Football player comprehensive ability evaluation method, system and equipment based on big data and medium

Abstract

The invention provides a football player comprehensive ability evaluation method, system and device based on big data and a medium, and belongs to the technical field of sports data analysis. The method comprises the steps that technical data, tactical data, physical ability data and physiological data of football players are collected; performing feature extraction and standardization processing on the data; acquiring contextual information of the match, presetting weights of technical features, tactical features, physical fitness features and physiological features in different match scenes, and weighting the standardized features to obtain a fused weighted feature vector; constructing a high-order ability quantification model, and calculating a high-order ability index of the player according to the weighted data; simulating the ability performance of the players under different opponent strengths based on the CGAN, calculating the deviation degree between the performance of the players and the reference performance, determining the stability coefficient, generating the comprehensive contribution index of the players in combination with the high-order ability index, and evaluating the comprehensive ability of the players. According to the method, the accuracy of player ability evaluation is improved, and the method can be used for optimizing team lineup configuration.

Description

Technical Field

[0001] This application belongs to the field of sports data analysis technology, specifically involving a method, system, equipment, and medium for evaluating the comprehensive abilities of football players based on big data. Background Technology

[0002] In the field of sports data analysis technology, comprehensive player ability assessment is the core support for professional teams' talent selection, tactical optimization, and transfer decisions.

[0003] Existing player evaluation techniques mainly fall into three categories: First, they rely on basic statistical data such as goals, assists, and running distance, calculating player contribution values ​​through simple weighting methods like the Opta index or regression models like Expected Goals. These methods only reflect explicit player performance and cannot quantify abstract abilities such as creativity and defensive anticipation, leading to blind spots in the evaluation of a player's tactical value. Second, the analytical dimensions are fragmented. Physical fitness monitoring data such as GPS and heart rate operate independently from technical and tactical data obtained through video tracking, lacking correlation analysis. For example, the physical fitness system only focuses on running parameters, and technical and tactical tools only track player positions, resulting in one-sided evaluation results that fail to reflect the player's comprehensive role in actual match scenarios. Third, the evaluation system lacks objectivity, with an excessive reliance on coach experience evaluation, leading to significant subjective bias. Different evaluators have vastly different judgments of the same player, making it difficult to establish standardized evaluation criteria.

[0004] Existing technologies cannot transform technical, tactical, physical, and physiological data into a unified evaluation dimension; they cannot adjust the importance of each data dimension according to the stage of the game, the strength of the opponent, and other scenarios, resulting in a static evaluation method; and they lack quantitative algorithms for advanced football intelligence, making it difficult to transform abstract abilities into objective indices. Ultimately, the evaluation results cannot truly reflect the player's comprehensive contribution and potential value, which restricts the accuracy of team training, selection, and tactical deployment. Summary of the Invention

[0005] The purpose of this application is to provide a method, system, device, and medium for evaluating the comprehensive ability of football players based on big data, in order to solve the problems in the existing technology that it is impossible to transform multi-source data into a unified evaluation dimension, the evaluation method is static, and the evaluation results cannot truly reflect the comprehensive contribution and potential value of players.

[0006] In a first aspect, embodiments of this application provide a method for evaluating the comprehensive abilities of football players based on big data, comprising the following steps: S1. Collect multi-source data, including player technical data, tactical data, physical fitness data, and physiological data; S2. Perform feature extraction and standardization processing on the collected multi-source data; S3. Obtain the match context information, which includes the match stage, opponent strength, tactical role, and intensity of offensive and defensive transitions. Preset the weights of technical features, tactical features, physical features, and physiological features under different match scenarios, and weight the standardized features to obtain a fused weighted feature vector. S4. Construct a high-level ability quantification model, which calculates the player's high-level ability index based on weighted data. The high-level ability index includes a creativity index and a defensive reading ability index. S5. Construct an adversarial scenario simulator based on Conditional Generative Adversarial Network (CGAN), using fused weighted feature vectors and match context information as input to simulate player performance under different opponent strengths, calculate the deviation of player performance from benchmark performance, determine the stability coefficient, and combine the higher-order ability index and stability coefficient to generate a player comprehensive contribution index to evaluate the player's comprehensive ability.

[0007] Furthermore, the specific steps of step S1 are as follows: S11. Use computer vision equipment to perform frame-by-frame tracking and analysis of the game video to extract technical data, including: pass success rate, number of breakthroughs, and accuracy of shooting angles; S12. Construct a tactical analysis framework using graph neural networks (GNNs) and collect tactical data, including positional heatmaps, defensive pressure frequency, and passing network centrality. S13. Collect physical fitness data through GPS / IMU sensors, including high-speed running data and acceleration data when changing direction; S14. Collect physiological data, including heart rate variability (HRV) and muscle oxygen saturation, using heart rate monitoring devices and near-infrared spectroscopy (NIRS) sensors.

