Real-time game platform fight dynamic adjustment method

Abstract

The invention relates to the technical field of real-time game data processing and intelligent matching, and discloses a real-time game platform fight dynamic adjustment method which comprises the following steps: acquiring player attribute data and game session dynamic data in a game platform, performing player skill level evaluation processing based on the player attribute data and the game session dynamic data to generate player skill index data; player attribute data and game session dynamic data are collected in real time, player skill level evaluation and battle matching fairness judgment are carried out, and dynamic matching adjustment is started when matching is judged to be unfair, so that real-time monitoring of player battle states and immediate correction of matching strategies are achieved, the accuracy and fairness of battle matching are improved, and the battle matching efficiency is improved. And matching with the standard parameter combination data to determine target dynamic adjustment algorithm type data so as to generate dynamic adjustment summarized data and execute real-time parameter adjustment, thereby enhancing the pertinence and timeliness of dynamic adjustment of the battle environment.

Description

Technical Field

[0001] This invention relates to the field of real-time game data processing and intelligent matchmaking technology, specifically a method for dynamically adjusting battles on a real-time game platform. Background Technology

[0002] Real-time dynamic adjustment, as a core technology of real-time gaming platforms, plays a crucial role in maintaining the fairness and dynamic balance of game battles, ensuring the competitive experience of players, and protecting the health of the platform ecosystem. Its technology covers multiple dimensions, including real-time parameter calculation and dynamic adjustment of player skill matching, network status synchronization, resource balancing, and behavioral feedback.

[0003] Currently, in the matchmaking process of real-time gaming platforms, the system relies solely on static historical data of players for matching. It cannot continuously assess players' skill levels and dynamically judge the fairness of matchmaking based on real-time dynamic data collected during the game session. Furthermore, it cannot promptly generate adjustment parameters and adaptation algorithms when matchmaking is unfair. At the same time, the lack of a collaborative processing mechanism based on a distributed architecture for the aforementioned data collection, core calculations, and cross-platform interactions leads to lag, deviations, and platform compatibility issues in matchmaking adjustments, affecting the real-time performance and accuracy of match balance and player experience.

[0004] Therefore, a method for dynamic adjustment of real-time game platform battles is proposed to solve the above problems. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method for dynamic adjustment of battles on a real-time game platform. This method solves the problems of lag, deviation, and platform compatibility in battle adjustments mentioned in the background, which affect the real-time performance and accuracy of battle balance and player experience.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for dynamic adjustment of real-time game platform battles, the method comprising the following steps: S1. Collect player attribute data and game session dynamic data from the game platform; S2. Based on the player attribute data and game session dynamic data, perform player skill level assessment processing to generate player skill index data; S3. Based on the player skill index data, perform a fair matchmaking judgment to generate fair matchmaking judgment data; S4. When it is determined that the matching is unfair, dynamic matching adjustment is performed based on the player attribute data and player skill index data to generate dynamic matching adjustment data; S5. Based on the dynamic matching adjustment data and combined with the game session dynamic data, perform battle parameter combination processing to construct battle parameter combination data; S6. Based on the battle parameter combination data and the pre-stored standard parameter combination data, perform matching processing to generate target dynamic adjustment algorithm type data; S7. Integrate the battle parameter combination data, target dynamic adjustment algorithm type data, and game session dynamic data to form dynamic adjustment summary data, and dynamically adjust the game battle parameters accordingly to generate game battle adjustment data. S8. Based on the game battle adjustment data and the collected player feedback data, perform dynamic adjustment effect evaluation and parameter optimization processing to generate adjustment effect evaluation data; S9, based on a distributed architecture, collects player attribute data and game session dynamic data, performs core calculations for player skill level assessment and dynamic matching adjustment, and interacts with multiple game platforms to adjust data.

[0007] Preferably, the step S1 of collecting player attribute data and game session dynamic data from the game platform includes the following steps: S11. Collect player registration information and historical behavior data through the game platform server interface to obtain the player attribute data; S12. During the game session, the monitoring module embedded in the client collects network status, operation command stream and resource consumption information to obtain the game session dynamic data. S13. Preprocess the collected player attribute data and game session dynamic data, including data cleaning, normalization and denoising operations, to generate a standardized initial dataset.

