect the compatibility between the item and the current team, so that the effect is more in line with the actual team needs.

[0145] After calculating the matching scores across the three dimensions mentioned above, the matching scores for the usage scenario, task type, and team configuration can be integrated based on whether the current scenario is a multiplayer mode or a special map, resulting in the final scenario matching score parameter. For example, in a single-player, regular, non-special map scenario, only the usage scenario matching score and task type matching score need to be integrated; the team configuration matching score does not need to be included separately, further simplifying the calculation logic and saving computational resources. In a multiplayer special event map scenario, all three matching scores can be included in the integrated calculation, and a comprehensive scenario matching score parameter can be obtained according to preset weights, ensuring that the scenario dimension judgment covers all key information and does not miss any core constraints. This decomposed calculation method not only ensures the completeness of the scenario matching judgment but also flexibly adjusts the calculation content according to different scenarios, further balancing the accuracy of the judgment and the computational efficiency.

[0146] In one embodiment, the player operation data may include at least one of the player's operation trajectory, operation speed, and operation pressure during the process of activating items.

[0147] In S1222, determining the operation standardization parameter based on the player behavior analysis layer and the player operation data may include:

[0148] S221: Determine the trajectory matching parameters based on the matching degree between the operation trajectory and the preset reference dataset.

[0149] S222: Determine the speed consistency parameter based on the stability of the operating speed.

[0150] S223: Determine the pressure uniformity parameter based on the range of change of the operating pressure.

[0151] S224: Use the trajectory matching parameter, the speed consistency parameter, and / or the pressure uniformity parameter as the operation standardization parameter.

[0152] In this embodiment, when the player operation data includes at least one of the player's operation trajectory, operation speed, and operation pressure during the process of activating the item, this application can decompose and calculate from three dimensions respectively, obtain the corresponding sub-parameters, and then integrate them into the operation standardization parameter, which can more accurately reflect the true state of the player's operation and will not misjudge the player's operation intention due to the deviation of a single dimension.

[0153] The trajectory matching parameter is calculated by matching the continuous trajectory points left by the player during operation with the standard operation trajectory in the preset reference dataset. The overall trajectory matching degree is obtained by calculating the average Euclidean distance between each trajectory point and the standard trajectory. The smaller the distance, the closer the player's operation trajectory is to the standard trajectory, and the higher the score of the trajectory matching parameter. This dimension can directly reflect whether the player's operation meets the basic path requirements for using the prop. For example, when using a drawing prop to draw the outline of a specific pattern, the trajectory matching degree can directly reflect the degree of fit between the outline drawn by the player and the required outline.

[0154] The speed consistency parameter is calculated based on the fluctuations in the player's operation speed during the operation. This application divides the entire operation into multiple time segments of equal length, counts the displacement within each time segment, and calculates the variance of multiple displacements. The smaller the variance, the more uniform and stable the player's operation speed, and the higher the score of the speed consistency parameter. This dimension can identify whether the player's operation is a continuous, proactive creation or a coincidental result of accidental touches. For example, some players may ultimately draw a graphic that coincidentally meets the requirements, but if the operation process is intermittent and multiple drags and adjustments deviate from the path, the speed consistency parameter will give a low score, thereby reducing the final probability of success and avoiding unreasonable accidental success.

[0155] The pressure uniformity parameter is designed for devices that support pressure-sensitive operation. For example, when a player uses a stylus or pressure-sensitive drawing tablet to operate on a mobile device, different operating pressures are recorded by the device. In standard operation, key feature points usually require greater operating pressure to emphasize the features, while the pressure at non-key locations is relatively stable. This application can statistically analyze the range of changes in the player's operating pressure throughout the entire process, compare it with the standard pressure change patterns in the reference dataset, and calculate the similarity between the two to obtain the pressure uniformity parameter. If the player's pressure changes conform to the standard pattern, the parameter score is higher, and vice versa. This dimension can further identify the player's operating intent, distinguish between intentional drawing and misoperation, and further improve the accuracy of the judgment.

[0156] After obtaining the three sub-parameters, this application can also flexibly integrate them to obtain the operation standardization parameter based on the type of device and item currently in use. If the player is using a regular mouse that does not support pressure sensing, only the trajectory matching parameter and speed consistency parameter are integrated, without including the pressure uniformity parameter. This ensures that the accuracy of the judgment is not affected and also avoids invalid calculations. If the player is using a professional device that supports pressure sensing, all three parameters can be integrated and calculated, making full use of the operation data provided by the device to improve the accuracy of the judgment. This flexible and adaptable calculation method can be compatible with the operation needs of different devices and different scenarios, providing fair and accurate effective judgments for players on different devices.

[0157] In one embodiment, the item identification information may include at least one of virtual item type and virtual item level.

[0158] In step S1223, determining the item capability parameters based on the item characteristic adaptation layer and the item identification information may include:

[0159] S231: Determine the basic function parameters based on the type of virtual prop.