[0008] Furthermore, the specific steps of step S12 are as follows: S121. By using the positioning chip worn by the players, the player's on-field coordinates are collected every 10 seconds to generate raw data for the position heat map; S122. Using a motion trajectory matching algorithm, identify the player's approaching action towards the opponent's player with the ball, count the number of defensive presses per minute, and obtain the defensive press frequency; S123. Construct a passing network by treating players as graph nodes and passing relationships as edges. Use the node importance calculation algorithm of GNN to obtain the passing network centrality of each player.

[0009] Furthermore, the specific steps of step S2 are as follows: S21. Select the passing success rate, number of breakthroughs, and shooting angle accuracy from the technical data to form a technical feature vector. Standardize each technical feature in the feature vector and control the range of the standardized technical feature values ​​within the range of [-3,3]. S22. Extract the original data of position heatmap, defensive pressure frequency, and passing network centrality from the tactical data. The position heatmap is converted into quantitative features by the number of hot spots to form a tactical feature vector. Based on the attention mechanism of the graph neural network, the original passing network centrality is optimized. The optimized tactical feature vector is scaled to the [0,1] interval by Min-Max. S23. Extract high-speed running data and change-of-direction acceleration data from physical fitness data to form physical fitness feature vectors. Use box plot method to identify abnormal values ​​of physical fitness features. Replace the abnormal values ​​with the feature mean of adjacent 5-minute time periods. Scale the processed physical fitness features to the [0,1] interval using Min-Max. S24. Extract heart rate variability (HRV) and muscle oxygen saturation from physiological data to form a physiological feature vector. Use the average physiological data of the player 30 minutes before the game as the baseline value and the fluctuation range of the player's physiological data in the last 5 games as the dynamic interval for dynamic normalization calculation.

[0010] Furthermore, the specific steps of step S3 are as follows: S31. Obtain contextual information through the competition data platform, including competition stage, opponent strength, tactical role, and intensity of offensive and defensive transitions. Based on the collected contextual information, divide the competition into 8 scenarios: high-pressure defense, fast counterattack, positional attack, positional defense, medium-intensity confrontation, low-intensity confrontation, critical period attack, and critical period defense. S32. Based on expert ratings and historical match data, conduct correlation analysis to calculate the Pearson correlation coefficient between each feature and the player's actual contribution value. Use weighted summation to determine the weights of technical features, tactical features, physical features, and physiological features in each scenario. The sum of the weights is 1, forming a weight table for 8 match scenarios. S33. Based on the current competition scenario, retrieve the corresponding weights from the weight table, assign weights to the standardized technical feature vector, tactical feature vector, physical feature vector, and physiological feature vector respectively, and perform weighted calculations to form a fused weighted feature vector.

[0011] Furthermore, the specific steps of step S4 are as follows: S41. Extract the parameters required for the creativity index calculation from the fused weighted feature vector and the collected multi-source data, including the number of successful high-risk passes, the total number of high-risk passes, the frequency of use of each type of pass, the density of the opponent's defensive players in the passing area, and their movement speed. S42. Calculate the probability P of a successful high-risk pass using the following formula:

[0012] Based on the information entropy theory, the pass selection entropy value E is calculated using the following formula:

[0013] in, p i Let i be the frequency percentage of the i-th type of pass, i=1,2,3,4, corresponding to short pass, long pass, through ball, and diagonal pass respectively, with an entropy range of [0.2, 1.4], where 0.2 is the lowest entropy value for a single pass type and 1.4 is the highest entropy value for a variety of passes. The formula for calculating the opponent's pressure intensity I is:

[0014] The passing area is a circular area with a radius of 10 meters centered on the starting point of the pass, and the calculation results are scaled to the [0,1] interval using Min-Max. S43. Calculate the creativity index C using the following formula:

[0015] Where P is the probability of a successful high-risk pass, E is the pass selection entropy, and I is the opponent's pressure intensity; S44. Extract the predicted number of interceptions, total number of defensive actions, real-time player position on the field, and distance from key defensive areas from the fused weighted feature vector and collected multi-source data; S45. Calculate the predicted interception efficiency R, using the following formula:

[0016] To calculate the position correction coefficient K, first determine the base coefficient based on the on-field position: defender 1.2, midfielder 1.0, forward 0.8, and then adjust it according to the distance correction rules; let the distance between the player and the key defensive area be d meters. If d≤d0, the coefficient remains unchanged; if d0<d≤2d0, the base coefficient is multiplied by 0.8; if d>2d0, the coefficient is multiplied by 0.5. S46. Calculate the defensive reading ability index using the following formula:

[0017] Where R is the predicted interception efficiency and K is the position correction coefficient.

[0018] Furthermore, the specific steps of step S5 are as follows: S51. Encode the game context information into a vector, normalize it, and fuse the weighted feature vector as the CGAN input vector. At the same time, take the average of the player's high-order ability index in the same scenario in the last 3 games as the benchmark performance vector. S52. Build CGAN, the generator outputs player performance data under different opponent strengths, the discriminator uses real player performance data under different opponent strengths or player performance data output by the generator to conduct adversarial training to optimize the generator; generate 5 sets of effective simulated data for each opponent strength level, remove outliers, and take the mean as the simulated performance result under that strength. S53. Calculate the deviation of creativity and the deviation of defensive reading ability. After normalization, sum the weighted deviations with a weight of 0.4:0.6. Obtain the stability coefficient through the formula: stability coefficient = 1 - weighted deviation. The stability coefficient ranges from [0,1], and the closer it is to 1, the stronger the stability. S54. The Creativity Index, Defensive Reading Ability Index, and Stability Coefficient are weighted and summed at a ratio of 0.3:0.3:0.4 to obtain the Comprehensive Contribution Index (CCI), which is used to evaluate the player's overall ability.