[0008] Preferably, the generation of player skill index data in S2 includes the following steps: S21. Extract feature vectors for skill assessment from the standardized initial dataset. The feature vectors include historical win / loss records, number of successful operations per unit time, and tactical execution success rate. S22. Input the feature vector into a pre-trained player skill evaluation model, and the model outputs a quantitative score corresponding to the current player's skill level; S23. Fine-tune the quantitative score based on the real-time performance data of the current session to generate the player skill index data, which includes core skill values ​​and dynamic correction values.

[0009] Preferably, the generation of matchmaking fairness judgment data in S3 includes the following steps: S31. For a queue of players to be matched, calculate the overall skill dispersion of the queue based on the player skill index data of all players in the queue. S32. Compare the skill dispersion with a preset fairness threshold range; S33. When the skill dispersion is within the fairness threshold range, generate the matchmaking fairness judgment data indicating fair matchmaking; S34. When the skill dispersion exceeds the fairness threshold range, generate the matchmaking fairness judgment data indicating unfair matchmaking.

[0010] Preferably, generating dynamic matching adjustment data in step S4 includes the following steps: S41. When it is determined that the match is not fair, calculate the match priority weight of each player based on the player skill index data; S42. Based on the matching priority weight and player queue status, reorder and schedule candidate players in the matching pool across pools. S43. Generate a new matching scheme based on the scheduling result, use it as the dynamic matching adjustment data, and update the matching queue.

[0011] Preferably, the construction of battle parameter combination data in S5 includes the following steps: S51. Obtain the dynamic matching adjustment data as the current baseline parameter set; S52. Align and merge the baseline parameter set with the game session dynamic data according to the time window; S53. Encode the fused multidimensional data to construct the battle parameter combination data that represents the current battle scenario.

[0012] Preferably, the step S6 of generating target dynamic adjustment algorithm type data includes the following steps: S61. Predefine multiple dynamic adjustment algorithm types and configure a corresponding standard parameter combination template for each algorithm type; S62. Calculate the matching degree between the battle parameter combination data and each of the standard parameter combination templates; S63. Select the dynamic adjustment algorithm type corresponding to the standard parameter combination template with the highest matching degree as the target dynamic adjustment algorithm type data.

[0013] Preferably, generating game battle adjustment data in S7 includes the following steps: S71. Package the battle parameter combination data, the target dynamic adjustment algorithm type data, and the related game session dynamic data to create the dynamic adjustment summary data; S72. Based on the algorithm identifier specified in the target dynamic adjustment algorithm type data, call the corresponding dynamic adjustment execution engine; S73. The dynamic adjustment execution engine parses the dynamic adjustment summary data, and calculates and applies new game parameter configurations in real time to generate the game battle adjustment data.

[0014] Preferably, generating adjustment effect evaluation data in step S8 includes the following steps: S81. After the game session ends and within a preset period, collect player feedback data, including subjective ratings, mid-game exit records in the behavior log, and complaint information. S82. Perform correlation analysis between the game battle adjustment data and the player feedback data, evaluate the actual effect of this dynamic adjustment, and generate an adjustment effect evaluation report; S83. Based on the adjustment effect evaluation report, dynamically correct and optimize the algorithm matching logic and the fairness threshold range.

[0015] Preferably, adjusting the data interaction in S9 includes the following steps: S91. By deploying computing nodes at the network edge, collect player attribute data and game session dynamic data, and align the collected data format with the player feedback data on which the evaluation is based, in order to support subsequent correlation analysis; S92. The core calculations for player skill level assessment and dynamic matching adjustment are performed by the cloud computing platform, wherein the parameters of the core calculations are dynamically optimized based on the optimization criteria generated by the assessment. S93. Through a unified application programming interface, adjust the data interaction with multiple game platforms, and adapt the content and format of the interacted data according to the optimized core calculation parameters.

