Basketball training effect evaluation method and system based on data analysis

Abstract

The application discloses a basketball training effect evaluation method and system based on data analysis, relates to the technical field of training effect evaluation, collects physiological index data and establishes a first time axis, establishes a pass network and a defense cooperation network and establishes a second time axis, calculates pass practical effect data and defense cooperation practical effect data according to the pass network and the defense cooperation network, carries out correlation analysis on the data, outputs correlation analysis results, carries out screening according to the correlation analysis results, outputs screening results and generates an evaluation result signal, the application combines physiological indexes and practical effect data, comprehensively evaluates the basketball training effect, accurately locates training requirements, provides personalized guidance, improves training efficiency and quality, objectively reflects the training effect through data analysis, reduces subjective judgment, promotes the scientific and data-based development of basketball training, and provides strong support for the continuous progress and innovation of basketball sports.

Description

Technical Field

[0001] The present invention relates to the technical field of training effect evaluation, and in particular to a basketball training effect evaluation method and system based on data analysis. Background Art

[0002] With the rapid development of sports science and big data technology, basketball training has shifted from the traditional experience-driven model to data-driven precision training. Existing technologies mostly use a separate analysis method: on the one hand, wearable devices are used to collect athletes' physiological indicators such as body fat percentage and heart rate; on the other hand, video analysis technology is used to count performance data on the court, such as the number of passes and defensive actions.

[0003] Currently, the Chinese invention patent application number CN202410112337.5 discloses a method and system for quantitatively evaluating the effect of sports training. The method includes: storing the data of the first training mode and the second training mode to obtain training data sets for practice and assessment; obtaining a previous training data set, each training data set is a practice training data set, a psychological evaluation index of the first training mode, an assessment training data set, and a psychological evaluation index of the second training mode; using a dual-channel twin network to identify the previous training data set and the current training data set to obtain a twin training data set; outputting the first evaluation index using the current training data set and the twin training data set. This method can solve the problem of low accuracy and poor evaluation stability of training evaluation results due to the failure to consider the impact of the athlete's psychological fluctuations on the training evaluation results. It can improve the accuracy and stability of training evaluation, thereby helping athletes to develop more accurate training plans.

[0004] The above technology makes it difficult to combine physiological indicator data with actual combat effects that are not recorded manually to generate evaluation results that evaluate training effects based on actual combat effects and changes in physiological indicators, which affects the adaptability and accuracy of training effect evaluation. Summary of the Invention

[0005] The technical problem solved by the present invention is that it is difficult in the existing technology to combine physiological indicator data with actual combat effects that are not recorded manually to generate evaluation results for evaluating training effects based on actual combat effects and changes in physiological indicators, which affects the adaptability and accuracy of training effect evaluation.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] The basketball training effect evaluation method based on data analysis includes the following steps:

[0008] Step S1, collecting physiological indicator data and establishing a first time axis;

[0009] Step S2, establishing a passing network and a defensive coordination network and establishing a second timeline;

[0010] Step S3, calculating the actual passing effect data and the actual defensive coordination effect data according to the passing network and the defensive coordination network;

[0011] Step S4, performing correlation analysis on the passing actual combat effect data, the defensive coordination actual combat effect data, and the physiological index data, and outputting the correlation analysis results;

[0012] Step S5: screening the passing actual combat effect data, the defensive cooperation actual combat effect data and the physiological index data according to the correlation analysis result, outputting the screening result and generating an evaluation result signal.

[0013] Preferably, step S1 includes the following sub-steps:

[0014] Step S101, collecting physiological indicator data, wherein the physiological indicator data includes body fat percentage data, heart rate data, maximum oxygen uptake data, muscle content data and lactate threshold data;

[0015] Step S102: establishing a first time axis and storing the physiological indicator data in the time sequence of the first time axis.

[0016] Preferably, step S2 includes the following sub-steps:

[0017] Step S201: define the first node, the first edge, and the first edge weight of the passing network. The first node, the first edge, and the first edge weight are:

[0018] first nodes, each of which represents a player;

[0019] First edge: if the player represented by any first node passes the ball to the player represented by another first node, then there exists a first edge from any first node to the other first node;

[0020] a first edge weight, the first edge weight being a pass success rate, the first edge corresponding to the first edge weight;

[0021] The steps for constructing the passing network are:

[0022] Statistics on the passing success rate between players in each game;

[0023] The pass success rates are organized into an adjacency matrix and output as a pass network.

[0024] Preferably, the step S2 further includes:

[0025] Step S202: define the second node, the second edge, and the second edge weight of the defense cooperation network. The second node, the second edge, and the second edge weight are:

[0026] second nodes, each of the second nodes representing a player;

[0027] For the second side, if the player represented by any second node and the player represented by another second node jointly defend the player represented by the same second node, then there exists a second side from any second node to the other second node;

[0028] A second edge weight, where the second edge weight is a defense success rate, and the second edge corresponds to the second edge weight;

[0029] The steps for constructing the defensive collaboration network are:

[0030] Statistics on the defensive success rate between players in each game;

[0031] The defense success rate is organized into an adjacency matrix and output as a defense collaboration network;

[0032] Step S203: Establish a second time axis, and store the passing network and the defensive coordination network in the time sequence of the second time axis.

[0033] Preferably, step S3 includes the following sub-steps:

[0034] Step S301, calculating centrality index data based on the passing network and the defensive collaboration network, wherein the centrality index data includes passing network degree centrality data, defensive collaboration network degree centrality data, passing network betweenness centrality data, defensive collaboration network betweenness centrality data, passing network degree proximity data, and defensive collaboration network proximity centrality data.

[0035] Preferably, the step S3 further includes:

[0036] Step S302, calculating clustering coefficient data based on the passing network and the defensive coordination network, wherein the clustering coefficient data includes clustering coefficient data of the passing network and clustering coefficient data of the defensive coordination network;

[0037] Step S303: Merge the passing network degree centrality data, the passing network betweenness centrality data, the passing network degree proximity data, and the passing network clustering coefficient data to output as passing actual combat effect data; merge the defensive collaboration network degree centrality data, the defensive collaboration network betweenness centrality data, the defensive collaboration network degree proximity data, and the defensive collaboration network clustering coefficient data to output as defensive collaboration actual combat effect data; and insert the second time axis into the passing actual combat effect data and the defensive collaboration actual combat effect data.

[0038] Preferably, step S4 includes the following sub-steps:

[0039] Step S401: performing correlation analysis on the passing actual combat effect data, the defensive coordination actual combat effect data, and the physiological index data. The logic of the correlation analysis is:

[0040] A linear regression model was established, and the influence weights corresponding to the body fat percentage data in actual passing, the influence weights corresponding to the heart rate data in actual passing, the influence weights corresponding to the maximum oxygen uptake data in actual passing, the influence weights corresponding to the muscle content data in actual passing, the influence weights corresponding to the lactate threshold data in actual passing, the influence weights corresponding to the body fat percentage data in actual defensive cooperation, the influence weights corresponding to the heart rate data in actual defensive cooperation, the influence weights corresponding to the maximum oxygen uptake data in actual defensive cooperation, the influence weights corresponding to the muscle content data in actual defensive cooperation, and the influence weights corresponding to the lactate threshold data in actual defensive cooperation were calculated using the least squares method.