[0019] Secondly, embodiments of this application also provide a comprehensive ability assessment system for football players based on big data, the system comprising: The data acquisition module is used to collect multi-source data, including players' technical data, tactical data, physical data, and physiological data. The feature processing module is used to extract and standardize features from the collected multi-source data. The feature weighting module is used to obtain match context information, which includes match stage, opponent strength, tactical role, and intensity of offensive and defensive transitions. It presets the weights of technical features, tactical features, physical features, and physiological features under different match scenarios, and weights the standardized features to obtain a fused weighted feature vector. The advanced ability quantification module is used to construct an advanced ability quantification model. The model calculates the advanced ability index of the player based on the weighted data. The advanced ability index includes the creativity index and the defensive reading ability index. The adversarial scenario simulation and comprehensive evaluation module is used to construct an adversarial scenario simulator based on the conditional generative adversarial network CGAN. It uses fused weighted feature vectors and game context information as input to simulate the player's ability performance under different opponent strengths, calculates the deviation of the player's performance from the benchmark performance, determines the stability coefficient, and generates a player comprehensive contribution index by combining the higher-order ability index and the stability coefficient to evaluate the player's comprehensive ability.

[0020] Thirdly, embodiments of this application also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the big data-based comprehensive ability assessment method for football players as described in the first aspect.

[0021] Fourthly, embodiments of this application also provide a storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the big data-based comprehensive ability evaluation method for football players as described in the first aspect.

[0022] As can be seen from the above technical solutions, the present invention has the following advantages: This application discloses a method, system, device, and medium for comprehensive evaluation of football players' abilities based on big data. It achieves full data coverage by collecting multi-source data on players' techniques, tactics, physical fitness, and physiology. Through dimensional standardization processing, it effectively breaks down data silos caused by fragmented data dimensions in traditional evaluations, constructing a unified and standardized data processing link. This provides a high-quality, highly consistent data foundation for subsequent evaluations, ensuring the reliability of the evaluation results. It divides typical scenarios based on match context information and pre-sets scenario-based weights by combining expert scores with correlation analysis of historical data, solving the problem of the disconnect between traditional static evaluations and actual match scenarios, making feature weighting more closely aligned with players' actual contribution scenarios. By quantifying players' creativity and defensive reading ability, it transforms abstract abilities into objective indices, improving the scientific rigor and standardization of abstract ability evaluation and overcoming the limitations of traditional reliance on subjective experience. Based on a conditional generative adversarial network, it constructs an adversarial scenario simulator to simulate player performance under different opponent intensities and calculates stability coefficients. Combined with higher-order ability indices, it generates a comprehensive contribution index, which can identify undervalued players and predict their performance under pressure, providing precise decision support for team talent selection, tactical optimization, and transfer pricing, fully exploring players' comprehensive contributions and potential value. This application improves the accuracy of player ability assessments, which can be used to optimize team lineup configurations. Attached Figure Description

[0023] To more clearly illustrate the technical solution of this application, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating the comprehensive ability assessment method for football players based on big data, as described in this invention.

[0025] Figure 2 This is a schematic diagram of the big data-based comprehensive ability assessment system for football players according to the present invention. Detailed Implementation

[0026] The various embodiments of this disclosure will be described more fully in the following detailed description of the specific steps of the big data-based comprehensive ability assessment method for football players. This disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of this disclosure to the specific embodiments disclosed herein, but rather this disclosure should be understood to cover all adjustments, equivalents, and / or alternatives falling within the spirit and scope of the various embodiments of this disclosure.