[0016] Compared with existing technologies, this invention provides a method for dynamic adjustment of real-time game platform battles, which has the following beneficial effects: 1. In this invention, by collecting player attribute data and game session dynamic data in real time, the system evaluates player skill levels and judges the fairness of matchmaking. When the matchmaking is determined to be unfair, dynamic matchmaking adjustment is initiated, thereby realizing real-time monitoring of player battle status and instant correction of matchmaking strategies, thus improving the accuracy and fairness of matchmaking.

[0017] 2. In this invention, battle parameter combination data is constructed based on dynamic matching adjustment data and game session dynamic data, and matched with standard parameter combination data to determine the target dynamic adjustment algorithm type data. Then, dynamic adjustment summary data is generated and real-time parameter adjustment is performed, forming a closed-loop control mechanism from data matching to algorithm invocation, which enhances the pertinence and timeliness of dynamic adjustment of the battle environment.

[0018] 3. In this invention, data collection, core computing and cross-platform data interaction are carried out collaboratively based on a distributed architecture, and dynamic adjustment effect evaluation and parameter optimization are completed by combining player feedback data. This achieves unified coordination and continuous optimization of the entire battle adjustment process, and improves the overall system response efficiency and multi-platform collaborative adaptation capabilities. Attached Figure Description

[0019] Figure 1 This is a flowchart of the real-time game platform dynamic adjustment method of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] For specific implementation examples, please refer to: Figure 1 A method for dynamically adjusting real-time game platform battles, characterized by the following steps: S1. Collect player attribute data and game session dynamic data from the game platform; S2. Based on player attribute data and game session dynamic data, evaluate and process player skill level to generate player skill index data; S3. Based on player skill indicator data, perform matchmaking fairness judgment and generate matchmaking fairness judgment data; S4. When it is determined that the matchmaking is unfair, dynamic matchmaking adjustments are made based on player attribute data and player skill index data, and dynamic matchmaking adjustment data is generated. S5. Based on the dynamic matching adjustment data and combined with the game session dynamic data, perform battle parameter combination processing to construct battle parameter combination data; S6. Based on the combination data of battle parameters and the pre-stored standard parameter combination data, the target dynamic adjustment algorithm type data is generated by matching the battle parameter combination data with the pre-stored standard parameter combination data. S7. Integrate battle parameter combination data, target dynamic adjustment algorithm type data, and game session dynamic data to form dynamic adjustment summary data, and dynamically adjust game battle parameters accordingly to generate game battle adjustment data. S8. Based on game battle adjustment data and collected player feedback data, conduct dynamic adjustment effect evaluation and parameter optimization processing, and generate adjustment effect evaluation data; S9, based on a distributed architecture, collects player attribute data and game session dynamic data, performs core calculations for player skill level assessment and dynamic matching adjustment, and interacts with multiple game platforms to adjust data.

