Disclosure of Invention
Therefore, the invention provides a game data mining method which is used for solving the problem of low efficiency of optimizing the performance of a virtual reality game in the prior art.
In order to achieve the above object, the present invention provides a game data mining method, comprising:
step S1, collecting virtual reality game performance data;
step S2, preprocessing the virtual reality game performance data to obtain preprocessed data;
s3, analyzing the bottleneck factors of the virtual reality game performance according to the preprocessing data;
s4, inputting the bottleneck factors of the virtual reality game performance into an optimization decision model, and outputting a virtual reality game performance optimization strategy;
step S5, judging the use condition of the user behavior data according to the repetition times of the user behavior data, and correcting the virtual reality game performance optimization process according to the judgment result;
and S6, judging the relevance between the use condition of the user behavior data and the bottleneck factor of the virtual reality game performance according to the relevance rule mining algorithm, and adjusting the optimization process of correcting the virtual reality game performance according to the judging result.
Further, in the step S2, the preprocessing method includes:
step S20, data cleaning is carried out on the virtual reality game performance data to obtain effective data;
step S21, dividing the effective data into image data and data point data;
Step S22, carrying out standardization processing on the data point data through a standardization algorithm to obtain standardized data point data;
And S23, carrying out feature extraction on the image data and the standardized data point data, and combining the data after feature extraction to obtain preprocessing data.
Further, in the step S20, when the virtual reality game performance data is cleaned, the missing value of the virtual reality game performance data is filled by interpolation, the abnormal value of the virtual reality game performance data is detected by the IQR method, the representative information of the abnormal value is collected, the importance of the representative information of the abnormal value is determined, and the abnormal value is deleted and marked according to the determination result, so as to obtain effective data, wherein:
when the representative information of the outlier is important information, reserving and marking the outlier;
When the representative information of the outlier is unimportant information, the outlier is deleted.
Further, in the step S21, the effective data stored in the form of pixels is divided into image data according to the storage form of the effective data, and the effective data stored in the form of numerical values is divided into data point data;
In the step S22, the data point data is normalized by a Z-score normalization method in a normalization algorithm, and a calculation formula of the normalization algorithm is set to be z= (X- μ)/σ, where X is the data point data, μ is the mean value of the data point data, σ is the standard deviation of the data point data, and Z is the normalized data point data;
The calculation formula of the mean value mu is set asWhere N is the number of data points in the data point data, X i is the i-th data point in the data point data, e.g., the first data point in the data point data is X1 and the second data point is X2;
Setting the calculation formula of standard deviation sigma as
Further, in the step S23, feature extraction is performed on the image data by a Canny edge detection method to obtain feature image data, where the Canny edge detection method includes:
Step G100, removing noise from the image data by using Gaussian filtering to obtain noise-removed image data;
step G200, calculating the gradient direction of the image data after noise removal by using a Sobe l operator to obtain the gradient direction of the image data;
Step G300, performing non-maximum suppression in the gradient direction of the image data by using a non-maximum suppression method, and reserving local maximum points to obtain the maximum point distribution of the image data;
step G400, judging each point in the maximum point distribution of the image data by using a double-threshold processing method, performing edge connection according to a judging result, setting the value of each maximum point as V, the maximum threshold as Vmax and the minimum threshold as Vmin, wherein:
when V > Vmax, judging the maximum point as a strong edge;
When Vmin is less than or equal to V is less than or equal to Vmax, judging the maximum point as a weak edge;
When V < Vmin, judging the maximum value point as a non-edge;
and G500, connecting the strong edge and the weak edge, and outputting the connected image as characteristic image data.
Further, in the step S23, feature extraction is performed on the normalized data points by a principal component analysis method to obtain feature data points, the feature image data and the feature data points are input into a depth feature fusion model for merging, and the depth feature fusion model outputs the preprocessing data.
Further, in the step S3, when the bottleneck factors of the performance of the virtual reality game are analyzed according to the preprocessing data, the preprocessing data are input into the bottleneck factor analysis model, and the bottleneck factor analysis model is used for outputting the bottleneck factors of the performance of the virtual reality game.