[0041] Preferably, the step S4 further includes:

[0042] Step S402: sorting the influence weights corresponding to the body fat percentage data in the actual passing game, the influence weights corresponding to the heart rate data in the actual passing game, the influence weights corresponding to the maximum oxygen uptake data in the actual passing game, the influence weights corresponding to the muscle content data in the actual passing game, and the influence weights corresponding to the lactate threshold data in the actual passing game in descending order, and outputting a physiological indicator weight sequence for the actual passing game; sorting the influence weights corresponding to the heart rate data in the actual defensive collaboration game, the influence weights corresponding to the maximum oxygen uptake data in the actual defensive collaboration game, the influence weights corresponding to the muscle content data in the actual defensive collaboration game, and the influence weights corresponding to the lactate threshold data in the actual defensive collaboration game in descending order, and outputting a physiological indicator weight sequence for the actual defensive collaboration game;

[0043] The weight sequence of the actual passing physiological index and the weight sequence of the actual defensive cooperation physiological index are output as the correlation analysis results.

[0044] Preferably, step S5 includes the following sub-steps:

[0045] Step S501: align the first time axis and the second time axis, and perform a primary screening on the actual passing effect data according to the weight sequence of the actual passing physiological index and the preset abnormal physiological index threshold range group. The logic of the primary screening is:

[0046] The physiological indicator abnormal threshold range group includes a body fat rate abnormal threshold range, a heart rate abnormal threshold range, a maximum oxygen uptake abnormal threshold range, a muscle content abnormal threshold range and a lactate threshold abnormal threshold range, and obtains the passing network degree centrality data group, the passing network betweenness centrality data group, the passing network degree proximity data group and the passing network clustering coefficient data group corresponding to the body fat rate abnormal threshold range, the heart rate abnormal threshold range, the maximum oxygen uptake abnormal threshold range, the muscle content abnormal threshold range and the lactate threshold abnormal threshold range, and calculates the average values ​​of the passing network degree centrality data group, the passing network betweenness centrality data group, the passing network degree proximity data group and the passing network clustering coefficient data group respectively, and selects the passing network degree centrality corresponding to the average value with the largest value among the average values. The data group, the passing network betweenness centrality data group, the passing network degree proximity data group, or the passing network clustering coefficient data group are used as the passing network degree centrality average, the passing network betweenness centrality average, the passing network degree proximity average, or the passing network clustering coefficient average corresponding to the body fat rate data, the heart rate data, the maximum oxygen uptake data, the muscle content data, or the lactate threshold data; a first correspondence is established between the body fat rate data, the heart rate data, the maximum oxygen uptake data, the muscle content data, the lactate threshold data, the passing network degree centrality average, the passing network betweenness centrality average, the passing network degree proximity average, and the passing network clustering coefficient average; the first correspondence is sorted in descending order according to the weight sequence of the actual passing physiological indicators; and the output is the actual passing effect evaluation result;

[0047] Step S502: Perform secondary screening on the defense cooperation actual combat effect data according to the defense cooperation actual combat physiological index weight sequence and the physiological index abnormal threshold range group. The logic of the secondary screening is:

[0048] Obtain the defense collaboration network degree centrality data group, defense collaboration network betweenness centrality data group, defense collaboration network degree proximity data group and defense collaboration network clustering coefficient data group corresponding to the abnormal threshold range of body fat rate, abnormal threshold range of heart rate, abnormal threshold range of maximum oxygen uptake, abnormal threshold range of muscle content and abnormal threshold range of lactate threshold, calculate the average values ​​of the defense collaboration network degree centrality data group, defense collaboration network betweenness centrality data group, defense collaboration network degree proximity data group and defense collaboration network clustering coefficient data group respectively, and select the defense collaboration network degree centrality data group, defense collaboration network betweenness centrality data group, defense collaboration network degree proximity data group or defense collaboration network corresponding to the average value with the largest value among the average values. The clustering coefficient data group is used as the average degree centrality of the defense collaboration network, the average betweenness centrality of the defense collaboration network, the average degree closeness of the defense collaboration network, or the average clustering coefficient of the defense collaboration network corresponding to the body fat rate data, heart rate data, maximum oxygen uptake data, muscle content data, or lactate threshold data. A second corresponding relationship is established between the body fat rate data, heart rate data, maximum oxygen uptake data, muscle content data, lactate threshold data, the average degree centrality of the defense collaboration network, the average betweenness centrality of the defense collaboration network, the average degree closeness of the defense collaboration network, and the average clustering coefficient of the defense collaboration network. The second corresponding relationship is sorted in descending order according to the weight sequence of the physiological indicators of actual defense collaboration, and the output is the evaluation result of the actual defense collaboration effect.

[0049] Step S503: Combine the passing actual effect evaluation result and the defensive coordination actual effect evaluation result to output as a screening result and generate an evaluation result signal.

[0050] A basketball training effect evaluation system based on data analysis, which is applied to the basketball training effect evaluation method based on data analysis, includes a data acquisition module, a network establishment module, a data calculation module, a correlation analysis module and a result evaluation module;

[0051] The data acquisition module is used to collect physiological indicator data and establish a first time axis;

[0052] The network establishment module is used to establish a passing network and a defensive cooperation network and establish a second time axis;

[0053] The data calculation module is used to calculate the passing actual combat effect data and the defensive coordination actual combat effect data according to the passing network and the defensive coordination network, and insert the second time axis into the passing actual combat effect data and the defensive coordination actual combat effect data;

[0054] The correlation analysis module is used to perform correlation analysis on the passing actual combat effect data, the defensive cooperation actual combat effect data and the physiological index data, and output the correlation analysis results;

[0055] The result evaluation module is used to screen the passing actual effect data, the defensive cooperation actual effect data and the physiological index data according to the correlation analysis result, output the screening result and generate an evaluation result signal.

[0056] The beneficial effects of the present invention are as follows: The present invention combines physiological indicators with actual combat effect data to comprehensively evaluate the effect of basketball training, accurately locate training needs, provide personalized guidance, improve training efficiency and quality, and through data analysis, realize objective reflection of training effects, reduce subjective judgment, promote the development of basketball training in a scientific and data-based direction, and provide strong support for the continuous progress and innovation of basketball. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A flowchart of the steps of a basketball training effect evaluation method based on data analysis provided by one embodiment of the present invention;

[0058] Figure 2 A schematic diagram of the basic flow of a basketball training effect evaluation system based on data analysis provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0060] Example 1, reference Figure 1 , provides a basketball training effect evaluation method based on data analysis, including the following steps:

[0061] Step S1: Collect physiological indicator data and establish a first time axis.