[0160] S232: Determine the function unlock status parameters based on the virtual item level.

[0161] S233: Use the basic function parameters and / or the function unlock status parameters as the item ability parameters.

[0162] In this embodiment, when the item identification information includes at least one of virtual item type and virtual item level, the item ability parameters that accurately reflect the item's current actual ability can be obtained by decomposing and calculating the item's own attributes. This avoids using the judgment logic of fully equipped items to evaluate low-level items with unlocked functions, making the effectiveness judgment more consistent with the item's actual growth status.

[0163] The basic function parameters are determined based on the core type of the virtual prop. Different types of virtual props correspond to different core function ranges. For example, the core function output of creative props is work matching, the core function output of beneficial props is attribute enhancement, and the core function output of interactive props is social connection. This application can directly give the basic score of the corresponding core function based on the prop's preset type label. For example, a full score is given for usage scenarios that match the core function positioning, and a low basic score is given for usage attempts that exceed the core function positioning, thus clarifying the basic boundaries of the prop's capabilities from the root.

[0164] The function unlock status parameter is determined by the current level of the virtual item. In most games, virtual items unlock additional functions as the level increases. For example, a level 1 painting item can only unlock the basic template drawing function, while a level 3 item can unlock the custom creation function (e.g., Figure 5 The game features functions such as drawing portals, summoning spirit pets, modifying terrain, reviving teammates, and invisibility cloaks. At level 5, the customizable concubine preference function can be unlocked. If the current item level meets the unlock requirements of the corresponding function, the function unlock status parameter will be given full marks. If it does not meet the requirements, the corresponding function's score will be deducted and it will be excluded from the judgment logic. This will prevent the unreasonable situation of triggering the effect of a high-level function before the item level is high, thus ensuring the stability of the game's numerical system.

[0165] Furthermore, this application can be flexibly adjusted according to the actual situation during integration. For example, if the current prop only has a type label and no level growth system, the basic function parameters can be directly used as the final prop ability parameters. If the prop has both type classification and level growth, the two sub-parameters can be integrated according to the preset weight to obtain the final prop ability parameters. This set of decomposition calculation logic is consistent with the calculation logic of the previous scene and operation dimensions, and also takes into account the judgment accuracy and calculation efficiency.

[0166] In one embodiment, determining the interaction adaptation parameters based on the real-time interaction state layer and the current game scene data in step S1224 may include:

[0167] S241: When the current game scene data indicates team mode, determine the enhancement adaptation parameters of the item effect based on the role class of other members of the current team.

[0168] S242: and / or, determine the conflict detection parameters for item effects based on the active items of other members of the current team.

[0169] In this embodiment, when determining the interaction adaptation parameters based on the real-time interaction state layer and the current game scene data, this application can start from the actual interaction situation in team mode and calculate the parameters of the two dimensions of gain adaptation and effect conflict respectively, so that the final evaluation result is more in line with the actual state of the team battle.

[0170] The buff adaptation parameter is mainly calculated based on the class distribution of other members in the current team. For example, if a player uses a group magic buff item, but more than 90% of the current team members are physical damage dealers and there is only one magic damage dealer, then the actual buff coverage of this item will be very limited, the buff adaptation parameter will give a lower score, and the actual buff strength after the item takes effect will also be reduced. If most of the team members are magic damage dealers, then the buff adaptation parameter will give a high score, and the effect after the item takes effect will be calculated at full strength. This judgment method is more in line with the actual benefits of the team and can also guide players to choose the use of items reasonably according to the team configuration, thus improving the strategic nature of the game.

[0171] The conflict detection parameter is used to detect whether the item to be used conflicts with other items already in effect in the team. For example, if a player has already used the same type of "spell power increase" item, and the same type of buff effect does not stack in the game rules, only the one with the highest strength will take effect, then if the player uses the same type of item later, the conflict detection parameter will give a very low score, directly determining that the item cannot trigger the normal effect, and will only give the feedback "effect conflict cannot take effect". This is in line with the game's own rule setting, and can also prevent players from wasting valuable items, thus improving the overall game experience.

[0172] If there is no effect conflict, the conflict detection parameter will remain at full score and will not have a negative impact on the final evaluation result. This set of interactive adaptation calculation logic specifically solves the special judgment requirements in the multi-player team mode, fills the gap in the traditional single judgment scheme that does not consider team interaction, and makes the item effect evaluation more in line with the actual situation of complex battles.

[0173] In one embodiment, such as Figure 6 As shown, Figure 6 This application provides a flowchart illustrating the process for verifying whether player operation data meets functional structural standards; the method may further include:

[0174] S140: Obtain the functional structural standard corresponding to the current game scene. The functional structural standard is used to define the structural elements that the player's operation input must satisfy.

[0175] S141: Verify whether the player operation data meets the functional structural standard.