Further, in the step S4, an optimization decision model is constructed through an expert optimization strategy data set, and a virtual reality game performance bottleneck factor is input into the optimization decision model, so as to output a virtual reality game performance optimization strategy.
Further, in the step S5, user behavior data is obtained, the number of repetitions J of the user behavior data is compared with the number of repetitions J0 of the preset user behavior data, the use condition of the user behavior data is judged according to the comparison result, and the condition that the performance optimization process of the virtual reality game meets the standard is judged and corrected according to the judgment result, wherein:
When J < J0, judging that the user behavior data is low-frequency user behavior data, wherein the virtual reality game performance optimization process reaches the standard, and correcting the virtual reality game performance optimization process is not performed:
when J is more than or equal to J0, judging that the user behavior data is high-frequency user behavior data, correcting the virtual reality game performance optimization process, pushing an expert optimization strategy data set update window to a virtual reality game equipment manager, updating the expert optimization strategy data set by the virtual reality game equipment manager, acquiring an updated expert optimization strategy data set, and optimizing the virtual reality game performance according to the updated expert optimization strategy data set.
Further, in the step S6, when judging the relevance between the usage of the user behavior data and the bottleneck factor of the virtual reality game according to the association rule mining algorithm, calculating the confidence coefficient P between the usage of the user behavior data and the bottleneck factor of the virtual reality game according to the association rule mining algorithm, comparing the confidence coefficient P with a preset confidence coefficient P0, judging the relevance between the usage of the user behavior data and the bottleneck factor of the virtual reality game according to the comparison result, and adjusting the correction process according to the judgment result, wherein:
when P < P0, judging that the relevance between the use condition of the user behavior data and the bottleneck factor of the virtual reality game performance is low, and not adjusting the optimization process of correcting the virtual reality game performance;
When P is more than or equal to P0, judging that the relevance between the use condition of the user behavior data and the bottleneck factor of the virtual reality game performance is high, adjusting the correction process, setting an adjustment coefficient Q, wherein Q=1+ (P-P0), adjusting the preset user behavior data repetition number J0 according to the adjustment coefficient Q, and setting the preset user behavior data repetition number after adjustment to be Jq0, wherein Jq0=J0×Q.
Compared with the prior art, the method has the advantages that the method collects the virtual reality game performance data through the step S1, comprehensively knows key indexes such as running state, resource consumption, frame rate, delay and the like of the game, provides data support for performance optimization, pre-processes the virtual reality game performance data through the step S2 to obtain pre-processed data, eliminates noise, errors or inconsistency in original data, improves data quality, thereby more accurately reflecting the real situation of the game performance, provides more reliable basis for subsequent analysis and optimization, analyzes the bottleneck factors of the virtual reality game performance according to the pre-processed data through the step S3 so as to accurately obtain key factors which lead to the reduction of the game performance, inputs the bottleneck factors of the virtual reality game performance into an optimization decision model through the step S4, outputs the optimization strategy of the virtual reality game performance, correlates the performance factors with the optimization strategy to the corresponding optimization strategy recommended for different, further improves the optimization efficiency of the virtual reality game performance, repeatedly reflects the real situation of the game performance according to the step S5, judges the bottleneck factors of the virtual reality game performance according to the pre-processed data, makes the correlation result of the virtual reality game performance data according to the step S6, makes the correlation result of the virtual reality game performance data more correct the user performance and the bottleneck factors according to the correlation rule, and the optimization result is more relevant to the optimization rule, and the user performance is judged to be closest to the optimization result, and the performance is optimized to the user is required to the optimization process, and the performance is optimized to the optimization model is improved by the optimization algorithm, the influence mechanism of the user behavior data use condition and the virtual reality game performance bottleneck factor is disclosed, deeper insight is provided for the formulation of the optimization strategy, and the effectiveness and pertinence of the optimization strategy are further improved by adjusting the optimization process to adapt to the relevance, so that the optimization efficiency of the virtual reality game performance is further improved.