[0062] Step S2: establishing a passing network and a defensive coordination network and establishing a second timeline.

[0063] Step S3, calculating the passing actual combat effect data and the defensive coordination actual combat effect data according to the passing network and the defensive coordination network, and inserting the second time axis into the passing actual combat effect data and the defensive coordination actual combat effect data.

[0064] Step S4: performing correlation analysis on the passing actual combat effect data, the defensive coordination actual combat effect data and the physiological index data, and outputting the correlation analysis result.

[0065] Step S5: screening the passing actual combat effect data, the defensive cooperation actual combat effect data and the physiological index data according to the correlation analysis result, outputting the screening result and generating an evaluation result signal.

[0066] Step S1 includes the following sub-steps:

[0067] Step S101, collect physiological index data, including body fat rate data, heart rate data, maximum oxygen uptake data, muscle content data and lactate threshold data.

[0068] Step S101 ensures that the athlete's body fat rate data, heart rate data, maximum oxygen uptake data, muscle content data and lactate threshold data are obtained, which can reflect the athlete's physical condition and health status. Through professional measuring equipment and standardized measuring method, it is ensured that the collected physiological index data is accurate and reliable.

[0069] Step S102, establish a first time axis, and store the physiological index data in time sequence according to the first time axis.

[0070] Step S102 provides a clear time frame for physiological index data, facilitating subsequent data analysis and processing. The physiological index data is stored in the database in time sequence, ensuring data continuity and traceability, and providing convenience for subsequent data correlation analysis and training effect evaluation.

[0071] Step S1 aims to comprehensively collect and orderly store the physiological index data of basketball players, providing a basis for subsequent data analysis and training effect evaluation. Through this step, all key physiological indicators can be accurately recorded and arranged in time sequence, facilitating subsequent data processing and correlation analysis.

[0072] Step S2 includes the following sub-steps:

[0073] Step S201, define the first node, first edge and first edge weight of the pass network, the first node, first edge and first edge weight are:

[0074] The first node, each first node represents a player.

[0075] The first edge, if any first node represents a player passing to another first node representing a player, there is a first edge from any first node to another first node.

[0076] The first edge weight, the first edge weight is the pass success rate, and the first edge corresponds to the first edge weight.

[0077] The construction steps of the pass network are:

[0078] Statistical pass success rate between players in each game.

[0079] Arrange the pass success rate as an adjacency matrix and output as a pass network.

[0080] Step S201 defines the nodes, edges, and edge weights in the passing network, providing a theoretical basis for constructing the passing network. By counting the passing success rate in the game and organizing it into an adjacency matrix, the passing relationship and efficiency between players are intuitively displayed, providing data support for analyzing the passing ability of players and team cooperation.

[0081] Step S2 further includes:

[0082] Step S202: define the second node, the second edge, and the second edge weight of the defense cooperation network. The second node, the second edge, and the second edge weight are:

[0083] The second nodes each represent a player.

[0084] For the second side, if the player represented by any second node and the player represented by another second node jointly defend the player represented by the same second node, then there exists a second side from any second node to the other second node.

[0085] The second side weight, the second side weight is the defense success rate, and the second side corresponds to the second side weight.

[0086] The steps to build a defensive collaboration network are:

[0087] Statistics on the defensive success rate between players in each game.

[0088] The defense success rates are organized into an adjacency matrix and output as a defense collaboration network.

[0089] Step S202 defines the nodes, edges, and edge weights in the defensive coordination network, providing a theoretical framework for constructing the defensive coordination network. By statistically analyzing the defensive success rate in the game and organizing it into an adjacency matrix, the defensive coordination patterns and effects among players are revealed, which helps to evaluate the defensive ability and team coordination level of the players.

[0090] Step S203: Establish a second time axis, and store the passing network and the defensive coordination network in the time sequence of the second time axis.

[0091] Step S203 provides a time dimension for the passing network and the defensive coordination network to ensure the orderly storage and traceability of the network data. The passing network and the defensive coordination network data are stored in the database in chronological order to facilitate subsequent data analysis and comparison, and provide time series data support for evaluating the training effect.

[0092] Step S2 aims to comprehensively capture the interactive behaviors between players in basketball games by constructing passing networks and defensive coordination networks, and to store these network data in an orderly manner for subsequent analysis. This provides a data basis for an in-depth understanding of the players' passing efficiency and defensive coordination capabilities in the game, and helps to evaluate the effect of training on improving players' tactical understanding and teamwork capabilities.

[0093] Step S3 includes the following sub-steps:

[0094] Step S301, calculate centrality index data based on the passing network and the defensive collaboration network, the centrality index data including passing network degree centrality data, defensive collaboration network degree centrality data, passing network betweenness centrality data, defensive collaboration network betweenness centrality data, passing network degree proximity data, and defensive collaboration network proximity centrality data.

[0095] The mathematical expression of the degree centrality data of the passing network is:

[0096] ;

[0097] in, is the degree centrality data of the passing network, is the first node corresponding to the player to be evaluated, The first node in the passing network The number of connections, is the total number of players;

[0098] The mathematical expression of the degree centrality data of the defense collaboration network is:

[0099] ;

[0100] in, To defend the degree centrality data of collaborative networks, is the second node corresponding to the player to be evaluated, The second node in the passing network The number of connections.

[0101] The mathematical expression of the betweenness centrality data of the passing network is:

[0102] ;

[0103] in, is the betweenness centrality data of the passing network, and is the first node corresponding to other players in the passing network except the player to be evaluated, The first node passed by the passing network The number of shortest paths, From the first node in the passing network To the first node The number of shortest paths.

[0104] The mathematical expression of the betweenness centrality data of the defensive collaboration network is:

[0105] ;

[0106] wherein, is the defensive cooperation network betweenness centrality data, and is the second node corresponding to the other player in the defensive cooperation network except the player to be evaluated, is the number of shortest paths passing through the second node in the defensive cooperation network, is the number of shortest paths from the second node to the second node in the defensive cooperation network.

[0107] The mathematical expression of the pass network degree closeness data is:

[0108] ;

[0109] wherein, is the pass network degree closeness data, is the first node in the pass network except the first node , the first node and the first node , is the shortest path length from the first node to the first node in the pass network.

[0110] The mathematical expression of the defensive cooperation network degree closeness data is:

[0111] ;

[0112] wherein, is the defensive cooperation network degree closeness data, is the second node in the defensive cooperation network except the second node , the second node and the second node , is the shortest path length from the second node to the second node in the defensive cooperation network.

[0113] Step S301 can evaluate the pass and defensive cooperation activity, influence and proximity to the core position of the player in the game by calculating the degree centrality, betweenness centrality and degree closeness data in the pass network and the defensive cooperation network, which helps to understand the key role and influence of the player in the game.