[0022] The steps involved in collecting player attribute data and game session dynamic data from the game platform in S1 are as follows: S11. Collect player registration information and historical behavior data through the game platform server interface to obtain player attribute data; S12. During the game session, the monitoring module embedded in the client collects network status, operation command stream and resource consumption information to obtain dynamic data of the game session. S13. Preprocess the collected player attribute data and game session dynamic data, including data cleaning, normalization, and noise reduction, to generate a standardized initial dataset. This includes the following steps: Data cleaning includes the following steps: S131. Identify and remove abnormal registration information and historical behavior data that exceed reasonable limits from player attribute data; ; in, Represents a player's attribute data point. Represents the mean. Represents standard deviation; S132. Detect and fix packet loss and out-of-order issues in game session dynamic data caused by network transmission failure; S133. For missing values ​​in player attribute data and game session dynamic data, an imputation algorithm based on the historical mean of similar players is used to fill in the missing values. The specific calculation formula is as follows: ; in, This represents the player attribute value obtained after filling in the missing values. This represents the number of players whose attributes are similar to those of the player with the missing value. Representing the Corresponding attribute values ​​for similar players; Normalization includes the following steps: S134. Using the minimum-maximum scaling algorithm, player attribute data and game session dynamic data of different dimensions are linearly transformed to the [0,1] interval; ; in, This represents the normalized player attribute values ​​and dynamic data values ​​from the game session. Represents the original data value. and These represent the maximum and minimum values ​​in the feature data column, respectively. S135. For data features that follow a normal distribution, the Z-score standardization algorithm is used to transform them into a standard distribution with a mean of 0 and a standard deviation of 1. The standardization algorithm calculation formula is as follows: ; in, For standardized data values, The original data values, The mean, Standard deviation; The noise reduction process includes the following steps: S136. Apply a sliding window-based mean filtering algorithm to smooth high-frequency noise in dynamic game session data caused by instantaneous network jitter. The specific calculation formula is as follows: ; in, Represents the moment Data values ​​after mean filtering and smoothing. Represents the first in the window The original data values ​​at each moment. This represents the duration of the sliding window. This represents the current moment when the filtering calculation is being performed. S137. Using wavelet transform threshold denoising algorithm, separate and suppress random fluctuation interference in player attribute data, specifically including the following steps: S1371. Perform discrete wavelet decomposition on the player attribute data sequence containing random fluctuations to obtain the approximation coefficients and detail coefficients of each layer. The decomposition is calculated using the following formula: ; in, This represents the player attribute data sequence to be denoised. For indexing data points, The preset total number of wavelet decomposition layers. For the first Approximate coefficients obtained from layer wavelet decomposition For the first Detail coefficients obtained from layer wavelet decomposition and They are respectively through scale Peaceful relocation The scaling function and wavelet function obtained after transformation; S1372. Apply a soft thresholding function to each layer of detail coefficients to suppress random fluctuation noise. The soft thresholding is calculated using the following formula: ; in, The detail coefficients after thresholding. For symbolic functions, For the first The threshold for layer decomposition, For the first Detail coefficients obtained from layer wavelet decomposition It is the maximum value; S1373. Wavelet reconstruction is performed using the detail coefficients after thresholding and the retained approximation coefficients to obtain the denoised player attribute data sequence. The reconstruction is calculated using the following inverse transform formula: ; in, The output sequence of player attribute data after denoising. For the first Approximate coefficients obtained from layer wavelet decomposition and They are respectively through scale Peaceful relocation The scaling function and wavelet function obtained after the transformation The detail coefficients are the result of thresholding.

[0023] Generating player skill metric data in S2 includes the following steps: S21. Extract feature vectors for skill assessment from the standardized initial dataset. The feature vectors include historical win / loss records, number of successful operations per unit time, and tactical execution success rate. S22. Input the feature vector into the pre-trained player skill evaluation model. The model outputs a quantitative score corresponding to the current player's skill level. This includes the following steps: S221. Extract feature vectors of a large number of players and their corresponding final battle results from historical game logs as a training sample set. ; in, Represents the total number of training samples. This represents the total number of player samples drawn from historical logs. Representing the The feature vector of each player Representative and The corresponding final battle result quantification value; S222. Construct an initial evaluation model with a multilayer perceptron and gradient boosting decision tree as its core architecture. Its core mathematical definitions are as follows: The multilayer perceptron model is defined through multilayer nonlinear transformations: ; ; in, This represents the total number of layers in the neural network. For the first The activation vector of the layer, and The first Layer weight matrix and bias vector, For the first The activation function of the layer, Represents the set of all trainable parameters. This is the predicted value for the final skill score; The gradient boosting decision tree model is defined using an additive model: ; in, For player feature vectors Using the input as input, the predicted output value of the player's skill level is obtained after calculation through the gradient boosting decision tree model. The total number of base learners, For the initial prediction, Representing the A decision tree, For parameters, The learning rate; S223. Use the training sample set to perform supervised training on the initial evaluation model, and iteratively optimize the model parameters through the backpropagation algorithm until the model prediction accuracy converges. ; in, Representative model for samples The predicted output, To measure the difference between predicted values ​​and true labels The loss function for the difference between them. For regularization terms, The regularization coefficient is . For loss; S23. Fine-tune the quantitative score based on the real-time performance data of the current session to generate player skill index data that includes core skill values ​​and dynamic correction values.