In particular, the step S20 fills in missing values of the virtual reality game performance data by interpolation, detects the missing values of the virtual reality game performance data by IQR, retains the missing values of the representative information as important information, and deletes the missing values of the representative information as unimportant information to improve the data quality, thereby reflecting the actual situation of the game performance more accurately.
In particular, the step S21 divides the effective data stored in the form of pixels into image data according to the storage form of the effective data, and divides the effective data stored in the form of numerical values into data point data in order to improve efficiency of management and utilization of the virtual reality game performance data.
In particular, in the step S22, when the data point data is normalized by the Z-score normalization method in the normalization algorithm, a calculation formula of the normalization algorithm is set to improve the data comparability and enhance the data analysis accuracy, thereby improving the optimization efficiency of the virtual reality game performance.
In particular, the step S23 performs feature extraction on the image data step by using a Canny edge detection method, connects the strong edge and the weak edge, retains the connected image, and deletes the isolated edge points, so as to improve the accuracy and parameter adjustability of the image data.
In particular, the step S23 uses a principal component analysis method to perform feature extraction on the standardized data point data, so as to reduce the dimension of the data, reduce the calculation cost, facilitate the subsequent data processing and analysis, and input the feature image data and the feature data point data into a depth feature fusion model for merging, so that the virtual reality game performance is optimized by using the key information in the two data at the same time, thereby further improving the optimization efficiency of the virtual reality game performance.
In particular, the step S3 inputs the preprocessing data into the bottleneck factor analysis model, and outputs the performance bottleneck factor of the virtual reality game through the bottleneck factor analysis model, so as to accurately position the performance bottleneck, thereby improving the efficiency of problem solving, avoiding the waste of resources, and improving the overall cost benefit.
In particular, the step S4 constructs an optimization decision model through the expert optimization strategy data set and outputs the virtual reality game performance optimization strategy to realize targeted optimization, so that the optimization efficiency of the virtual reality game performance is improved, and the optimization decision model supports rapid iteration to further improve the optimization efficiency of the virtual reality game performance.
In particular, when the number of repeating actions of the data is greater than or equal to the preset number of repeating actions of the user, the step S5 updates the expert optimization strategy data set by pushing the expert optimization strategy data set update window to the virtual reality game device administrator, so as to improve the instantaneity and accuracy of the optimization strategy and further improve the optimization efficiency of the virtual reality game performance.
In particular, when the confidence coefficient of the user behavior data usage situation and the virtual reality game performance bottleneck factor is greater than the preset confidence coefficient in step S6, the setting adjustment coefficient is increased along with the increase of the difference between the confidence coefficient and the preset confidence coefficient, so as to adjust the change of the repetition number of the preset user behavior data when the relevance between the user behavior data usage situation and the virtual reality game performance bottleneck factor is strong, thereby further improving the sensitivity of the game performance optimization process and the optimization efficiency of the virtual reality game performance.
Detailed Description
The invention will be further described with reference to examples for the purpose of making the objects and advantages of the invention more apparent, it being understood that the specific examples described herein are given by way of illustration only and are not intended to be limiting.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.
Referring to fig. 1, a flow chart of a method for mining game data according to the present embodiment is shown, where the method includes:
step S1, collecting virtual reality game performance data;
step S2, preprocessing the virtual reality game performance data to obtain preprocessed data;
s3, analyzing the bottleneck factors of the virtual reality game performance according to the preprocessing data;
s4, inputting the bottleneck factors of the virtual reality game performance into an optimization decision model, and outputting a virtual reality game performance optimization strategy;
step S5, judging the use condition of the user behavior data according to the repetition times of the user behavior data, and correcting the virtual reality game performance optimization process according to the judgment result;
and S6, judging the relevance between the use condition of the user behavior data and the bottleneck factor of the virtual reality game performance according to the relevance rule mining algorithm, and adjusting the optimization process of correcting the virtual reality game performance according to the judging result.