[0027] It should be understood that, when used in this specification, the term "comprising" indicates the presence of the described feature, integral, step, operation, element, and / or component, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or collections thereof. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0028] The terms "one embodiment" or "some embodiments" used in this application mean that one or more embodiments of this application include the specific features, structures, or characteristics described in that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this application do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0029] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0030] Please see Figure 1 The diagram shown is a flowchart of a big data-based comprehensive ability assessment method for football players, including the following steps: S1. Collect multi-source data, including player technical data, tactical data, physical fitness data, and physiological data; It should be noted that by collecting multi-source data on players' skills, tactics, physical fitness, and physiology, the limitations of traditional single-dimensional data are broken, providing comprehensive data support for subsequent comprehensive evaluation and avoiding evaluation bias caused by one-sided data. S2. Perform feature extraction and standardization processing on the collected multi-source data; It should be noted that by extracting features from multi-source data in different dimensions and standardizing them, we can unify the data format, eliminate differences in units, lay the foundation for subsequent dynamic weighting and capability quantification, and improve the effectiveness and consistency of the evaluation data. S3. Obtain the match context information, which includes the match stage, opponent strength, tactical role, and intensity of offensive and defensive transitions. Preset the weights of technical features, tactical features, physical features, and physiological features under different match scenarios, and weight the standardized features to obtain a fused weighted feature vector. It should be noted that by dividing the scenario into scenarios based on the competition context and pre-setting feature weights, the standardized data is weighted to obtain a fusion vector, which adapts to dynamic competition scenarios, avoids static evaluation bias, and improves the fit between the evaluation and the actual scenario. S4. Construct a high-level ability quantification model, which calculates the player's high-level ability index based on weighted data. The high-level ability index includes a creativity index and a defensive reading ability index. It should be noted that constructing a high-level ability quantification model to calculate creativity and defensive reading ability indices transforms abstract abilities into objective indicators, breaking through the limitations of traditional subjective assessments and improving the accuracy of player ability assessments. S5. Construct an adversarial scenario simulator based on Conditional Generative Adversarial Network (CGAN), using fused weighted feature vectors and match context information as input to simulate player performance under different opponent strengths, calculate the deviation of player performance from benchmark performance, determine the stability coefficient, and generate a player comprehensive contribution index by combining the higher-order ability index and the stability coefficient to evaluate the player's comprehensive ability. It should be noted that a CGAN-based adversarial scenario simulator, using a fusion of weighted feature vectors and match context as input, can simulate player performance under varying opponent strengths. By calculating deviation to determine the stability coefficient, and combining it with higher-order ability indices to generate a comprehensive contribution index, this approach overcomes the limitations of traditional methods that only assess current performance, while also accurately uncovering a player's resilience and potential value.

[0031] Furthermore, as a refinement and extension of the specific implementation methods of the above embodiments, and to fully illustrate the specific implementation process of this embodiment, another method for evaluating the comprehensive ability of football players based on big data is provided. The method includes: S1. Collect multi-source data, including player technical data, tactical data, physical fitness data, and physiological data; the specific steps of step S1 are as follows: S11. Use computer vision equipment to perform frame-by-frame tracking and analysis of the game video to extract technical data, including: pass success rate, number of breakthroughs, and accuracy of shooting angles; S12. Construct a tactical analysis framework using a graph neural network (GNN) and collect tactical data, including positioning heatmaps, defensive pressing frequency, and passing network centrality; the specific steps of step S12 are as follows: S121. By using the positioning chip worn by the players, the player's on-field coordinates are collected every 10 seconds to generate raw data for the position heat map; S122. Using a motion trajectory matching algorithm, identify the player's approaching action towards the opponent's player with the ball, count the number of defensive presses per minute, and obtain the defensive press frequency; S123. Construct a passing network by treating players as graph nodes and passing relationships as edges. Use the node importance calculation algorithm of GNN to obtain the passing network centrality of each player. S13. Collect physical fitness data through GPS / IMU sensors, including high-speed running data and acceleration data when changing direction; S14. Collect physiological data, including heart rate variability (HRV) and muscle oxygen saturation, using heart rate monitoring equipment and near-infrared spectroscopy NIRS sensors; For example, in a professional football league match, 8K high-definition cameras are used to analyze the match video frame by frame to extract player A's passing success rate, number of breakthroughs, and shooting angle accuracy; the positioning chip worn by the player is used to collect coordinates every 10 seconds to generate a position heatmap, and a motion trajectory matching algorithm is used to count the frequency of defensive pressing, and a GNN is used to construct a passing network to obtain centrality; the proportion of high-speed running and acceleration due to change of direction are collected with the help of GPS / IMU sensors; and heart rate monitoring equipment and NIRS sensors are used to collect HRV and muscle oxygen saturation. S2. Perform feature extraction and standardization processing on the collected multi-source data; the specific steps of step S2 are as follows: S21. Select the passing success rate, number of breakthroughs, and shooting angle accuracy from the technical data to form a technical feature vector. Standardize each technical feature in the feature vector and control the range of the standardized technical feature values ​​within the range of [-3,3]. S22. Extract the original data of position heatmap, defensive pressure frequency, and passing network centrality from the tactical data. The position heatmap is converted into quantitative features by the number of hot spots to form a tactical feature vector. Based on the attention mechanism of the graph neural network, the original passing network centrality is optimized. The optimized tactical feature vector is scaled to the [0,1] interval by Min-Max. S23. Extract high-speed running data and change-of-direction acceleration data from physical fitness data to form physical fitness feature vectors. Use box plot method to identify abnormal values ​​of physical fitness features. Replace the abnormal values ​​with the feature mean of adjacent 5-minute time periods. Scale the processed physical fitness features to the [0,1] interval using Min-Max. S24. Extract heart rate variability (HRV) and muscle oxygen saturation from physiological data to form a physiological feature vector. Use the average physiological data of the player 30 minutes before the game as the baseline value and the fluctuation range of the player's physiological data in the last 5 games as the dynamic interval for dynamic normalization calculation. For example, the multi-source data of player A is processed, and the passing success rate and other data are filtered to form vectors. After Z-score standardization, the values ​​are 1.2, 0.8, and -0.5, all within [-3,3]. On the tactical side, the positioning heatmap is converted into quantified values ​​of 12 hotspot areas. After optimizing the centrality of the passing network, the vector is scaled to [0.6, 0.4, 0.7] using Min-Max. On the physical side, outliers are removed using a box plot and the vector is scaled to [0.5, 0.3]. On the physiological side, the vector is dynamically normalized to [0.6, 0.7] based on the pre-match mean. S3. Obtain match context information, which includes match stage, opponent strength, tactical role, and intensity of offensive and defensive transitions. Preset weights for technical, tactical, physical, and physiological characteristics under different match scenarios, and weight the standardized features to obtain a fused weighted feature vector. The specific steps of step S3 are as follows: S31. Obtain contextual information through the competition data platform, including competition stage, opponent strength, tactical role, and intensity of offensive and defensive transitions. Based on the collected contextual information, divide the competition into 8 scenarios: high-pressure defense, fast counterattack, positional attack, positional defense, medium-intensity confrontation, low-intensity confrontation, critical period attack, and critical period defense. S32. Based on expert ratings and historical match data, conduct correlation analysis to calculate the Pearson correlation coefficient between each feature and the player's actual contribution value. Use weighted summation to determine the weights of technical features, tactical features, physical features, and physiological features in each scenario. The sum of the weights is 1, forming a weight table for 8 match scenarios. S33. Based on the current competition scenario, retrieve the corresponding weights from the weight table, assign weights to the standardized technical feature vector, tactical feature vector, physical feature vector, and physiological feature vector respectively, and perform weighted calculations to form a fused weighted feature vector; For example, the weight table for the eight scenarios is shown in Table 1: Table 1 Weighting Table for 8 Scenarios