[0114] Step S3 further comprises:

[0115] Step S302 : calculating clustering coefficient data based on the passing network and the defensive coordination network. The clustering coefficient data includes the passing network clustering coefficient data and the defensive coordination network clustering coefficient data.

[0116] The mathematical expression of the clustering coefficient data of the passing network is:

[0117]

[0118] in, is the clustering coefficient data of the passing network, The first node The degree, The first node The actual number of edges between it and other adjacent first nodes.

[0119] The mathematical expression of the clustering coefficient data of the defensive collaboration network is:

[0120]

[0121] in, To defend the clustering coefficient data of the collaborative network, For the second node The degree, For the second node The actual number of edges between it and other adjacent second nodes.

[0122] The calculation of the clustering coefficient data in step S302 can reveal the local closeness of the players in the passing and defensive coordination networks. A high clustering coefficient means that the passing or defensive coordination between the player and his teammates is closer, which helps to evaluate the player's teamwork ability.

[0123] Step S303: Merge the passing network degree centrality data, the passing network betweenness centrality data, the passing network degree proximity data, and the passing network clustering coefficient data to output as passing actual combat effect data; merge the defensive collaboration network degree centrality data, the defensive collaboration network betweenness centrality data, the defensive collaboration network degree proximity data, and the defensive collaboration network clustering coefficient data to output as defensive collaboration actual combat effect data; and insert the second time axis into the passing actual combat effect data and the defensive collaboration actual combat effect data.

[0124] Step S303 combines the various indicator data in the passing network and the defensive coordination network and outputs them as actual combat effect data, so that the player's performance can be presented more intuitively and comprehensively. At the same time, inserting a second time axis into the data helps to analyze the performance changes of players in different time periods and provide the coaching team with a more detailed player performance analysis.

[0125] Step S3 uses complex network analysis methods to comprehensively evaluate players' passing and defensive coordination performance during the game. By calculating various centrality metrics and clustering coefficients within the passing and defensive coordination networks, Step S3 generates detailed data reflecting the players' actual performance, including actual passing and defensive coordination performance data. This provides a valuable reference for subsequent player performance analysis, tactical development, and player selection.

[0126] Step S4 includes the following sub-steps:

[0127] Step S401: performing correlation analysis on the passing actual combat effect data, the defensive coordination actual combat effect data, and the physiological index data. The logic of the correlation analysis is:

[0128] Establish a linear regression model, the linear regression model is:

[0129]

[0130] in, For the actual effect data of passing, is the preset constant term for the actual passing effect. For body fat percentage data, is the influence weight corresponding to the body fat percentage data in actual passing, For heart rate data, is the influence weight corresponding to the actual passing center rate data, is the maximum oxygen uptake data, is the influence weight corresponding to the maximum oxygen uptake data in actual passing, is the muscle content data, is the influence weight corresponding to the muscle content data in actual passing, is the lactate threshold data, It is the influence weight corresponding to the lactate threshold data in actual passing.

[0131]

[0132] in, For the actual combat effect data of defensive cooperation, is the preset constant term for the actual combat effect of defensive cooperation, The corresponding influence weight of body fat percentage data in actual defensive cooperation. is the impact weight corresponding to the center rate data of the actual defense cooperation, The influence weight corresponding to the maximum oxygen uptake data in the actual defensive cooperation, The influence weight of muscle content data in actual defensive cooperation, It is the influence weight corresponding to the lactate threshold data in actual defensive cooperation.

[0133] The influence weights corresponding to the body fat percentage data in actual passing, the influence weights corresponding to the heart rate data in actual passing, the influence weights corresponding to the maximum oxygen uptake data in actual passing, the influence weights corresponding to the muscle content data in actual passing, the influence weights corresponding to the lactate threshold data in actual passing, the influence weights corresponding to the body fat percentage data in actual defensive cooperation, the influence weights corresponding to the heart rate data in actual defensive cooperation, the influence weights corresponding to the maximum oxygen uptake data in actual defensive cooperation, the influence weights corresponding to the muscle content data in actual defensive cooperation, and the influence weights corresponding to the lactate threshold data in actual defensive cooperation are calculated by the least squares method.

[0134] Step S401 establishes a linear regression model and uses the least squares method to calculate the weight of each physiological indicator's impact on actual passing and defensive coordination. This can quantify the specific contribution of each physiological indicator to game performance and provide data support for subsequent weight ranking. Through this method, the coaching team can clearly see which physiological indicators have a significant impact on the players' game performance, thereby conducting more targeted training and adjustments.

[0135] Step S4 further includes:

[0136] Step S402, sort the influence weights corresponding to the body fat percentage data in actual passing, the influence weights corresponding to the heart rate data in actual passing, the influence weights corresponding to the maximum oxygen uptake data in actual passing, the influence weights corresponding to the muscle content data in actual passing, and the influence weights corresponding to the lactate threshold data in actual passing in descending order, and output a weight sequence of physiological indicators in actual passing; sort the influence weights corresponding to the heart rate data in actual defensive cooperation, the influence weights corresponding to the maximum oxygen uptake data in actual defensive cooperation, the influence weights corresponding to the muscle content data in actual defensive cooperation, and the influence weights corresponding to the lactate threshold data in actual defensive cooperation in descending order, and output a weight sequence of physiological indicators in actual defensive cooperation.

[0137] The weight sequence of the actual passing physiological index and the weight sequence of the actual defensive cooperation physiological index are output as the correlation analysis results.

[0138] Step S402 sorts the calculated influence weights in descending order and outputs a weighted sequence of physiological indicators related to passing and defensive coordination. This allows the coaching team to intuitively identify which physiological indicators have the greatest impact on game performance, allowing them to more accurately formulate training plans. This also provides athletes with a clear training direction, enabling them to improve their physiological qualities in a targeted manner to enhance their game performance.

[0139] Step S4 aims to explore the relationship between passing effectiveness, defensive coordination effectiveness, and player physiological indicators through correlation analysis. By calculating the weight of each physiological indicator's impact on passing effectiveness and defensive coordination effectiveness, step S4 can reveal which physiological indicators have the greatest impact on a player's performance during the game. This helps coaches and athletes better understand how physiological factors influence performance, allowing them to develop more scientific training and diet plans to improve their overall competitive level.