Specifically, the method collects virtual reality game performance data through the step S1, comprehensively knows key indexes such as running state, resource consumption, frame rate and delay of a game, and provides data support for performance optimization, the method preprocesses the virtual reality game performance data through the step S2 to obtain preprocessing data, eliminates noise, errors or inconsistencies in the original data, improves data quality, thereby more accurately reflecting the real situation of game performance, and provides more reliable basis for subsequent analysis and optimization, the method analyzes the bottleneck factors of the virtual reality game performance according to the preprocessing data through the step S3 so as to accurately obtain key factors causing the game performance to be reduced, and the method inputs the bottleneck factors of the virtual reality game performance into an optimization decision model and outputs a virtual reality game performance optimization strategy through the step S4; the method comprises the steps of associating performance bottleneck factors with optimization strategies, recommending corresponding optimization strategies for different bottleneck factors, thereby improving the performance optimization efficiency of the virtual reality game, judging the use condition of the user behavior data according to the repetition times of the user behavior data in step S5, correcting the performance optimization process of the virtual reality game according to the judgment result, ensuring that the optimization process and the optimization result are closer to the user demand and the expectation, judging the association between the use condition of the user behavior data and the performance bottleneck factors of the virtual reality game according to the association rule mining algorithm in step S6, adjusting the performance optimization process of the corrected virtual reality game according to the judgment result, revealing the influence mechanism of the use condition of the user behavior data and the performance bottleneck factors of the virtual reality game, the method provides deeper insight for the formulation of the optimization strategy, and further improves the effectiveness and pertinence of the optimization strategy by adjusting the optimization process to adapt to the relevance, thereby further improving the optimization efficiency of the virtual reality game performance.
Specifically, in the step S1, when the virtual reality game performance data is collected, the virtual reality game performance data is collected by the internal monitoring system of the virtual reality game device.
Specifically, the internal monitoring system of the virtual reality game device refers to a set of system integrated in the virtual reality game device and game software for capturing, analyzing and displaying various performance data and user behavior data in the game running process in real time, the virtual reality game is a game created by utilizing a virtual reality technology and capable of enabling players to interact and experience in a virtual environment, the virtual reality game performance data refers to a series of quantization indexes for describing and evaluating the performance of the virtual reality game system, the specific content of the virtual reality game performance data is not limited, and related technicians in the field can freely set the system, and only the requirements of subsequent data processing and analysis are met, for example, the virtual reality game performance data including frame rate, delay, memory occupation, processor utilization rate, resolution and perception can be set.
Specifically, in the step S2, the virtual reality game performance data is preprocessed by a preprocessing method, so as to obtain preprocessed virtual reality game performance data.
Referring to fig. 2, a flow chart of the pretreatment method of the present embodiment is shown, where the pretreatment method includes:
step S20, data cleaning is carried out on the virtual reality game performance data to obtain effective data;
step S21, dividing the effective data into image data and data point data;
Step S22, carrying out standardization processing on the data point data through a standardization algorithm to obtain standardized data point data;
And S23, carrying out feature extraction on the image data and the standardized data point data, and combining the data after feature extraction to obtain preprocessing data.
Specifically, in the step S20, when the virtual reality game performance data is subjected to data cleaning, the missing value of the virtual reality game performance data is filled in by an interpolation method, the abnormal value of the virtual reality game performance data is detected by an IQR method, representative information of the abnormal value is collected, importance of the representative information of the abnormal value is determined, and the abnormal value is deleted and marked according to the determination result, so that effective data is obtained, wherein:
when the representative information of the outlier is important information, reserving and marking the outlier;
When the representative information of the outlier is unimportant information, the outlier is deleted.
Specifically, the interpolation method is a numerical calculation method for constructing a function having the same properties as the original function on the data points based on given finite data, the missing values are clusters, groups, deletions and truncations of data due to lack of information in virtual reality game performance data, the IQR method is a method for describing data distribution in statistics, the outliers are measured values having deviations from the average value exceeding a preset multiple such as twice standard deviation in a set of measured values, the representative information of the outliers is specific events in the virtual reality game such as network abnormality, character model loading failure and game rendering failure, and the important information is information having a significant influence on virtual reality game performance such as network abnormality, packet reading failure and virtual reality game card.