[0032] S4. Construct a high-order ability quantification model. This model calculates the player's high-order ability index based on weighted data. The high-order ability index includes a creativity index and a defensive reading ability index. The specific steps of step S4 are as follows: S41. Extract the parameters required for the creativity index calculation from the fused weighted feature vector and the collected multi-source data, including the number of successful high-risk passes, the total number of high-risk passes, the frequency of use of each type of pass, the density of the opponent's defensive players in the passing area, and their movement speed. S42. Calculate the probability P of a successful high-risk pass using the following formula:

[0033] Based on the information entropy theory, the pass selection entropy value E is calculated using the following formula:

[0034] in, p i Let i be the frequency percentage of the i-th type of pass, i=1,2,3,4, corresponding to short pass, long pass, through ball, and diagonal pass respectively, with an entropy range of [0.2, 1.4], where 0.2 is the lowest entropy value for a single pass type and 1.4 is the highest entropy value for a variety of passes. The formula for calculating the opponent's pressure intensity I is:

[0035] The passing area is a circular area with a radius of 10 meters centered on the starting point of the pass, and the calculation results are scaled to the [0,1] interval using Min-Max. S43. Calculate the creativity index C using the following formula:

[0036] Where P is the probability of a successful high-risk pass, E is the pass selection entropy, and I is the opponent's pressure intensity; S44. Extract the predicted number of interceptions, total number of defensive actions, real-time player position on the field, and distance from key defensive areas from the fused weighted feature vector and collected multi-source data; S45. Calculate the predicted interception efficiency R, using the following formula:

[0037] To calculate the position correction coefficient K, first determine the base coefficient based on the on-field position: defender 1.2, midfielder 1.0, forward 0.8, and then adjust it according to the distance correction rules; let the distance between the player and the key defensive area be d meters. If d≤d0, the coefficient remains unchanged; if d0<d≤2d0, the base coefficient is multiplied by 0.8; if d>2d0, the coefficient is multiplied by 0.5. S46. Calculate the defensive reading ability index using the following formula:

[0038] Where R is the predicted interception efficiency and K is the position correction coefficient; For example, taking a professional league midfielder B as an example, we calculate his advanced ability index. Parameters are extracted from the fused weighted feature vector and multi-source data: 8 successful high-risk passes, 12 total passes, the frequency ratios of short passes, long passes, through balls, and diagonal passes are 0.4, 0.2, 0.3, and 0.1 respectively, the density of defensive players in the passing area is 0.3 players / m², the average movement speed of defensive players is 5 m / s, and the player passing speed is 8 m / s.