[0024] The steps involved in generating matchmaking fairness assessment data in S3 are as follows: S31. For a queue of players to be matched, calculate the overall skill dispersion of the queue based on the skill index data of all players in the queue. S32. Compare the skill dispersion with a preset fairness threshold range; S33. When the skill dispersion is within the fairness threshold range, generate matchmaking fairness judgment data to indicate fair matchmaking. S34. When the skill dispersion exceeds the fairness threshold range, generate matchmaking fairness judgment data indicating unfair matchmaking.

[0025] Generating dynamic matching adjustment data in S4 includes the following steps: S41. When it is determined that the matchmaking is unfair, calculate the matchmaking priority weight of each player based on the player skill index data; S42. Based on the matching priority weight and player queue status, reorder and schedule candidate players in the matching pool across pools. S43. Generate a new matching scheme based on the scheduling results, use it as dynamic matching adjustment data, and update the matching queue.

[0026] The steps to construct battle parameter combination data in S5 are as follows: S51. Obtain dynamic matching adjustment data as the current baseline parameter set; S52. Align and merge the baseline parameter set with the game session dynamic data according to the time window, specifically including the following steps: S521. Using the timestamps of the game session dynamic data as a reference, the nearest neighbor interpolation method is used to align the reference parameter set to the same time axis. The specific calculation formula is as follows: ; in, Represents alignment to The reference parameter value at time, Representing the Timestamps of dynamic data from each game session. This represents the original set of timestamps for the baseline parameter set. This represents the original set of timestamps for the baseline parameter set; S522. The aligned baseline parameter set and the game session dynamic data are concatenated according to the feature dimension to form an enhanced feature vector. S523. Perform principal component analysis on the enhanced feature vectors and reduce their dimensionality to obtain fused data that represents the current battle scenario; ; in, Represents a low-dimensional feature matrix. Represents the original enhanced eigenvector matrix. The projection matrix represents the eigenvectors. S53. Encode the fused multidimensional data to construct a combination of battle parameters that characterizes the current battle scenario.

[0027] The steps involved in generating target dynamic adjustment algorithm type data in S6 are as follows: S61. Predefine multiple dynamic adjustment algorithm types and configure a corresponding standard parameter combination template for each algorithm type; S62. Calculate the matching degree between the battle parameter combination data and each standard parameter combination template, specifically including the following steps: S621. Calculate the Euclidean distance between the battle parameter combination data and the corresponding feature dimension of each standard parameter combination template. ; in, Representative battle parameter combination data With the Standard parameter combination template The Euclidean distance between them The dimension of the parameter vector. represent The 3D eigenvalues Representing the Standard template The 3D eigenvalues; S622. Map the reciprocal of the Euclidean distance to a similarity score in the [0,1] interval, and use it as the matching degree; ; in, This represents the similarity score calculated based on Euclidean distance. Representative battle parameter combination data With the Standard parameter combination template The Euclidean distance between them; S63. Select the dynamic adjustment algorithm type corresponding to the standard parameter combination template with the highest matching degree as the target dynamic adjustment algorithm type data.

[0028] Generating game match adjustment data in S7 includes the following steps: S71. Package the battle parameter combination data, target dynamic adjustment algorithm type data, and related game session dynamic data into dynamic adjustment summary data; S72. Based on the algorithm identifier specified in the target dynamic adjustment algorithm type data, call the corresponding dynamic adjustment execution engine; S73: The dynamic adjustment execution engine parses the dynamically adjusted summary data and calculates and applies new game parameter configurations in real time to generate game battle adjustment data.