Specifically, in step S21, the effective data stored in the form of pixels is divided into image data according to the storage form of the effective data, and the effective data stored in the form of numerical values is divided into data point data.
In particular, the storage form refers to a manner of representing and storing effective data in a virtual reality system, for example, the data may be stored in binary, text, image, audio and video forms, the effective data stored in pixel form refers to the data stored in image and video forms, and the effective data stored in numerical form refers to the data stored in numerical data forms.
Specifically, in the step S22, the data point data is normalized by a Z-score normalization method in a normalization algorithm, and a calculation formula of the normalization algorithm is set to be z= (X- μ)/σ, where X is the data point data, μ is the mean value of the data point data, σ is the standard deviation of the data point data, and Z is the normalized data point data;
The calculation formula of the mean value mu is set asWhere N is the number of data points in the data point data, X i is the i-th data point in the data point data, e.g., the first data point in the data point data is X1 and the second data point is X2;
Setting the calculation formula of standard deviation sigma as
Specifically, the normalization algorithm refers to converting data into a distribution having a mean value of 0 and a standard deviation of 1, the Z-score normalization method refers to a data preprocessing technique for scaling data of different features to the same scale such that the data of each feature has a standard normal distribution having a mean value of 0 and a standard deviation of 1, the mean value of the data point data refers to the sum of all data points divided by the number of data points, the standard deviation of the data point data refers to an index measuring the degree of dispersion of the data distribution, the number of data points in the data point data refers to the total number of data points contained in the data point data,
Specifically, in the step S23, feature extraction is performed on the image data by a Canny edge detection method to obtain feature image data, where the Canny edge detection method includes:
Step G100, removing noise from the image data by using Gaussian filtering to obtain noise-removed image data;
step G200, calculating the gradient direction of the image data after noise removal by using a Sobe l operator to obtain the gradient direction of the image data;
Step G300, performing non-maximum suppression in the gradient direction of the image data by using a non-maximum suppression method, and reserving local maximum points to obtain the maximum point distribution of the image data;
step G400, judging each point in the maximum point distribution of the image data by using a double-threshold processing method, performing edge connection according to a judging result, setting the value of each maximum point as V, the maximum threshold as Vmax and the minimum threshold as Vmin, wherein:
when V > Vmax, judging the maximum point as a strong edge;
When Vmin is less than or equal to V is less than or equal to Vmax, judging the maximum point as a weak edge;
When V < Vmin, judging the maximum value point as a non-edge;
and G500, connecting the strong edge and the weak edge, and outputting the connected image as characteristic image data.
Specifically, the Canny edge detection method refers to an edge image processing technology used for extracting an image, the Gaussian filtering refers to an edge thinning technology used for removing pixels which are not true edges, the thinning edge refers to an image smoothing processing method used for reducing noise by means of a Gaussian function, the noise is removed refers to high-frequency noise in the image through a filter, edges and features in the image are more obvious, the Sobe operator refers to a convolution kernel used for calculating gradient of the image, the gradient strength refers to the amplitude of change of pixel values in the image, the gradient direction refers to the direction of change of pixel values in the image, the non-maximum suppressing method refers to an edge thinning technology used for removing pixels which are not true edges, the thinning edge refers to edge thinning and sharpness by means of restraining non-maximum points in the gradient direction, the local maximum points refer to pixel points in the gradient direction, the gradient value is larger than the pixel points of adjacent points, the dual threshold processing method refers to a method used for calculating gradient strength of the image, the gradient strength of the gradient is changed by using two thresholds in the Canny edge detection, the method refers to the fact that each point in the gradient is connected with the two threshold values in the image, the two threshold values are connected with the two threshold values, the two threshold values are connected with the edge thinning edge, and the edge thinning edge is formed, and the edge thinning edge is sharp by means.