[0039] The calculated success probability of a high-risk pass is P = 8 / 12 ≈ 0.67; the pass selection entropy is E = -(0.4 × ln0.4 + 0.2 × ln0.2 + 0.3 × ln0.3 + 0.1 × ln0.1) ≈ 1.18; the opponent's pressure intensity is I = (0.3 × 0.6 + (5 / 8) × 0.4) ≈ 0.43, which becomes 0.45 after Min-Max scaling; and the creativity index is C = 0.67 × 1.18 × 0.45 ≈ 0.36. Additionally, 6 anticipatory interceptions and 20 total defensive actions were extracted. The player was a midfielder, 7 meters away from the key defensive area (d0=8 meters, d≤d0, coefficient remains unchanged); the anticipatory interception efficiency R=6 / 20=0.3, and the defensive reading ability index=0.3×1.0=0.3; S5. Construct an adversarial scenario simulator based on the Conditional Generative Adversarial Network (CGAN). Using fused weighted feature vectors and match context information as input, simulate player performance under different opponent strengths, calculate the deviation of player performance from benchmark performance, determine the stability coefficient, and combine the higher-order ability index and stability coefficient to generate a comprehensive player contribution index to evaluate the player's overall ability. The specific steps of step S5 are as follows: S51. Encode the game context information into a vector, normalize it, and fuse the weighted feature vector as the CGAN input vector. At the same time, take the average of the player's high-order ability index in the same scenario in the last 3 games as the benchmark performance vector. S52. Build CGAN, the generator outputs player performance data under different opponent strengths, the discriminator uses real player performance data under different opponent strengths or player performance data output by the generator to conduct adversarial training to optimize the generator; generate 5 sets of effective simulated data for each opponent strength level, remove outliers, and take the mean as the simulated performance result under that strength. S53. Calculate the deviation of creativity and the deviation of defensive reading ability. After normalization, sum the weighted deviations with a weight of 0.4:0.6. Obtain the stability coefficient through the formula: stability coefficient = 1 - weighted deviation. The stability coefficient ranges from [0,1], and the closer it is to 1, the stronger the stability. S54. The Creativity Index, Defensive Reading Index, and Stability Coefficient are weighted and summed at a ratio of 0.3:0.3:0.4 to obtain the Comprehensive Contribution Index (CCI), which is used to evaluate the player's overall ability. For example, taking a professional league forward player C as an example, the game context information (second half, opponent's high intensity, core offensive role, high intensity of offensive and defensive transition) is encoded into a vector and normalized. The fused weighted feature vector of player C is concatenated as the input of CGAN, and the average of the high-order ability index of the player in the same scenario in the last 3 games (creativity index 0.42, defensive reading ability index 0.28) is taken as the baseline vector. CGAN was built, and the generator output simulated performance data under three levels of opponent strength: low, medium and high. After generating 5 sets of data for each level, 1 outlier was removed and the average value was taken: low strength (creativity 0.41, defense reading 0.27), medium strength (0.40, 0.26), high strength (0.39, 0.25). The single-dimensional deviation was calculated, normalized, and weighted at 0.4:0.6 to obtain a weighted deviation of 0.08. The stability coefficient was 1-0.08=0.92. Calculated with a weighting of 0.3:0.3:0.4, the comprehensive contribution index (CCI) is 0.42×0.3+0.28×0.3+0.92×0.4=0.62, which is assessed as an above-average comprehensive capability, and the high intensity of the operation demonstrates excellent stability.

[0040] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0041] Please see Figure 2 As shown below, the following is an embodiment of the comprehensive ability assessment system for football players based on big data provided in this application. This system and the comprehensive ability assessment method for football players based on big data in the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the comprehensive ability assessment system for football players based on big data, please refer to the embodiments of the comprehensive ability assessment method for football players based on big data.

[0042] The system includes: The data acquisition module is used to collect multi-source data, including players' technical data, tactical data, physical data, and physiological data. The feature processing module is used to extract and standardize features from the collected multi-source data. The feature weighting module is used to obtain match context information, which includes match stage, opponent strength, tactical role, and intensity of offensive and defensive transitions. It presets the weights of technical features, tactical features, physical features, and physiological features under different match scenarios, and weights the standardized features to obtain a fused weighted feature vector. The advanced ability quantification module is used to construct an advanced ability quantification model. The model calculates the advanced ability index of the player based on the weighted data. The advanced ability index includes the creativity index and the defensive reading ability index. The adversarial scenario simulation and comprehensive evaluation module is used to construct an adversarial scenario simulator based on the conditional generative adversarial network CGAN. It uses fused weighted feature vectors and game context information as input to simulate the player's ability performance under different opponent strengths, calculates the deviation of the player's performance from the benchmark performance, determines the stability coefficient, and generates a player comprehensive contribution index by combining the higher-order ability index and the stability coefficient to evaluate the player's comprehensive ability.

[0043] This embodiment achieves a comprehensive quantitative assessment of a football player's abilities through the interactive collaboration of a data acquisition module, a feature processing module, a feature weighting module, a high-order ability quantification module, and a confrontation scenario simulation and comprehensive evaluation module.

[0044] The big data-based comprehensive ability assessment method for football players provided in this application can be applied to electronic devices. Those skilled in the art will understand that the electronic device structure involved in the embodiments of this invention does not constitute a limitation on the electronic device. An electronic device may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. In the embodiments of this invention, the electronic device includes, but is not limited to, laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of this application described and / or claimed herein.

[0045] Electronic devices may include processors, external memory interfaces, internal memory, universal serial bus (USB) interfaces, charging management modules, power management modules, batteries, wireless communication modules, audio modules, speakers, microphones, sensor modules, buttons, cameras, displays, and SIM card interfaces, etc.