[0140] Step S5 includes the following sub-steps:

[0141] Step S501: align the first time axis and the second time axis, and perform a screening on the actual passing effect data according to the weight sequence of the actual passing physiological index and the preset abnormal physiological index threshold range group. The logic of the screening is:

[0142] The physiological indicator abnormal threshold range group includes the body fat rate abnormal threshold range, the heart rate abnormal threshold range, the maximum oxygen uptake abnormal threshold range, the muscle content abnormal threshold range and the lactate threshold abnormal threshold range. The passing network degree centrality data group, the passing network betweenness centrality data group, the passing network degree proximity data group and the passing network clustering coefficient data group corresponding to the body fat rate abnormal threshold range, the heart rate abnormal threshold range, the maximum oxygen uptake abnormal threshold range, the muscle content abnormal threshold range and the lactate threshold abnormal threshold range are obtained. The average values ​​of the passing network degree centrality data group, the passing network betweenness centrality data group, the passing network degree proximity data group and the passing network clustering coefficient data group are calculated respectively, and the passing network degree centrality data group corresponding to the average value with the largest value among the average values ​​is selected. The data group, the passing network betweenness centrality data group, the passing network degree proximity data group or the passing network clustering coefficient data group are used as the passing network degree centrality average value, the passing network betweenness centrality average value, the passing network degree proximity average value or the passing network clustering coefficient average value corresponding to the body fat rate data, heart rate data, maximum oxygen uptake data, muscle content data or lactate threshold data, and a first correspondence is established between the body fat rate data, heart rate data, maximum oxygen uptake data, muscle content data, lactate threshold data, the passing network degree centrality average value, the passing network betweenness centrality average value, the passing network degree proximity average value and the passing network clustering coefficient average value. The first correspondence is sorted in descending order according to the weight sequence of the actual passing physiological indicators, and the output is the passing actual effect evaluation result.

[0143] Step S501 ensures the temporal synchronization of the passing performance data and the physiological indicator data by aligning the first and second timelines. The passing performance data is then screened based on the passing performance physiological indicator weight sequence and the preset physiological indicator abnormality threshold range. This screening process takes into account the players' actual passing performance and physiological state. By calculating the average of the passing network centrality data corresponding to each physiological indicator, a first correspondence relationship is established between the physiological indicators and the passing network centrality data. Finally, the first correspondence relationship is sorted in descending order based on the passing performance physiological indicator weight sequence, and the passing performance evaluation results are output, which helps the coaching team understand the players' passing strengths and weaknesses, as well as the relationship between these performances and physiological states.

[0144] Step S502: Perform secondary screening on the defense cooperation actual combat effect data according to the defense cooperation actual combat physiological index weight sequence and the physiological index abnormal threshold range group. The logic of the secondary screening is:

[0145] Obtain the defense collaboration network degree centrality data group, defense collaboration network betweenness centrality data group, defense collaboration network degree proximity data group and defense collaboration network clustering coefficient data group corresponding to the abnormal threshold range of body fat rate, abnormal threshold range of heart rate, abnormal threshold range of maximum oxygen uptake, abnormal threshold range of muscle content and abnormal threshold range of lactate threshold, calculate the average values ​​of the defense collaboration network degree centrality data group, defense collaboration network betweenness centrality data group, defense collaboration network degree proximity data group and defense collaboration network clustering coefficient data group respectively, and select the defense collaboration network degree centrality data group, defense collaboration network betweenness centrality data group, defense collaboration network degree proximity data group or defense collaboration network corresponding to the average value with the largest value among the average values. The clustering coefficient data group is used as the average degree centrality of the defense collaboration network, the average betweenness centrality of the defense collaboration network, the average degree closeness of the defense collaboration network, or the average clustering coefficient of the defense collaboration network corresponding to the body fat rate data, heart rate data, maximum oxygen uptake data, muscle content data, or lactate threshold data. A second corresponding relationship is established among the body fat rate data, heart rate data, maximum oxygen uptake data, muscle content data, lactate threshold data, the average degree centrality of the defense collaboration network, the average betweenness centrality of the defense collaboration network, the average degree closeness of the defense collaboration network, and the average clustering coefficient of the defense collaboration network. The second corresponding relationship is sorted in descending order according to the weight sequence of the physiological indicators of actual combat of defense collaboration, and the output is the evaluation result of the actual combat effect of defense collaboration.

[0146] Step S502 is similar to step S501, but targets actual defensive coordination effectiveness data. It also considers the players' actual defensive coordination performance and physiological state. By calculating the average of the defensive coordination network centrality data corresponding to each physiological indicator, a second correspondence between the physiological indicators and the defensive coordination network centrality data is established. This second correspondence is then sorted in descending order based on the weight sequence of the defensive coordination physiological indicators, outputting the defensive coordination effectiveness evaluation results. This result helps the coaching team understand the players' defensive coordination performance and the relationship between this performance and physiological state.

[0147] Step S503: Combine the passing actual effect evaluation result and the defensive coordination actual effect evaluation result to output as a screening result and generate an evaluation result signal.

[0148] Step S503 combines the passing effectiveness evaluation results and the defensive coordination effectiveness evaluation results into a screening result and generates an evaluation result signal. This step comprehensively considers the player's passing and defensive coordination performance, providing the coaching team with a more comprehensive and intuitive evaluation result. Through the evaluation result signal, the coaching team can quickly understand the player's overall performance, as well as which players excelled or needed improvement in which areas, helping the coaching team develop more targeted training plans to improve the team's overall performance.

[0149] Step S5 aims to integrate the actual passing effect, defensive coordination effect, and the players' physiological indicator data. Through a series of screening and sorting operations, it generates detailed evaluation results of the players' passing and defensive coordination performance in the game. It not only takes into account the players' actual performance in the game, but also combines their physiological status, thereby providing a more comprehensive and in-depth evaluation. By outputting the screening results and generating the evaluation result signal, step S5 provides important reference information for the coaching team, helping them to better understand the players' status, formulate targeted training plans, and improve the team's overall performance.

[0150] This method combines physiological indicator data with actual performance data to comprehensively evaluate training effectiveness from multiple dimensions. This not only reflects changes in an athlete's physical fitness but also reveals improvements in passing efficiency, defensive coordination, and other aspects of their game, providing more comprehensive feedback on training effectiveness. Through in-depth analysis of actual passing and defensive coordination data, it can precisely identify technical and tactical shortcomings. Integrating physiological indicator data allows coaches to more scientifically formulate targeted training plans to help athletes effectively improve their fitness and game performance. Because athletes vary in their physiological characteristics and skill levels, this method, through correlation analysis of physiological indicator data and actual performance data, can provide personalized training guidance for each athlete. This helps athletes tailor their training strategies to their individual circumstances, achieving more efficient and personalized training. Traditional training effectiveness evaluation methods often rely on coaches' subjective judgment and accumulated experience, which can be subject to certain uncertainties. This method, based on data analysis, objectively and accurately reflects training effectiveness, enabling coaches to adjust training plans in a timely manner and improve training efficiency and quality. The application of this method marks a step forward in the scientific and data-driven development of basketball training. By deeply exploring and analyzing game data and physiological indicator data, we can provide richer data support and theoretical basis for basketball training and research, and promote the continuous development and innovation of basketball.

[0151] Example 2, reference Figure 2 , provides a basketball training effect evaluation system based on data analysis, including data acquisition module, network establishment module, data calculation module, correlation analysis module and result evaluation module.

[0152] The data acquisition module is used to collect physiological indicator data and establish a first time axis.