Specifically, in the step S23, feature extraction is performed on the normalized data point data by the principal component analysis method to obtain feature data point data, the feature image data and the feature data point data are input into the depth feature fusion model to be combined, and the depth feature fusion model outputs the preprocessing data.
Specifically, the principal component analysis method is a statistical technique for converting a set of variables with possibly correlation into a set of variables with linear independence through forward-reverse conversion, the deep feature fusion model is a model which is obtained by training a convolutional neural network model according to a preset feature fusion data set and meets a preset merging success rate, the preset feature fusion data set is a data set of feature image data and feature data point data which are preset by training the convolutional neural network model, the training mode of the convolutional neural network model is not limited, a person skilled in the art can freely set according to actual requirements, if the requirement of merging the feature image data and the feature data point data is met, the situation can be set that the preset feature fusion data set is divided into a training set, 20% is divided into a test set, 10% is divided into a verification set, the training set is input into the convolutional neural network model for training, the test set is input into the trained convolutional neural network model, parameters in the convolutional neural network model are optimized, the verification set is input into the convolutional neural network model, the convolutional neural network is verified, and the training mode is carried out until the correct depth of the convolutional neural network model is output when the correct depth of the convolutional neural network model is 99%.
Specifically, in the step S3, when the bottleneck factors of the performance of the virtual reality game are analyzed according to the preprocessing data, the preprocessing data is input into the bottleneck factor analysis model, and the bottleneck factor analysis model is used for outputting the bottleneck factors of the performance of the virtual reality game.
Specifically, the virtual reality performance bottleneck factor refers to a factor causing performance degradation when a virtual reality game is executed and applied, for example, hardware performance, software performance, network performance, game content and scene complexity, the bottleneck factor analysis model refers to a model which is obtained by training a random forest model according to a preset game performance bottleneck factor data set and meets a preset analysis accuracy, the preset game performance bottleneck factor data set refers to preset virtual reality game performance bottleneck factors corresponding to preset preprocessing data, the training mode of the random forest model is not limited, a person skilled in the art can freely set according to actual requirements, only needs to meet the requirement of analyzing the virtual reality game performance bottleneck factor, for example, the situation can be set that 60% of the preset game performance bottleneck factor data set is divided into a training set, 20% of the training set is divided into a verification set, the training set is input into the random forest model for training, the test set is input into the trained random forest model, the parameter of the random forest model is corrected according to the test result, the verification set is input into the random forest model for verification, and the random forest model is output as a correct result when the random forest model is output as the bottleneck factor 80%.
Specifically, in the step S4, an optimization decision model is constructed through an expert optimization strategy data set, and a virtual reality game performance bottleneck factor is input into the optimization decision model to output a virtual reality game performance optimization strategy.
Specifically, the expert optimization strategy data set refers to a strategy set designed by an expert in the field of virtual reality games based on experience and knowledge and used for optimizing virtual reality game performance, wherein the strategy set comprises a virtual reality game performance optimization strategy corresponding to a preset virtual reality game performance bottleneck factor and a preset virtual reality game performance bottleneck factor, the optimization decision model refers to a decision tree model which is trained according to the expert optimization strategy data set and meets the preset decision accuracy, the training mode of the decision tree model is not limited, the relevant technical staff in the field can be freely set and only needs to meet the requirement of the optimization decision model, if the training mode of the settable decision tree model is that 90% of the expert optimization strategy data set is divided into a training set, 10% of the training set is divided into a verification set, the maximum depth of a tree in parameters of the decision tree model is set to be the number of virtual reality game performance optimization strategies in the expert optimization strategy data set, the minimum sample number is set to be 1, the training set is input into the decision tree model after the training set, the verification set is input into the decision tree model after training, parameters of the decision tree model are corrected, the decision tree model is output until the maximum depth of the decision tree model reaches the maximum depth of a node in the virtual reality game performance optimization strategy model, the maximum value is reached to the maximum point, the maximum point of the real game performance optimization strategy is reached in the virtual reality game performance optimization strategy is reached, and the maximum point is reached to the maximum point node performance optimization performance is reached, and the maximum point is reached to the real game performance optimization performance is reached, and the maximum point is reached, and the real node performance is optimized node, and the real game performance is optimized, optimizing memory management, adopting a low polygon model, reducing cutscene and balancing server load.