[0046] It is understood that the structures illustrated in the embodiments of this application do not constitute a specific limitation on the electronic device. In other embodiments of this application, the electronic device may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0047] A processor may include one or more processing units, such as a central processing unit (CPU), an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural network processing unit (NPU). Different processing units may be independent devices or integrated into one or more processors.

[0048] The processor can serve as the nerve center and command center of an electronic device. The controller can generate operation control signals based on the instruction opcode and timing signals to control the fetching and execution of instructions.

[0049] The processor may also include memory for storing instructions and data. In some embodiments, the memory in the processor is a cache memory. This memory can store instructions or data that the processor has just used or that are used repeatedly. If the processor needs to use the instruction or data again, it can retrieve it directly from this memory. This avoids repeated accesses, reduces processor latency, and thus improves system efficiency.

[0050] The storage medium provided in this application stores a program product capable of implementing a comprehensive ability assessment method for football players based on big data.

[0051] In some possible implementations, the big data-based comprehensive ability assessment method for football players disclosed herein can be implemented as a program product comprising program code that, when the program product is run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure.

[0052] The storage medium disclosed herein may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0053] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. 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 the invention. Therefore, the invention 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 comprehensive abilities of football players based on big data, characterized in that, The method includes: S1. Collect multi-source data, including player technical data, tactical data, physical fitness data, and physiological data; S2. Perform feature extraction and standardization processing on the collected multi-source data; S3. Obtain the match context information, which includes the match stage, opponent strength, tactical role, and intensity of offensive and defensive transitions. Preset the weights of technical features, tactical features, physical features, and physiological features under different match scenarios, and weight the standardized features to obtain a fused weighted feature vector. S4. Construct a high-level ability quantification model, which calculates the player's high-level ability index based on weighted data. The high-level ability index includes a creativity index and a defensive reading ability index. S5. Construct an adversarial scenario simulator based on Conditional Generative Adversarial Network (CGAN), using fused weighted feature vectors and match context information as input to simulate player performance under different opponent strengths, calculate the deviation of player performance from benchmark performance, determine the stability coefficient, and combine the higher-order ability index and stability coefficient to generate a player comprehensive contribution index to evaluate the player's comprehensive ability.

2. The method for evaluating the comprehensive ability of football players based on big data according to claim 1, characterized in that, The specific steps of step S1 are as follows: S11. Use computer vision equipment to perform frame-by-frame tracking and analysis of the game video to extract technical data, including: pass success rate, number of breakthroughs, and accuracy of shooting angles; S12. Construct a tactical analysis framework using graph neural networks (GNNs) and collect tactical data, including positional heatmaps, defensive pressure frequency, and passing network centrality. S13. Collect physical fitness data through GPS / IMU sensors, including high-speed running data and acceleration data when changing direction; S14. Collect physiological data, including heart rate variability (HRV) and muscle oxygen saturation, using heart rate monitoring devices and near-infrared spectroscopy (NIRS) sensors.

3. The method for evaluating the comprehensive ability of football players based on big data according to claim 1, characterized in that, The specific steps of step S12 are as follows: S121. By using the positioning chip worn by the players, the player's on-field coordinates are collected every 10 seconds to generate raw data for the position heat map; S122. Using a motion trajectory matching algorithm, identify the player's approaching action towards the opponent's player with the ball, count the number of defensive presses per minute, and obtain the defensive press frequency; S123. Construct a passing network by treating players as graph nodes and passing relationships as edges. Use the node importance calculation algorithm of GNN to obtain the passing network centrality of each player.

4. The method for evaluating the comprehensive ability of football players based on big data according to claim 1, characterized in that, The specific steps of step S2 are as follows: S21. Select the passing success rate, number of breakthroughs, and shooting angle accuracy from the technical data to form a technical feature vector. Standardize each technical feature in the feature vector and control the range of the standardized technical feature values ​​within the range of [-3,3]. S22. Extract the original data of position heatmap, defensive pressure frequency, and passing network centrality from the tactical data. The position heatmap is converted into quantitative features by the number of hot spots to form a tactical feature vector. Based on the attention mechanism of the graph neural network, the original passing network centrality is optimized. The optimized tactical feature vector is scaled to the [0,1] interval by Min-Max. S23. Extract high-speed running data and change-of-direction acceleration data from physical fitness data to form physical fitness feature vectors. Use box plot method to identify abnormal values ​​of physical fitness features. Replace the abnormal values ​​with the feature mean of adjacent 5-minute time periods. Scale the processed physical fitness features to the [0,1] interval using Min-Max. S24. Extract heart rate variability (HRV) and muscle oxygen saturation from physiological data to form a physiological feature vector. Use the average physiological data of the player 30 minutes before the game as the baseline value and the fluctuation range of the player's physiological data in the last 5 games as the dynamic interval for dynamic normalization calculation.