[0176] S142: If the conditions are not met, the virtual item is directly determined to be invalid, and no evaluation of the predicted effect is performed.

[0177] In this embodiment, in addition to the above-mentioned effectiveness evaluation method, this application can also screen out operations that obviously do not meet the effectiveness conditions in advance through hard structural requirements, without having to enter the multi-dimensional parameter calculation process, thus further saving computing resources.

[0178] For example, when creating clothing using drawing tools, the functional structural standard requires drawing three core structural elements: the collar, the body, and the cuffs. If a player only draws an incomplete outline of the body, lacking the other two core structures, it will be directly judged as not meeting the standard, and the result of the tool failing to take effect will be output directly. There is no need to continue calculating parameters of multiple dimensions such as scene matching and operation specifications, which greatly shortens the judgment process and reduces the computing pressure on the server. Especially during peak periods when there are many players online at the same time, this pre-screening mechanism can effectively reduce server load and ensure the smooth operation of the game as a whole.

[0179] Understandably, the functional structural standards in this application will be flexibly set according to the usage requirements of different items. Not all items need to have multi-layered structural requirements. For example, simple one-time buff items only need to meet the single structural requirement of "click to activate," which will not add unnecessary verification thresholds to normal usage scenarios. This differentiated pre-verification mechanism filters out invalid operations without affecting the normal usage process, further balancing judgment efficiency and player experience.

[0180] In one embodiment, the process of determining the functional structural criteria may include:

[0181] S401: When the current game scene is a clothing design scene, the functional structural standard includes at least the neckline structural element and the cuff structural element.

[0182] S402: When the current game scene is a building construction scene, the functional structural standard includes at least door structural elements and window structural elements.

[0183] In this embodiment, the functional structural standard will preset corresponding core elements for different types of creative scenarios, without the need to set rules temporarily for each use. This ensures the rationality of advance screening without adding extra rule configuration burden.

[0184] For example, in a clothing design scenario, the core is to generate a wearable clothing appearance. The collar and cuffs are the core structures that distinguish different clothing styles and meet basic wearing requirements. Without these two elements, it is impossible to generate qualified clothing that meets the requirements. Therefore, these two elements must be included in the hard verification standards. In a building construction scenario, to generate a usable qualified building, the door is a necessary structure for entering the building, and the window is a necessary structure for meeting the requirements of lighting and the integrity of the building's appearance. Without these two structures, the generated building does not meet the basic usage requirements. Therefore, these two structures are considered necessary verification elements.

[0185] Of course, developers can also flexibly expand the content of the functional structural standards based on the new scene types added in the game. For example, when weapon forging scenes are opened later, "grip structure elements" and "blade structure elements" can be added as hard standards. The entire set of rules is highly scalable and can adapt to the new content requirements after the game version is updated. The rules can be upgraded without modifying the overall evaluation framework.

[0186] In one embodiment, the method may further include:

[0187] S150: When the virtual props take effect, generate the effective work.

[0188] S151: In response to the sharing instruction for the effective work, publish the effective work to the shared work pool.

[0189] S152: Determine the ranking information of the effective work based on the interaction data of the effective work in the shared work pool.

[0190] S153: Provide additional economic benefits to the players corresponding to the effective works based on the ranking information.

[0191] In this embodiment, after the virtual props take effect and generate works that meet the needs of players, this application also provides a supporting mechanism for sharing works and incentivizing revenue, which can stimulate players' creative enthusiasm, enrich the social interaction atmosphere in the game, and at the same time accumulate more high-quality user-generated content for the official team, enriching the game's content ecosystem.

[0192] When a player chooses to share their work to the shared work pool, the system will collect various interaction data for that work in real time, including likes, favorites, downloads, comments, and shares. Different types of interaction data can be converted into a total interaction score according to preset weights. For example, a download has a higher weight than a like because downloading and using represents other players' actual recognition of the work; comments have a higher weight than simple likes because they represent players' willingness to invest time in expressing their appreciation for the work. Based on the total interaction score, the system can automatically generate the work's ranking information in the corresponding category, for example, as shown below. Figure 7 As shown, Figure 7 This is a display diagram of the clothing drawing ranking interface after using drawing tools to draw clothing, provided in an embodiment of this application. This application scores the effective works drawn by each user in the shared work pool, and ranks them according to the total interaction score. Then, it displays each user and the clothing drawn using drawing tools in the clothing drawing ranking interface under the clothing design category. Users can view the ranking through this interface.

[0193] Furthermore, creators of top-ranked works can receive additional in-game economic benefits, including bound ingots, game coins, exclusive creation titles, and even limited-edition creation item fragments. The higher the ranking, the richer the rewards. Creators of top-ranked popular works can even receive exclusive limited-edition avatar frames and chat icons, satisfying players' sense of creative accomplishment.