Specifically, in the step S5, user behavior data is obtained, the number of repetitions J of the user behavior data is compared with the number of repetitions J0 of the preset user behavior data, the use condition of the user behavior data is judged according to the comparison result, and the standard condition of the virtual reality game performance optimization process is judged and corrected according to the judgment result, wherein:
When J < J0, judging that the user behavior data is low-frequency user behavior data, wherein the virtual reality game performance optimization process reaches the standard, and correcting the virtual reality game performance optimization process is not performed:
when J is more than or equal to J0, judging that the user behavior data is high-frequency user behavior data, correcting the virtual reality game performance optimization process, pushing an expert optimization strategy data set update window to a virtual reality game equipment manager, updating the expert optimization strategy data set by the virtual reality game equipment manager, acquiring an updated expert optimization strategy data set, and optimizing the virtual reality game performance according to the updated expert optimization strategy data set.
Specifically, the user behavior data refers to data generated when a user uses a virtual reality game device, such as interaction data, motion data, virtual reality game device usage data, user preference data and social interaction data, the number of repetitions J of the user behavior data refers to the number of repetitions of the user in the virtual reality game device, the preset number of repetitions J0 of the user behavior data refers to a preset reference value for reflecting the usage situation of the user behavior data, the value of the preset number of repetitions J0 of the user behavior data is not limited in this embodiment, and a person skilled in the art can freely set according to actual requirements, for example, only needs to meet the requirement for judging the usage situation of the user behavior data, for example, preset number of repetitions J0 = 20 of the user behavior data can be set, the manager of the virtual reality game device refers to a professional responsible for managing, configuring and maintaining the virtual reality game device, the manner of acquiring the user behavior data is not limited in this embodiment, and the person skilled in the art can freely set according to actual requirements, for acquiring the user behavior data only needs to meet the requirement of acquiring the user behavior data, for example, the system can acquire user behavior data through internal monitoring of the virtual reality game device can be set.
Specifically, in the step S6, when the correlation between the usage of the user behavior data and the bottleneck factor of the virtual reality game performance is determined according to the association rule mining algorithm, the confidence coefficient P between the usage of the user behavior data and the bottleneck factor of the virtual reality game performance is calculated according to the association rule mining algorithm, the confidence coefficient P is compared with the preset confidence coefficient P0, the correlation between the usage of the user behavior data and the bottleneck factor of the virtual reality game performance is determined according to the comparison result, and the correction process is adjusted according to the determination result, wherein:
when P < P0, judging that the relevance between the use condition of the user behavior data and the bottleneck factor of the virtual reality game performance is low, and not adjusting the optimization process of correcting the virtual reality game performance;
When P is more than or equal to P0, judging that the relevance between the use condition of the user behavior data and the bottleneck factor of the virtual reality game performance is high, adjusting the correction process, setting an adjustment coefficient Q, wherein Q=1+ (P-P0), adjusting the preset user behavior data repetition number J0 according to the adjustment coefficient Q, and setting the preset user behavior data repetition number after adjustment to be Jq0, wherein Jq0=J0×Q.
Specifically, the association rule mining algorithm is a data mining algorithm for finding association relationships between data items from a large-scale data set, the association is a relationship and a dependency between the data items, the confidence level is an index for quantifying the reliability of the association rule, the preset confidence level P0 is a preset value for judging the association between the usage of user behavior data and a virtual reality game performance bottleneck factor, the value of the preset confidence level P0 is not limited, and a person skilled in the art can freely set according to actual requirements, so long as the requirement for judging the association between the usage of user behavior data and the virtual reality game performance bottleneck factor is met, for example, the preset confidence level p0=0.75 can be set.
Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.