5. The method for evaluating the comprehensive ability of football players based on big data according to claim 1, characterized in that, The specific steps of step S3 are as follows: S31. Obtain contextual information through the competition data platform, including competition stage, opponent strength, tactical role, and intensity of offensive and defensive transitions. Based on the collected contextual information, divide the competition into 8 scenarios: high-pressure defense, fast counterattack, positional attack, positional defense, medium-intensity confrontation, low-intensity confrontation, critical period attack, and critical period defense. S32. Based on expert ratings and historical match data, conduct correlation analysis to calculate the Pearson correlation coefficient between each feature and the player's actual contribution value. Use weighted summation to determine the weights of technical features, tactical features, physical features, and physiological features in each scenario. The sum of the weights is 1, forming a weight table for 8 match scenarios. S33. Based on the current competition scenario, retrieve the corresponding weights from the weight table, assign weights to the standardized technical feature vector, tactical feature vector, physical feature vector, and physiological feature vector respectively, and perform weighted calculations to form a fused weighted feature vector.

6. The method for evaluating the comprehensive ability of football players based on big data according to claim 1, characterized in that, The specific steps of step S4 are as follows: S41. Extract the parameters required for the creativity index calculation from the fused weighted feature vector and the collected multi-source data, including the number of successful high-risk passes, the total number of high-risk passes, the frequency of use of each type of pass, the density of the opponent's defensive players in the passing area, and their movement speed. S42. Calculate the probability P of a successful high-risk pass using the following formula: Based on the information entropy theory, the pass selection entropy value E is calculated using the following formula: in, p i Let i be the frequency percentage of the i-th type of pass, i=1,2,3,4, corresponding to short pass, long pass, through ball, and diagonal pass respectively, with an entropy range of [0.2, 1.4], where 0.2 is the lowest entropy value for a single pass type and 1.4 is the highest entropy value for a variety of passes. The formula for calculating the opponent's pressure intensity I is: The passing area is a circular area with a radius of 10 meters centered on the starting point of the pass, and the calculation results are scaled to the [0,1] interval using Min-Max. S43. Calculate the creativity index C using the following formula: Where P is the probability of a successful high-risk pass, E is the pass selection entropy, and I is the opponent's pressure intensity; S44. Extract the predicted number of interceptions, total number of defensive actions, real-time player position on the field, and distance from key defensive areas from the fused weighted feature vector and collected multi-source data; S45. Calculate the predicted interception efficiency R, using the following formula: To calculate the position correction coefficient K, first determine the base coefficient based on the on-field position: defender 1.2, midfielder 1.0, forward 0.8, and then adjust it according to the distance correction rules; let the distance between the player and the key defensive area be d meters. If d≤d0, the coefficient remains unchanged; if d0<d≤2d0, the base coefficient is multiplied by 0.8; if d>2d0, the coefficient is multiplied by 0.

5. S46. Calculate the defensive reading ability index using the following formula: Where R is the predicted interception efficiency and K is the position correction coefficient.

7. The method for evaluating the comprehensive ability of football players based on big data according to claim 1, characterized in that, The specific steps of step S5 are as follows: S51. Encode the game context information into a vector, normalize it, and fuse the weighted feature vector as the CGAN input vector. At the same time, take the average of the player's high-order ability index in the same scenario in the last 3 games as the benchmark performance vector. S52. Build CGAN, the generator outputs the player's ability performance data under different opponent strengths, and the discriminator uses the real player's ability performance data under different opponent strengths or the player's ability performance data output by the generator to conduct adversarial training to optimize the generator. Five sets of valid simulation data were generated for each level of opponent strength, and outliers were removed. The mean value was taken as the simulation performance result at that level. S53. Calculate the deviation of creativity and the deviation of defensive reading ability. After normalization, sum the weighted deviations with a weight of 0.4:0.

6. Obtain the stability coefficient through the formula: stability coefficient = 1 - weighted deviation, where the stability coefficient ranges from [0,1]. S54. The Creativity Index, Defensive Reading Ability Index, and Stability Coefficient are weighted and summed at a ratio of 0.3:0.3:0.4 to obtain the Comprehensive Contribution Index (CCI), which is used to evaluate the player's overall ability.

8. A comprehensive ability evaluation system for football players based on big data, characterized in that, The system includes: The data acquisition module is used to collect multi-source data, including player technical data, tactical data, physical fitness data, and physiological data. The feature processing module is used to extract and standardize features from the collected multi-source data. The feature weighting module is used to obtain match context information, which includes match stage, opponent strength, tactical role, and intensity of offensive and defensive transitions. It presets the weights of technical features, tactical features, physical features, and physiological features under different match scenarios, and weights the standardized features to obtain a fused weighted feature vector. The advanced ability quantification module is used to construct an advanced ability quantification model. The model calculates the advanced ability index of the player based on the weighted data. The advanced ability index includes the creativity index and the defensive reading ability index. The adversarial scenario simulation and comprehensive evaluation module is used to construct an adversarial scenario simulator based on the conditional generative adversarial network CGAN. It uses fused weighted feature vectors and game context information as input to simulate the player's ability performance under different opponent strengths, calculates the deviation of the player's performance from the benchmark performance, determines the stability coefficient, and generates a player comprehensive contribution index by combining the higher-order ability index and the stability coefficient to evaluate the player's comprehensive ability.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the big data-based comprehensive ability assessment method for football players as described in any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the big data-based comprehensive ability assessment method for football players as described in any one of claims 1 to 7.

Images