[0153] The data acquisition module efficiently and accurately collects players' physiological indicators and establishes a primary timeline. This data forms the foundation for subsequent analysis and is crucial for assessing players' physical condition and training effectiveness. Real-time, continuous data collection ensures the timeliness and accuracy of the data, providing strong support for subsequent in-depth analysis.

[0154] The network building module is used to build a passing network and a defensive coordination network and to establish a second timeline.

[0155] The network building module is responsible for establishing the passing network and defensive coordination network, and constructing a secondary timeline that clearly displays the players' passing and defensive coordination behaviors during the game, as well as how these behaviors change over time. By building a network model, it captures the interactions and coordination patterns between players, providing key information for subsequent data calculations and performance evaluation. Furthermore, the establishment of a secondary timeline helps align network data with physiological indicator data, facilitating correlation analysis.

[0156] The data calculation module is used to calculate the passing actual combat effect data and the defensive coordination actual combat effect data according to the passing network and the defensive coordination network, and insert the second time axis into the passing actual combat effect data and the defensive coordination actual combat effect data.

[0157] The data calculation module calculates the actual passing effect data and the actual defensive coordination effect data based on the passing network and the defensive coordination network, and inserts the second time axis into these data, which can quantify the player's performance in the game. By calculating data such as centrality indicators and clustering coefficients, it reveals the player's activity and influence in passing and defensive coordination. At the same time, inserting the time axis into the data helps analyze the changes in player performance in different time periods, providing the coaching team with a more detailed player performance analysis.

[0158] The correlation analysis module is used to perform correlation analysis on the actual passing effect data, the actual defensive cooperation effect data and the physiological index data, and output the correlation analysis results.

[0159] The correlation analysis module performs correlation analysis on passing effectiveness data, defensive coordination effectiveness data, and physiological indicator data, and outputs correlation analysis results, which can reveal the potential connection between physiological indicators and game performance. By calculating the influence weight of each physiological indicator on game performance, it provides the coaching team with information on which physiological factors have a significant impact on player performance, helping the coaching team to develop more scientific training and diet plans to improve the overall competitive level of the players.

[0160] The result evaluation module is used to screen the passing actual effect data, defensive cooperation actual effect data and physiological index data according to the correlation analysis results, output the screening results and generate an evaluation result signal.

[0161] The result evaluation module filters actual passing effectiveness data, actual defensive coordination effectiveness data, and physiological indicator data based on the correlation analysis results, outputs the screening results, and generates an evaluation result signal. This module comprehensively considers the players' game performance and physiological state. Through screening and sorting operations, it generates detailed evaluation results of the players' passing and defensive coordination performance during the game. This helps the coaching team quickly understand the players' overall performance, as well as which players excelled or needed improvement in which areas, thereby developing more targeted training plans to improve the team's overall performance. Furthermore, the generation of the evaluation result signal allows the coaching team to intuitively see the evaluation results, facilitating decision-making and adjustments.

[0162] Example 3 is another embodiment of the present invention. This embodiment is different from Example 1 and Example 2 in that it provides an experimental verification of a basketball training effect evaluation method based on data analysis. In order to verify and illustrate the technical effects used in this system, this embodiment uses a traditional technical solution to conduct comparative tests with the present invention, and compares the test results by means of scientific demonstration to verify the actual effect of the present invention.

[0163] In step S1, as shown in Table 1, physiological indicator data of five players (A, B, C, D, and E) are collected. The physiological indicator data includes body fat percentage data, heart rate data, maximum oxygen uptake data, muscle content data, and lactate threshold data. Timestamps are inserted for subsequent establishment of a first timeline.

[0164] Table 1:

[0165]

[0166] When the passing network is established in step S2, the passing success rate between players in a game is counted, as shown in Table 2:

[0167] Table 2:

[0168]

[0169] Construct the adjacency matrix:

[0170]

[0171] When establishing the defensive cooperation network in step S3, the success rate of defensive cooperation between players in a game is counted, as shown in Table 3:

[0172] Table 3:

[0173]

[0174] Construct the adjacency matrix:

[0175]

[0176] In the calculation of step S3, taking player C as an example:

[0177] The degree centrality data of the passing network is:

[0178] ;

[0179] The betweenness centrality data of the passing network is (here C is the key path A→C→D):

[0180] ;

[0181] The pass network proximity data is:

[0182] ;

[0183] The clustering coefficient data of the passing network is (there is no direct connection between the neighbors of C here):

[0184]

[0185] Take player D as an example:

[0186] The degree centrality data of the defense collaboration network is:

[0187] ;

[0188] The betweenness centrality data of the defense collaboration network is:

[0189] ;

[0190] The defensive collaboration network proximity data is:

[0191] ;

[0192] The clustering coefficient data of the defense collaboration network is:

[0193]

[0194] Step S5 uses a linear regression model to analyze the relationship between heart rate and the betweenness centrality of the passing network as an example:

[0195]

[0196] Virtual regression results:

[0197] (p < 0.05), indicating that for every 1 bpm increase in heart rate, betweenness centrality increases by 0.03;

[0198] In actual basketball games, the data related to the passing network and the defensive coordination network can reflect the team's performance in offense and defense:

[0199] Passing network related data:

[0200] Passing network degree centrality data:

[0201] Meaning: The passing network centrality data reflects the degree of direct connection between players in the passing network, that is, the ability of a player to establish passing connections with other players.

[0202] Actual performance: Players with high degree centrality are usually the passing core of the team. They can connect the team's offense through passing and create scoring opportunities for their teammates.

[0203] Passing network betweenness centrality data:

[0204] Meaning: The betweenness centrality of a passing network measures a player's ability to connect other players as an intermediary in the passing network. Players with high betweenness centrality play a bridging role in the team's offense, and their passes often break down the opponent's defensive layout.

[0205] Actual performance: Players with high betweenness centrality can flexibly mobilize teammates during the game, find defensive loopholes in the opponent through accurate passes, and create scoring opportunities for the team.

[0206] Passing network proximity data:

[0207] Meaning: Passing Network Proximity reflects a player's proximity to other players in the passing network, i.e., the player's global accessibility within the passing network. Players with high Proximity are able to establish passing connections with other players more quickly, improving the team's offensive efficiency.

[0208] Actual performance: Players with high accessibility can quickly integrate into the team's offensive system during the game, form tacit cooperation with teammates, and improve the team's overall offensive level.

[0209] Pass network clustering coefficient data:

[0210] Meaning: The clustering coefficient of a passing network measures the closeness of the passing relationships between players. A high clustering coefficient means that the passes between players are more frequent and close, which is conducive to the team's effective offensive coordination.

[0211] Actual performance: Teams with a high clustering coefficient can show smoother offensive coordination in the game, and the passing between players is more tacit and accurate, thereby improving the team's scoring ability.

[0212] Defense collaboration network related data:

[0213] Defensive collaboration network degree centrality data:

[0214] Meaning: The defensive cooperation network degree centrality data reflects the player's direct connection degree in the defensive cooperation network, that is, the player's ability to establish cooperative relationships with teammates in defense.