[0194] This incentive mechanism effectively guides players to create higher-quality works, rather than randomly generating low-quality works to fill the shared pool. It ensures the quality of content in the shared pool, making it easier for other players to download high-quality custom content. In turn, it further increases the interactive benefits for high-quality creators, forming a positive cycle. At the same time, it can also extend the time players spend in the game, improving overall player stickiness and activity.

[0195] In one embodiment, the activation operation of the virtual item in S130 based on the predicted activation result may include:

[0196] S131: When the probability of effectiveness in the predicted effectiveness result reaches or exceeds a preset threshold, the virtual item is determined to be effective, and an effectiveness effect is generated based on the expected effect parameters in the predicted effectiveness result.

[0197] S132: When the probability of effectiveness in the predicted effectiveness result does not reach the preset threshold, it is determined that the virtual item has not taken effect, and a failure feedback message is generated.

[0198] In this embodiment, when performing the activation operation of virtual props based on the predicted activation result, this application can clearly define the boundary between activation and non-activation by setting a preset threshold, so that the entire evaluation result can be finally translated into clear and perceptible game feedback. This not only meets the player's expected understanding of the prop usage result, but also ensures the transparency of the judgment rules, avoiding situations where the result is ambiguous and causes confusion for the player.

[0199] If the calculated probability of effectiveness reaches or exceeds the developer's preset threshold, the virtual prop is directly determined to be effective. Furthermore, it will combine the expected effect parameters calculated from the previous dimensions to generate an effective effect that meets the parameter requirements. For example, when using creative props to generate clothing, this application can adjust the fit and detail of the generated clothing according to parameters such as scene matching degree and operation standardization. The higher the probability of effectiveness and the better the parameters, the more closely the appearance of the generated clothing matches the player's drawing ideas, and the closer the attribute bonus will be to the maximum value. If it is a buff prop, the actual buff magnitude can be adjusted according to the interaction adaptation parameters. The higher the adaptation degree, the stronger the buff effect, which fully conforms to the results of the previous multi-dimensional parameter calculations, ensuring that the effect of each prop use matches the actual operation and scene state.

[0200] If the success rate does not reach the preset threshold, the virtual item is directly deemed ineffective. Furthermore, based on the calculation results of the preceding parameters, targeted failure feedback is generated, going beyond a vague "failed to take effect" message to explain the specific reason for the failure. For example, if the failure is due to improper operation, the application could display a message such as "The drawn trajectory does not meet the operation requirements; the item has failed to take effect. Please try drawing again." If the failure is due to an insufficient item level, the application could display a message such as "The current item level does not unlock the corresponding function; please upgrade the item level before using it." If the failure is due to a conflicting effect in a team scenario, the application could display a message such as "The current team already has a similar buff item in effect; similar effects cannot be stacked; the item has failed to take effect." Clear failure feedback allows players to quickly identify the problem, facilitating adjustments to their actions or changing their strategies, significantly improving the player experience and reducing negative emotions caused by failures for unknown reasons.

[0201] Of course, the preset threshold can also be flexibly adjusted by the developers according to the positioning of different items. For example, ordinary low-value consumable items can be set with a lower threshold so that most normal operations can be effective, improving the smoothness of the player's experience; while high-value rare and powerful items can be set with a slightly higher threshold to ensure the threshold for using the items, and also to highlight the scarcity of high-value items, which is in line with the overall numerical balance design of the game.

[0202] In one embodiment, the preset threshold is dynamically adjusted based on the rarity of the virtual item and / or game balance requirements.

[0203] In this embodiment, for top-tier items with higher rarity, developers can appropriately increase the preset threshold and flexibly adjust the values ​​in combination with the numerical balance requirements of the current game version.

[0204] For example, if a certain type of powerful item is too dominant in the current version, affecting the fairness of matches between different player groups, the activation threshold of that type of item can be slightly increased, appropriately reducing its activation probability to prevent excessive abuse of that type of item and maintain the overall balance and stability of the match. Conversely, if a certain type of item has consistently had a low usage rate since its design, and most players do not choose to carry or use it, its activation threshold can be appropriately lowered to reduce the barrier to entry, encourage players to try this type of item, increase its usage rate, and allow more items to enter the players' selection range, enriching the strategic choices in the match.

[0205] This dynamic adjustment mechanism allows the entire effectiveness evaluation system to adapt to continuous game version updates. It can make minor adjustments to the numerical balance without modifying the core evaluation framework, greatly reducing the cost of rule modification during version updates and ensuring that the entire evaluation system always fits the current ecological state of the game.

[0206] In one embodiment, the activation operation of the virtual item based on the predicted activation result in S130 may further include:

[0207] S133: Generate multimodal effect feedback information and present it to the player; wherein, the multimodal effect feedback information includes text prompts, dynamic visual effects, and / or 3D preview effects.

[0208] In this embodiment, regardless of whether the virtual item ultimately takes effect or not, the system will provide players with multimodal feedback information, rather than just monotonous text prompts, so that players can more intuitively perceive the result of using the item and improve the overall experience of the operation process.