[0029] The steps involved in generating adjustment effect evaluation data in S8 are as follows: S81. Collect player feedback data after the game session ends and within a preset period. The feedback data includes subjective ratings, mid-game exit records in the behavior log, and complaint information. S82. Correlate and analyze the game battle adjustment data with player feedback data to evaluate the actual effect of this dynamic adjustment and generate an adjustment effect evaluation report, which includes the following steps: S821. Calculate the Pearson correlation coefficient between the changes in key parameters in game battle adjustment data and the satisfaction ratings in player feedback data. ; in, This represents the total number of data sample pairs used in the association analysis. Representative at the The values ​​in each sample Represents the corresponding number Player satisfaction ratings in the sample Representing all Key parameters in each sample The average value, Representing all Satisfaction rating in each sample The average value, Represents the Pearson correlation coefficient; S822. When the absolute value of the correlation coefficient is greater than the preset threshold, it is determined that the key parameter is strongly correlated with player satisfaction, and it is used as the core basis for evaluating the effect of this adjustment. ; in, Pearson correlation coefficient The absolute value, For a pre-set threshold; S83. Based on the adjustment effect evaluation report, dynamically correct and optimize the algorithm matching logic and fairness threshold range.

[0030] Adjusting data interaction in S9 includes the following steps: S91. By deploying computing nodes at the network edge, collect player attribute data and game session dynamic data, and align the collected data format with the player feedback data on which the evaluation is based, in order to support subsequent correlation analysis; S92. The core calculations for evaluating player skill levels and dynamically adjusting matchmaking are performed by a cloud computing platform. The parameters of the core calculations are dynamically optimized based on the optimization criteria generated by the evaluation. S93. Through a unified application programming interface, it adjusts data interaction with multiple game platforms, and the content and format of the interacted data are adapted according to the optimized core calculation parameters.

[0031] The steps for implementing the real-time game platform's dynamic battle adjustment method are as follows: Step 1: Data Collection and Standardization Processing First, data collection and standardization are performed. Player attribute data is collected through the game platform server interface, while game session dynamic data is collected in real time through the monitoring module embedded in the client. Then, data preprocessing is performed, including data cleaning, normalization and noise reduction, to generate a standardized initial dataset, providing a high-quality data foundation for subsequent analysis. Specifically, outliers are identified and removed, transmission faults are repaired, missing values ​​are filled, and the minimum-maximum scaling and Z-score algorithms are used to complete data normalization.

[0032] Step Two: Player Skill Level Assessment and Processing Based on standardized data, the process moves to the player skill level assessment stage. Feature vectors are extracted from the dataset and input into a pre-trained player skill assessment model. This model uses a multilayer perceptron and gradient boosting decision tree architecture to quantify and score player skills, ultimately generating player skill index data that includes core skill values ​​and dynamic correction values. The model training process adopts a supervised learning approach, and the accuracy of the assessment is ensured by optimizing the loss function parameters.

[0033] Step 3: Handling the fairness of matchmaking Based on the skill assessment results, the matchmaking fairness judgment process is initiated. The skill dispersion of the player queue is calculated and compared with a preset fairness threshold range. When the dispersion is within the threshold range, matchmaking fairness judgment data indicating fair matchmaking is generated. When it exceeds the threshold, matchmaking unfair judgment data is generated, triggering the subsequent adjustment mechanism.

[0034] Step 4: Dynamic Matching and Adjustment Processing When an unfair match is determined, dynamic matchmaking adjustment is performed. The matchmaking priority weight of each player is calculated based on the player's skill index data. Based on the weight and the real-time player queue status, the matchmaking pool is reordered and cross-pool scheduling is performed to generate a new matchmaking scheme as dynamic matchmaking adjustment data, and the matchmaking queue is updated synchronously.

[0035] Step 5: Processing Battle Parameter Combinations The next step is to process the battle parameter combination. The dynamically matched adjustment data is used as the baseline parameter set and is aligned and merged with the game session dynamic data according to the time window. Time synchronization is achieved by the nearest neighbor interpolation method, and the merged multidimensional data is encoded to construct battle parameter combination data that represents the current battle scene.