[0215] Performance in actual combat: Players with high degree centrality can actively cooperate with teammates in defense, jointly defend against opponents' attacks, and improve the team's defensive stability.

[0216] Defensive cooperation network betweenness centrality data:

[0217] Meaning: The defensive cooperation network betweenness centrality data measures the player's ability to connect other players as an intermediary in the defensive cooperation network. Players with high betweenness centrality can coordinate the defensive positions of teammates and form an effective defensive formation in defense.

[0218] Performance in actual combat: Players with high betweenness centrality can quickly identify the offensive intentions of opponents during the game, and disrupt the offensive rhythm of opponents by directing and coordinating the defensive positions of teammates.

[0219] Defensive cooperation network degree closeness data:

[0220] Meaning: The defensive cooperation network degree closeness data reflects the player's closeness to other players in the defensive cooperation network. Players with high closeness can more quickly establish cooperative relationships with other defensive players, improving the team's defensive efficiency.

[0221] Performance in actual combat: Players with high closeness can quickly integrate into the team's defense system and form a tight defensive network with teammates, effectively limiting the scoring opportunities of opponents.

[0222] Defensive cooperation network clustering coefficient data:

[0223] Meaning: The defensive cooperation network clustering coefficient data measures the tightness of the defensive cooperation relationship between players. A high clustering coefficient of the defensive cooperation network means that the defensive cooperation between players is more tacit and tight, which is conducive to the team forming an effective defensive system.

[0224] Performance in actual combat: Teams with high clustering coefficients can exhibit more rigorous defensive cooperation during the game, and the defensive cooperation between players is more tacit and coordinated, thereby improving the team's defensive ability.

[0225] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0226] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A basketball training effect evaluation method based on data analysis, characterized in that: The steps include: Step S1, collecting physiological indicator data and establishing a first time axis; Step S2, establishing a passing network and a defensive coordination network and establishing a second timeline; Step S3, calculating the actual passing effect data and the actual defensive coordination effect data according to the passing network and the defensive coordination network; Step S3 includes the following sub-steps: Step S301, calculating centrality index data based on the passing network and the defensive coordination network, the centrality index data including passing network degree centrality data, defensive coordination network degree centrality data, passing network betweenness centrality data, defensive coordination network betweenness centrality data, passing network degree closeness data, and defensive coordination network closeness centrality data; The mathematical expression of the degree centrality data of the passing network is: ； in, is the degree centrality data of the passing network, is the first node corresponding to the player to be evaluated, The first node in the passing network The number of connections, is the total number of players; The mathematical expression of the degree centrality data of the defense collaboration network is: ； in, To defend the degree centrality data of collaborative networks, is the second node corresponding to the player to be evaluated, The second node in the passing network Number of connections; The mathematical expression of the betweenness centrality data of the passing network is: ； in, is the betweenness centrality data of the passing network, and is the first node corresponding to other players in the passing network except the player to be evaluated, The first node passed by the passing network The number of shortest paths, From the first node in the passing network To the first node The number of shortest paths; The mathematical expression of the betweenness centrality data of the defensive collaboration network is: ； in, To defend the collaborative network betweenness centrality data, and is the second node corresponding to the players other than the player to be evaluated in the defensive collaboration network, To defend the collaborative network through the second node The number of shortest paths, To defend the collaborative network from the second node To the second node The number of shortest paths; The mathematical expression of the passing network proximity data is: ； in, is the pass network proximity data, Remove the first node from the passing network , first node and the first node The first node outside The first node in the passing network To the first node The shortest path length; The mathematical expression of the proximity data of the defensive collaboration network is: ； in, To defend collaborative network proximity data, To defend the collaborative network, remove the second node , second node and the second node The second node outside To defend the second node in the collaborative network To the second node The shortest path length; Step S302, calculating clustering coefficient data based on the passing network and the defensive coordination network, where the clustering coefficient data includes the passing network clustering coefficient data and the defensive coordination network clustering coefficient data; The mathematical expression of the clustering coefficient data of the passing network is: in, is the clustering coefficient data of the passing network, The first node The degree, The first node The actual number of edges between the first nodes and other adjacent nodes; The mathematical expression of the clustering coefficient data of the defensive collaboration network is: in, To defend the clustering coefficient data of the collaborative network, For the second node The degree, For the second node The actual number of edges between the node and the other adjacent second nodes; Step S303: Merge the passing network degree centrality data, the passing network betweenness centrality data, the passing network degree proximity data, and the passing network clustering coefficient data to output as passing actual combat effect data; merge the defensive coordination network degree centrality data, the defensive coordination network betweenness centrality data, the defensive coordination network degree proximity data, and the defensive coordination network clustering coefficient data to output as defensive coordination actual combat effect data; and insert the second time axis into the passing actual combat effect data and the defensive coordination actual combat effect data. Step S4, performing correlation analysis on the passing actual combat effect data, the defensive coordination actual combat effect data, and the physiological index data, and outputting the correlation analysis results; Step S5: screening the passing actual combat effect data, the defensive cooperation actual combat effect data and the physiological index data according to the correlation analysis result, outputting the screening result and generating an evaluation result signal.

2. The basketball training effect evaluation method based on data analysis according to claim 1, characterized in that: The step S1 includes the following sub-steps: Step S101, collecting physiological indicator data, wherein the physiological indicator data includes body fat percentage data, heart rate data, maximum oxygen uptake data, muscle content data and lactate threshold data; Step S102: establishing a first time axis and storing the physiological indicator data in the time sequence of the first time axis.

3. The basketball training effect evaluation method based on data analysis according to claim 2, characterized in that: The step S2 includes the following sub-steps: Step S201: define the first node, the first edge, and the first edge weight of the passing network. The first node, the first edge, and the first edge weight are: first nodes, each of which represents a player; First edge: if the player represented by any first node passes the ball to the player represented by another first node, then there exists a first edge from any first node to the other first node; a first edge weight, the first edge weight being a pass success rate, the first edge corresponding to the first edge weight; The steps for constructing the passing network are: Statistics on the passing success rate between players in each game; The pass success rates are organized into an adjacency matrix and output as a pass network.

4. The basketball training effect evaluation method based on data analysis according to claim 3, characterized in that: The step S2 further includes: Step S202: define the second node, the second edge, and the second edge weight of the defense cooperation network. The second node, the second edge, and the second edge weight are: second nodes, each of the second nodes representing a player; For the second side, if the player represented by any second node and the player represented by another second node jointly defend the player represented by the same second node, then there exists a second side from any second node to the other second node; A second edge weight, where the second edge weight is a defense success rate, and the second edge corresponds to the second edge weight; The steps for constructing the defensive collaboration network are: Statistics on the defensive success rate between players in each game; The defense success rate is organized into an adjacency matrix and output as a defense collaboration network; Step S203: Establish a second time axis, and store the passing network and the defensive coordination network in the time sequence of the second time axis.