[0209] Indicatively, such as Figure 8 , Figure 9 As shown, Figure 8 This is a schematic diagram of the dynamic visual effects display interface provided in the embodiments of this application. Figure 9 This is a 3D artwork preview interface shown in an embodiment of this application. In this application, after the prop successfully takes effect and generates the corresponding artwork, in addition to displaying a text prompt to inform the player that the effect has been successfully achieved, the system can also play corresponding dynamic visual effects, such as... Figure 8Rare-level creations in the game will be accompanied by exclusive flowing light effects and particle animations, while legendary-level creations will trigger changes in lighting and shadows across the entire scene, giving players a strong sense of ritual and accomplishment. For creative works that generate effects, the system will also automatically generate corresponding 3D preview effects, such as... Figure 9 As shown, players can rotate and zoom their artwork to view the final product from different angles. They can also adjust detailed parameters directly in the preview interface, without needing to enter the actual scene to wear or place the artwork, saving players adjustment time. Furthermore, if interactive promotional features are involved, this application can also display a message such as "Your artwork has entered the recommendation pool; you can invite friends to vote."

[0210] If an item fails to work, multimodal feedback allows players to quickly grasp the failure information. In addition to targeted text prompts, such as "Insufficient drawing matching (currently 65 points, requires 80 points or above), please try to get closer to the outline of the prompt graphic" or "Your drawing pattern lacks cuff structure, please complete it and try again," it also provides operation suggestions (such as "Slow down the drawing speed and pay attention to the continuity of the lines") to help players understand and improve their operation, increasing the success rate of the next use. It can also be paired with a soft failure prompt animation, so that it will not cause players to feel disgusted by the overly abrupt effect, while also allowing players to notice the result feedback immediately.

[0211] This multimodal feedback method meets the current needs of players for game interaction experience, making the entire process of using props more layered, and making the result information more clearly and accurately delivered, preventing players from missing key prompts.

[0212] In one embodiment, the multimodal effect feedback information may further include:

[0213] S134: When the effect is determined to be effective, display the dynamic effect process, the 3D preview effect of the effective work, and the estimated value information.

[0214] S135: When the condition is determined to be ineffective, display the specific reasons for the ineffectiveness and suggestions for improvement.

[0215] In this embodiment, the multimodal effect feedback information can include not only the feedback information when the determination is effective, but also the feedback information when the determination fails.

[0216] For example, once an item is determined to have taken effect, the system can display the complete dynamic effect process step by step from the moment the item is triggered, instead of directly displaying the finished product. This allows players to experience the magic of the item throughout the process and enhances the sense of ritual when using the item.

[0217] For example, in creative items, the system displays the complete process step-by-step, from the final stroke of the drawing to the glowing lines, the completed outline, the filling of colors, and the addition of detailed textures. The entire process is smooth and natural, gradually shaping the player's creative idea and providing a stronger sense of participation and satisfaction. After the dynamic display is complete, players can directly stop at the 3D preview interface. They can freely drag and rotate the artwork, zoom in to view details such as the collar and embroidery, and directly switch between different wearing scenes to preview the effect. For example, players can place the newly designed clothing on their character to see how it looks under different lighting conditions. Once satisfied, they can save the artwork; if not, they can go back and adjust it without repeatedly changing scenes. The interface also clearly displays the estimated value of the artwork, including how many in-game coins it will fetch when sold to an NPC, the price range it will get on the trading market, and the numerical value of its inherent attribute bonuses, allowing players to clearly understand the value of their creations.

[0218] If an item fails to function, in addition to a written explanation of the reason for the failure, the system can also provide targeted suggestions for improvement based on current operational data, helping players understand and improve their actions. For example, in the clothing design scenario mentioned above, if the insufficient matching score is due to broken lines, the system, besides explaining that "the current drawing matching score is 62 points, and it needs to reach 70 points to be effective," can also suggest, "Try slowing down the drawing speed and keeping the pen tip on the screen to complete the entire outline drawing, which can improve the matching score." If the core structure verification fails, such as forgetting to draw the neckline, the system can provide prompts such as... Figure 10 The message "The current work lacks a neckline structure and does not meet the basic requirements of clothing design. Please add the neckline outline and resubmit" can be displayed. The system can also automatically circle the position where the neckline needs to be added in the drawing interface, making it convenient for players to modify directly. If the item is used again before its cooldown time has ended, the system can prompt "The item is still on a 12-second cooldown. It can be used again after the cooldown ends" and directly display the remaining cooldown countdown to help players arrange their operation rhythm reasonably.

[0219] This clear explanation of the reasons and feedback method, along with suggestions for improvement, greatly reduces the learning cost for players, especially novice players who are just getting started with these custom creation tools. They can quickly master the correct usage, increase the success rate of subsequent use, and reduce the frustration caused by repeated failures.