[0036] Step Six: Target Algorithm Type Matching Process In the target algorithm type matching process, the matching degree of the battle parameter combination data and the pre-stored standard parameter combination template is calculated. The similarity is calculated by the Euclidean distance algorithm, and the template with the highest matching degree is selected to correspond to the algorithm type. The target dynamically adjusts the algorithm type data, thereby realizing the intelligent selection of the algorithm.

[0037] Step 7: Dynamically Adjust Execution Processing Next, dynamic adjustment execution processing is performed, integrating battle parameter combination data, target dynamic adjustment algorithm type data, and game session dynamic data to form dynamic adjustment summary data. Based on the algorithm identifier, the corresponding dynamic adjustment execution engine is called to parse the data and calculate and apply new game parameter configurations in real time to generate game battle adjustment data.

[0038] Step 8: Evaluation and Processing of Adjustment Effects After the adjustment is completed, the adjustment effect evaluation process is initiated. Player feedback data is collected and correlated with game battle adjustment data. The adjustment effect is evaluated by calculating the Pearson correlation coefficient. When the absolute value of the correlation coefficient exceeds the preset threshold, it is determined to be a strong correlation. An adjustment effect evaluation report is generated, and the algorithm matching logic and fairness threshold range are dynamically optimized and iterated based on this.

[0039] Step Nine: Distributed Architecture Collaborative Processing Finally, with the support of distributed architecture collaborative processing, the entire system achieves full-process collaboration based on distributed architecture. Data collection is completed through edge computing nodes, core computing is performed by the cloud computing platform, and cross-platform adjustment data interaction is achieved through a unified application programming interface, ensuring the efficient operation of the system and its multi-platform adaptability.

[0040] 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.

[0041] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for dynamically adjusting real-time game platform battles, characterized by: The method includes the following steps: S1. Collect player attribute data and game session dynamic data from the game platform; S2. Based on the player attribute data and game session dynamic data, perform player skill level assessment processing to generate player skill index data; S3. Based on the player skill index data, perform a fair judgment on matchmaking to generate fair judgment data for matchmaking. S4. When it is determined that the matching is unfair, dynamic matching adjustment is performed based on the player attribute data and player skill index data to generate dynamic matching adjustment data; S5. Based on the dynamic matching adjustment data and combined with the game session dynamic data, perform battle parameter combination processing to construct battle parameter combination data; S6. Based on the battle parameter combination data and the pre-stored standard parameter combination data, perform matching processing to generate target dynamic adjustment algorithm type data; S7. Integrate the battle parameter combination data, target dynamic adjustment algorithm type data, and game session dynamic data to form dynamic adjustment summary data, and dynamically adjust the game battle parameters accordingly to generate game battle adjustment data. S8. Based on the game battle adjustment data and the collected player feedback data, perform dynamic adjustment effect evaluation and parameter optimization processing to generate adjustment effect evaluation data; S9, based on a distributed architecture, collects player attribute data and game session dynamic data, performs core calculations for player skill level assessment and dynamic matching adjustment, and interacts with multiple game platforms to adjust data.

2. The method for dynamic adjustment of real-time game platform battles according to claim 1, characterized in that: The steps involved in S1 to collect player attribute data and game session dynamic data from the game platform are as follows: S11. Collect player registration information and historical behavior data through the game platform server interface to obtain the player attribute data; S12. During the game session, the monitoring module embedded in the client collects network status, operation command stream and resource consumption information to obtain the game session dynamic data. S13. Preprocess the collected player attribute data and game session dynamic data, including data cleaning, normalization and denoising operations, to generate a standardized initial dataset.

3. The method for dynamic adjustment of real-time game platform battles according to claim 2, characterized in that: The process of generating player skill indicator data in S2 includes the following steps: S21. Extract feature vectors for skill assessment from the standardized initial dataset. The feature vectors include historical win / loss records, number of successful operations per unit time, and tactical execution success rate. S22. Input the feature vector into a pre-trained player skill evaluation model, and the model outputs a quantitative score corresponding to the current player's skill level; S23. Fine-tune the quantitative score based on the real-time performance data of the current session to generate the player skill index data, which includes core skill values ​​and dynamic correction values.