5. The basketball training effect evaluation method based on data analysis according to claim 4, characterized in that: The step S4 includes the following sub-steps: Step S401: performing correlation analysis on the passing actual combat effect data, the defensive coordination actual combat effect data, and the physiological index data. The logic of the correlation analysis is: A linear regression model was established, and the influence weights corresponding to the body fat percentage data in actual passing, the influence weights corresponding to the heart rate data in actual passing, the influence weights corresponding to the maximum oxygen uptake data in actual passing, the influence weights corresponding to the muscle content data in actual passing, the influence weights corresponding to the lactate threshold data in actual passing, the influence weights corresponding to the body fat percentage data in actual defensive cooperation, the influence weights corresponding to the heart rate data in actual defensive cooperation, the influence weights corresponding to the maximum oxygen uptake data in actual defensive cooperation, the influence weights corresponding to the muscle content data in actual defensive cooperation, and the influence weights corresponding to the lactate threshold data in actual defensive cooperation were calculated using the least squares method.

6. The basketball training effect evaluation method based on data analysis according to claim 5, characterized in that: The step S4 further includes: Step S402: sorting the influence weights corresponding to the body fat percentage data in the actual passing game, the influence weights corresponding to the heart rate data in the actual passing game, the influence weights corresponding to the maximum oxygen uptake data in the actual passing game, the influence weights corresponding to the muscle content data in the actual passing game, and the influence weights corresponding to the lactate threshold data in the actual passing game in descending order, and outputting a physiological indicator weight sequence for the actual passing game; sorting the influence weights corresponding to the heart rate data in the actual defensive collaboration game, the influence weights corresponding to the maximum oxygen uptake data in the actual defensive collaboration game, the influence weights corresponding to the muscle content data in the actual defensive collaboration game, and the influence weights corresponding to the lactate threshold data in the actual defensive collaboration game in descending order, and outputting a physiological indicator weight sequence for the actual defensive collaboration game; The weight sequence of the actual passing physiological index and the weight sequence of the actual defensive cooperation physiological index are output as the correlation analysis results.

7. The basketball training effect evaluation method based on data analysis according to claim 6, characterized in that: The step S5 includes the following sub-steps: Step S501: align the first time axis and the second time axis, and perform a primary screening on the actual passing effect data according to the weight sequence of the actual passing physiological index and the preset abnormal physiological index threshold range group. The logic of the primary screening is: The physiological indicator abnormal threshold range group includes a body fat rate abnormal threshold range, a heart rate abnormal threshold range, a maximum oxygen uptake abnormal threshold range, a muscle content abnormal threshold range and a lactate threshold abnormal threshold range, and obtains the passing network degree centrality data group, the passing network betweenness centrality data group, the passing network degree proximity data group and the passing network clustering coefficient data group corresponding to the body fat rate abnormal threshold range, the heart rate abnormal threshold range, the maximum oxygen uptake abnormal threshold range, the muscle content abnormal threshold range and the lactate threshold abnormal threshold range, and calculates the average values ​​of the passing network degree centrality data group, the passing network betweenness centrality data group, the passing network degree proximity data group and the passing network clustering coefficient data group respectively, and selects the passing network degree centrality corresponding to the average value with the largest value among the average values. The data group, the passing network betweenness centrality data group, the passing network degree proximity data group, or the passing network clustering coefficient data group are used as the passing network degree centrality average, the passing network betweenness centrality average, the passing network degree proximity average, or the passing network clustering coefficient average corresponding to the body fat rate data, the heart rate data, the maximum oxygen uptake data, the muscle content data, or the lactate threshold data; a first correspondence is established between the body fat rate data, the heart rate data, the maximum oxygen uptake data, the muscle content data, the lactate threshold data, the passing network degree centrality average, the passing network betweenness centrality average, the passing network degree proximity average, and the passing network clustering coefficient average; the first correspondence is sorted in descending order according to the weight sequence of the actual passing physiological indicators; and the output is the actual passing effect evaluation result; Step S502: Perform secondary screening on the defense cooperation actual combat effect data according to the defense cooperation actual combat physiological index weight sequence and the physiological index abnormal threshold range group. The logic of the secondary screening is: Obtain the defense collaboration network degree centrality data group, defense collaboration network betweenness centrality data group, defense collaboration network degree proximity data group and defense collaboration network clustering coefficient data group corresponding to the abnormal threshold range of body fat rate, abnormal threshold range of heart rate, abnormal threshold range of maximum oxygen uptake, abnormal threshold range of muscle content and abnormal threshold range of lactate threshold, calculate the average values ​​of the defense collaboration network degree centrality data group, defense collaboration network betweenness centrality data group, defense collaboration network degree proximity data group and defense collaboration network clustering coefficient data group respectively, and select the defense collaboration network degree centrality data group, defense collaboration network betweenness centrality data group, defense collaboration network degree proximity data group or defense collaboration network corresponding to the average value with the largest value among the average values. The clustering coefficient data group is used as the average degree centrality of the defense collaboration network, the average betweenness centrality of the defense collaboration network, the average degree closeness of the defense collaboration network, or the average clustering coefficient of the defense collaboration network corresponding to the body fat rate data, heart rate data, maximum oxygen uptake data, muscle content data, or lactate threshold data. A second corresponding relationship is established between the body fat rate data, heart rate data, maximum oxygen uptake data, muscle content data, lactate threshold data, the average degree centrality of the defense collaboration network, the average betweenness centrality of the defense collaboration network, the average degree closeness of the defense collaboration network, and the average clustering coefficient of the defense collaboration network. The second corresponding relationship is sorted in descending order according to the weight sequence of the physiological indicators of actual defense collaboration, and the output is the evaluation result of the actual defense collaboration effect. Step S503: Combine the passing actual combat effect evaluation result and the defensive coordination actual combat effect evaluation result to output as a screening result and generate an evaluation result signal.

8. A basketball training effect evaluation system based on data analysis, which is applied to the basketball training effect evaluation method based on data analysis as described in any one of claims 1 to 7, characterized in that: It includes data acquisition module, network establishment module, data calculation module, correlation analysis module and result evaluation module; The data acquisition module is used to collect physiological indicator data and establish a first time axis; The network establishment module is used to establish a passing network and a defensive cooperation network and establish a second time axis; The data calculation module is used to calculate the passing actual combat effect data and the defensive coordination actual combat effect data according to the passing network and the defensive coordination network, and insert the second time axis into the passing actual combat effect data and the defensive coordination actual combat effect data; The correlation analysis module is used to perform correlation analysis on the passing actual combat effect data, the defensive cooperation actual combat effect data and the physiological index data, and output the correlation analysis results; The result evaluation module is used to screen the passing actual effect data, the defensive cooperation actual effect data and the physiological index data according to the correlation analysis result, output the screening result and generate an evaluation result signal.

Images