[0220] In one embodiment, the method may further include:

[0221] S160: Record the complete data of this virtual item usage, including the activation command, the current game scene data, the player operation data, the prediction effect result, and the result of the effective operation.

[0222] S161: Store the complete data in the player behavior database.

[0223] In this embodiment, after each player uses a virtual item, the system automatically records all complete data from the initiation of the command to the final result, without missing any key information. This data includes not only the content of the activation command initiated by the player, various parameters of the game scene at that time, the trajectory and input information of each step of the player's actual operation, but also the predicted effect result calculated by the system and the actual result after the final activation operation. All data is organized in a unified format and stored in a dedicated player behavior database for easy retrieval and analysis later.

[0224] The stored complete data has multiple applications. First, it provides real training samples for subsequent algorithm model iterations. Developers can access a large amount of actual player usage data to optimize the model parameters for calculating the probability of effectiveness and predicting results, making the predictions increasingly aligned with players' actual operational intentions and continuously improving the accuracy of effectiveness assessments. Second, when players have questions about the results of item usage, or encounter issues such as items being consumed but the results being abnormal, customer service and development personnel can directly access the complete usage data for the corresponding time period to quickly pinpoint the cause of the problem, providing players with accurate answers and solutions, improving the efficiency of problem handling and the service experience. Furthermore, this data can be used to analyze the usage habits of different player groups for various virtual items. For example, it can statistically analyze the frequency of use of creative items by players at different levels, the distribution of the probability of effectiveness for items of different rarities, and the types of operation failures players frequently experience. These analytical results can help developers better adjust item design, optimize usage rules, and even launch new item features that better meet player needs, making the entire game's item system increasingly aligned with player requirements.

[0225] In one embodiment, when the process of dynamically evaluating the predicted effectiveness of the virtual item calls a multi-dimensional effectiveness condition determination model for evaluation, the method may further include:

[0226] Based on the complete data of multiple players stored in the player behavior database, the parameters of the multi-dimensional effective condition determination model are optimized by machine learning algorithms. The parameters include the weights and evaluation thresholds of each dimension of data.

[0227] In this embodiment, the initial parameters of the multi-dimensional effective condition determination model are often preset by the developers based on experience, which may deviate from the actual player operation habits and scene distribution. The massive amount of real player usage data stored in the database can be used as training samples for machine learning to continuously optimize the various parameters of the model and make the model's evaluation results more and more accurate.

[0228] For example, developers initially thought player action data should be weighted at 0.6 and scene matching at 0.4. However, after training with a large amount of real data, they found that scene matching had a greater impact on the final effect in most custom creation scenarios. Therefore, the weight ratio of the two dimensions can be automatically adjusted to make the evaluation results more consistent with the needs of actual scenarios. Another example is that for a certain type of low-value consumable item, the initial preset effectiveness threshold was set at 60 points. However, statistical data showed that 80% of normal actions could reach above 70 points, indicating that the threshold was set too low, making this type of item too easy to activate and affecting the game's numerical balance. In this case, the threshold can be automatically and slightly increased through model training to make the evaluation rules more closely reflect actual usage.

[0229] This continuous optimization mechanism based on real player data allows the multi-dimensional condition determination model to evolve over time as the game operates, resulting in increasingly higher evaluation accuracy. It eliminates the need for developers to manually adjust parameters one by one and ensures that the model always adapts to changes in player operating habits, greatly reducing model maintenance costs and continuously improving the player experience when using virtual items.

[0230] The following describes the virtual item effectiveness evaluation device provided in the embodiments of this application. The virtual item effectiveness evaluation device described below and the virtual item effectiveness evaluation method described above can be referred to in correspondence.

[0231] In one embodiment, such as Figure 11 As shown, Figure 11 This application provides a schematic diagram of the structure of a virtual item effectiveness evaluation device according to an embodiment of the present application; the present application also provides a virtual item effectiveness evaluation device, which may include a data acquisition module 210, an effectiveness evaluation module 220, and an item effectiveness module 230, specifically including the following:

[0232] The data acquisition module 210 is used to respond to the activation command of the virtual item, acquire the item identification information of the virtual item and the current game scene data, and continuously acquire player operation data.

[0233] The effectiveness evaluation module 220 is used to dynamically evaluate the predicted effectiveness result of the virtual item based on the item identification information, the current game scene data, the player operation data, and the preset reference dataset.

[0234] The prop activation module 230 is used to perform the activation operation of the virtual prop based on the predicted activation result.