4. The method for dynamic adjustment of real-time game platform battles according to claim 3, characterized in that: The steps involved in generating matchmaking fairness assessment data in S3 are as follows: S31. For a queue of players to be matched, calculate the overall skill dispersion of the queue based on the player skill index data of all players in the queue. S32. Compare the skill dispersion with a preset fairness threshold range; S33. When the skill dispersion is within the fairness threshold range, generate the matchmaking fairness judgment data indicating fair matchmaking; S34. When the skill dispersion exceeds the fairness threshold range, generate the matchmaking fairness judgment data indicating unfair matchmaking.

5. The method for dynamic adjustment of real-time game platform battles according to claim 4, characterized in that: The process of generating dynamic matching adjustment data in S4 includes the following steps: S41. When it is determined that the match is not fair, calculate the match priority weight of each player based on the player skill index data; S42. Based on the matching priority weight and player queue status, reorder and schedule candidate players in the matching pool across pools. S43. Generate a new matching scheme based on the scheduling result, use it as the dynamic matching adjustment data, and update the matching queue.

6. The method for dynamic adjustment of real-time game platform battles according to claim 5, characterized in that: The steps involved in constructing the battle parameter combination data in S5 are as follows: S51. Obtain the dynamic matching adjustment data as the current baseline parameter set; S52. Align and merge the baseline parameter set with the game session dynamic data according to the time window; S53. Encode the fused multidimensional data to construct the battle parameter combination data that represents the current battle scenario.

7. The method for dynamic adjustment of real-time game platform battles according to claim 6, characterized in that: The step of generating target dynamic adjustment algorithm type data in S6 includes the following steps: S61. Predefine multiple dynamic adjustment algorithm types and configure a corresponding standard parameter combination template for each algorithm type; S62. Calculate the matching degree between the battle parameter combination data and each of the standard parameter combination templates; S63. Select the dynamic adjustment algorithm type corresponding to the standard parameter combination template with the highest matching degree as the target dynamic adjustment algorithm type data.

8. The method for dynamic adjustment of real-time game platform battles according to claim 7, characterized in that: The process of generating game battle adjustment data in S7 includes the following steps: S71. Package the battle parameter combination data, the target dynamic adjustment algorithm type data, and the related game session dynamic data to create the dynamic adjustment summary data; S72. Based on the algorithm identifier specified in the target dynamic adjustment algorithm type data, call the corresponding dynamic adjustment execution engine; S73. The dynamic adjustment execution engine parses the dynamic adjustment summary data, and calculates and applies new game parameter configurations in real time to generate the game battle adjustment data.

9. The method for dynamic adjustment of real-time game platform battles according to claim 8, characterized in that: The step of generating adjustment effect evaluation data in S8 includes the following steps: S81. After the game session ends and within a preset period, collect player feedback data, including subjective ratings, mid-game exit records in the behavior log, and complaint information. S82. Perform correlation analysis between the game battle adjustment data and the player feedback data, evaluate the actual effect of this dynamic adjustment, and generate an adjustment effect evaluation report; S83. Based on the adjustment effect evaluation report, dynamically correct and optimize the algorithm matching logic and the fairness threshold range.

10. The method for dynamic adjustment of real-time game platform battles according to claim 9, characterized in that: The adjustment of data interaction in S9 includes the following steps: S91. By deploying computing nodes at the network edge, collect player attribute data and game session dynamic data, and align the collected data format with the player feedback data on which the evaluation is based, in order to support subsequent correlation analysis; S92. The core calculations for evaluating player skill levels and dynamically adjusting matchmaking are performed by a cloud computing platform, wherein the parameters of the core calculations are dynamically optimized based on the optimization criteria generated by the evaluation. S93. Through a unified application programming interface, adjust the data interaction with multiple game platforms, and adapt the content and format of the data to be interacted with according to the optimized core calculation parameters.

Images