[0235] In the above embodiments, by responding to the activation command of virtual props, prop identification information and current game scene data are obtained, and player operation data is continuously acquired. The predicted effectiveness of virtual props is dynamically evaluated by comprehensively considering prop identification information, current game scene data, player operation data, and a preset reference dataset. Compared to traditional single-dimensional judgment logic, this application can fully consider the characteristics of the virtual props themselves, subtle differences in player operation behavior, and the specific needs of the current game scene, making the prop effectiveness result more personalized and scene-adaptable. For example, in a clothing design scenario, this application can not only determine whether the collar and cuff structures drawn by the player meet functional structural standards, but also combine the player's operational trajectory smoothness, speed stability, and other behavioral data, as well as the specific requirements of the current task for clothing style (such as ancient style or modern style), to conduct a comprehensive evaluation, making the prop effectiveness determination more comprehensive and accurate, and better reflecting the player's operational intentions and strategic choices. Furthermore, by acquiring current game scene data (such as the different requirements of silk shop design scenarios and building construction scenarios) and using it as an important dimension for evaluation, this application enables the prop effectiveness determination to produce differentiated results based on different scenarios. For example, in a clothing design scenario, the system focuses on evaluating whether the drawn graphic includes clothing structural elements such as collars and cuffs; in a building construction scenario, the system focuses on evaluating whether it includes architectural structural elements such as doors and windows. This scenario-differentiated judgment mechanism makes the use of props more closely aligned with the game context, enhancing the strategic depth and realism of prop usage. Furthermore, the prop identification information obtained in this application can include prop type and prop level, allowing the system to determine the function unlock status based on the prop level, thereby dynamically invoking corresponding judgment conditions during the evaluation process. For example, when the prop is in its initial stage, only basic graphic matching judgment is enabled; when the prop has been upgraded and unlocked with advanced functions, corresponding preference data or advanced judgment conditions can be additionally invoked. This dynamic adaptation mechanism allows the growth and functional expansion of props to be fully reflected in the effective evaluation, thus enriching the depth of prop usage.

[0236] In one embodiment, this application also provides a computer-readable storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the virtual item effectiveness evaluation method as described in any of the above embodiments.

[0237] In one embodiment, this application also provides a computer device, including: one or more processors, and memory;

[0238] The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the virtual item effectiveness evaluation method as described in any of the above embodiments.

[0239] Indicatively, such as Figure 12 As shown, Figure 12 This is a schematic diagram of the internal structure of a computer device 300 provided in an embodiment of this application. The computer device 300 can be provided as a server. (Refer to...) Figure 12 The computer device 300 includes a processing component 302, which further includes one or more processors, and memory resources represented by memory 301 for storing instructions executable by the processing component 302, such as application programs. The application programs stored in memory 301 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 302 is configured to execute instructions to perform the effectiveness evaluation method for virtual props in any of the above embodiments.

[0240] The computer device 300 may also include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 may operate on an operating system stored in memory 301, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0241] Those skilled in the art will understand that Figure 12 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0242] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0243] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0244] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for evaluating the effectiveness of virtual items, characterized in that, The method includes: In response to the activation command of the virtual item, the system obtains the item identification information of the virtual item and the current game scene data, and continuously obtains player operation data; The predicted effectiveness of the virtual item is dynamically evaluated based on the item identification information, the current game scene data, the player operation data, and the preset reference dataset. The activation operation of the virtual item will be performed based on the predicted activation result.

2. The method according to claim 1, characterized in that, The step of dynamically evaluating the predicted effectiveness of the virtual item based on the item identification information, the current game scene data, the player operation data, and a preset reference dataset includes: Invoke the multi-dimensional effective condition determination model; The item identification information, the current game scene data, the player operation data, and the preset reference dataset are input into the multi-dimensional effectiveness condition determination model, and the predicted effectiveness result of the virtual item is generated through the multi-dimensional effectiveness condition determination model.

3. The method according to claim 2, characterized in that, The multi-dimensional effective condition determination model includes a basic scene matching layer, a player behavior analysis layer, an item characteristic adaptation layer, and a real-time interaction state layer. The step of generating the predicted effectiveness result of the virtual item through the multi-dimensional effectiveness condition determination model includes: The scene matching degree parameter is determined based on the basic scene matching layer and the current game scene data; The operation standardization parameter is determined based on the player behavior analysis layer and the player operation data; The prop ability parameters are determined based on the prop characteristic adaptation layer and the prop identification information; The interaction adaptation parameters are determined based on the real-time interaction state layer and the current game scene data. The predicted effectiveness result of the virtual prop is generated by combining at least two of the scene matching degree parameter, the operation standardization degree parameter, the prop ability parameter, and the interaction adaptation parameter.

4. The method according to claim 3, characterized in that, The current game scene data includes at least one of the following: the map environment where the player character is located, the type of task being performed, and the current team's character composition. The step of determining the scene matching degree parameter based on the basic scene matching layer and the current game scene data includes: The matching degree of the usage scenario of the virtual props is determined based on the map environment; Determine the task type matching degree of the virtual item based on the task type; The team configuration matching degree of the virtual props is determined based on the current team's role composition; The usage scenario matching degree, the task type matching degree, and / or the team configuration matching degree are used as the scenario matching degree parameter.

5. The method according to claim 3, characterized in that, The player operation data includes at least one of the player's operation trajectory, operation speed, and operation pressure during the process of activating items; The step of determining the operation standardization parameter based on the player behavior analysis layer and the player operation data includes: The trajectory matching parameters are determined based on the degree of matching between the operation trajectory and the preset reference dataset; Determine the speed consistency parameter based on the stability of the operating speed; The pressure uniformity parameter is determined based on the range of variation of the operating pressure; The trajectory matching parameter, the speed consistency parameter, and / or the pressure uniformity parameter are used as the operation standardization parameter.

6. The method according to claim 3, characterized in that, The item identification information includes at least one of virtual item type and virtual item level; The step of determining the item ability parameters based on the item characteristic adaptation layer and the item identification information includes: Determine the basic functional parameters based on the type of virtual item; The function unlock status parameters are determined based on the virtual item level; The basic function parameters and / or the function unlock status parameters are used as the item ability parameters.

7. The method according to claim 3, characterized in that, The step of determining the interaction adaptation parameters based on the real-time interaction state layer and the current game scene data includes: When the current game scene data indicates team mode, the enhancement parameters of the item effect are determined based on the role classes of other members of the current team. And / or, determine the conflict detection parameters for item effects based on the active items of other members of the current team.

8. The method according to any one of claims 1 to 7, characterized in that, The method further includes: Obtain the functional structural standards corresponding to the current game scene. The functional structural standards are used to define the structural elements that player input must satisfy. Verify whether the player operation data meets the functional structural standards; If the conditions are not met, the virtual item will be deemed invalid, and no evaluation of the predicted effectiveness will be performed.

9. The method according to claim 8, characterized in that, The process of determining the functional structural criteria includes: When the current game scene is a clothing design scene, the functional structural standard includes at least the neckline structural element and the cuff structural element; When the current game scene is a building construction scene, the functional structural standard includes at least door structural elements and window structural elements.

10. The method according to any one of claims 1 to 7, characterized in that, The method further includes: When the virtual props take effect, a working artwork is generated; In response to the sharing instruction for the effective work, the effective work is published to the shared work pool; The ranking information of the effective works is determined based on the interaction data of the effective works in the shared works pool; Based on the ranking information, additional economic benefits will be provided to the players corresponding to the effective works.

11. The method according to any one of claims 1 to 7, characterized in that, The step of performing the activation operation of the virtual item based on the predicted activation result includes: When the probability of effectiveness in the predicted effectiveness result reaches or exceeds a preset threshold, the virtual item is determined to be effective, and an effectiveness effect is generated based on the expected effect parameters in the predicted effectiveness result. If the probability of effectiveness in the predicted effectiveness result does not reach the preset threshold, it is determined that the virtual item has not taken effect, and a failure feedback message is generated.

12. The method according to claim 11, characterized in that, The preset threshold is dynamically adjusted based on the rarity of the virtual item and / or game balance requirements.

13. The method according to claim 11, characterized in that, The step of performing the activation operation of the virtual item based on the prediction activation result further includes: Generate multimodal effect feedback information and present it to the player; The multimodal effect feedback information includes text prompts, dynamic visual effects, and / or 3D preview effects.

14. The method according to claim 13, characterized in that, The multimodal effect feedback information also includes: When the effect is determined to be effective, the dynamic process of the effect taking effect, the 3D preview effect of the effective work, and the estimated value information are displayed. If the condition is determined to be ineffective, the specific reasons for the ineffectiveness and suggestions for improvement will be displayed.

15. The method according to any one of claims 1 to 7, characterized in that, The method further includes: Record complete data of this virtual item usage, including the activation command, the current game scene data, the player operation data, the predicted effect result, and the result of the effective operation; The complete data is stored in the player behavior database.

16. The method according to claim 15, characterized in that, When the process of dynamically evaluating the predicted effectiveness of the virtual item calls the multi-dimensional effectiveness condition determination model for evaluation, the method further includes: Based on the complete data of multiple players stored in the player behavior database, the parameters of the multi-dimensional effective condition determination model are optimized by machine learning algorithms. The parameters include the weights and evaluation thresholds of each dimension of data.

17. A device for evaluating the effectiveness of virtual props, characterized in that, include: The data acquisition module is used to respond to the activation command of the virtual item, acquire the item identification information of the virtual item and the current game scene data, and continuously acquire player operation data; The effectiveness evaluation module is used to dynamically evaluate the predicted effectiveness result of the virtual item based on the item identification information, the current game scene data, the player operation data, and a preset reference dataset. The item activation module is used to execute the activation operation of the virtual item based on the predicted activation result.

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the method for evaluating the effectiveness of virtual items as described in any one of claims 1 to 16.

19. A computer device, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the method for evaluating the effectiveness of virtual items as described in any one of claims 1 to 16.

Images