Athlete career management method and system based on artificial intelligence

Abstract

The invention discloses an athlete career management method and system based on artificial intelligence. The method comprises the steps of data acquisition, data preprocessing, athlete state evaluation, athlete potential prediction, hyper-parameter optimization and comprehensive career management. The invention belongs to the technical field of athlete career information management, and particularly relates to an athlete career management method and system based on artificial intelligence, the scheme creatively combines athlete state evaluation and potential prediction, and provides comprehensive and accurate support for athlete career management; a state evaluation activation function and a state evaluation loss function are designed, the stability of the evaluation model is improved, and the problem that the athlete state evaluation result is not accurate enough is solved; according to the scheme, the algorithm for obtaining the optimal parameters of the prediction model is improved by adopting the search adaptation factor and the neighborhood adaptive search strategy, the optimal parameter combination is obtained, the accuracy of the model output result is improved, and the potential prediction of athletes is more accurate.

Description

Technical Field

[0001] The present invention belongs to the technical field of athlete career information management, and specifically relates to an athlete career management method and system based on artificial intelligence. Background Art

[0002] An athlete career management method and system based on artificial intelligence is an intelligent solution that uses artificial intelligence technology to comprehensively and efficiently manage the athlete's career. Through scientific data analysis and model prediction, the method can accurately plan the athlete's career development path, identify the key nodes of the career peak and retirement period, provide real-time advice and long-term planning, help athletes optimize their competitive performance and prolong their careers, and support them to make wiser and more forward-looking career choices at critical stages; however, in the existing methods for athlete career management, there is a risk of over-focusing on the current athlete's status and ignoring future potential, which leads to uneven development of the athlete's career and loss of overall benefits; the limitations of activation functions and loss functions in existing models for athlete status evaluation lead to technical problems such as inaccurate athlete status evaluation results; in existing models for athlete potential prediction models, there is a technical problem of improper built-in parameter settings, which leads to inaccurate output results of the prediction model. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an athlete career management method and system based on artificial intelligence. In view of the risk problem of excessive attention to the current athlete status and neglect of future potential in the existing athlete career management method, which leads to the unbalanced development of the athlete's career and the loss of overall benefits, the present solution creatively combines the athlete status evaluation with the potential prediction, provides comprehensive and accurate support for the athlete's career management, improves the athlete's short-term competitive performance, provides scientific guidance for the athlete's long-term development, helps the athlete to smoothly achieve career transformation and development after the peak period, and ensures the continuity and stability of the athlete's career; in view of the limitations of the activation function and the loss function in the existing model applicable to the athlete status evaluation, resulting in the technical problem that the athlete status evaluation result is not accurate enough, the present solution innovatively designs the status evaluation activation function and the status evaluation loss function, improves the stability of the evaluation model, thereby solving the problem that the athlete status evaluation result is not accurate enough; in view of the technical problem that the existing model applicable to the athlete potential prediction model has improper built-in parameter settings, resulting in inaccurate output results of the prediction model, the present solution adopts the search adaptation factor and the neighborhood adaptive search strategy to improve the algorithm for obtaining the optimal parameters of the prediction model, obtains the optimal parameter combination, improves the accuracy of the model output results, and makes the athlete potential prediction more accurate.

[0004] The technical solution adopted by the present invention is as follows: The present invention provides an athlete career management method based on artificial intelligence, the method comprising the following steps:

[0005] Step S1: data collection;

[0006] Step S2: data preprocessing;

[0007] Step S3: Athlete status assessment;

[0008] Step S4: Athlete potential prediction;

[0009] Step S5: hyperparameter optimization;

[0010] Step S6: Comprehensive career management.

[0011] Further, in step S1, the data collection is specifically to obtain the original data of athlete career management from the sports event platform and the athlete personal information system by collecting; the original data of athlete career management includes reference athlete status evaluation data, reference athlete potential prediction data, target athlete status evaluation data and target athlete potential prediction data; the reference athlete status evaluation data and the target athlete status evaluation data both contain physiological data, training data and psychological state data; the reference athlete status evaluation data also includes reference athlete status; the reference athlete potential prediction data and the target athlete potential prediction data both contain historical game data, historical training data and historical injury records; the physiological data includes heart rate, blood oxygen saturation, body temperature, fatigue and blood pressure; the training data includes training time, training intensity, training type and training effect feedback; the historical game data includes game results, game type and individual performance in the game; the historical training data includes long-term training data and training effects in different training cycles; the historical injury records include injury records, health files and recovery status data.

[0012] Furthermore, in step S2, the data preprocessing is used to perform data cleaning, standardization and feature selection on the original data of athlete career management to obtain preliminary data of athlete career management; the data cleaning is to process missing values, outliers and duplicate values; the standardization is to standardize the data based on the maximum and minimum normalization method; the feature selection is to use the correlation analysis method to screen out relevant features for athlete status assessment and athlete potential prediction from the original data of athlete career management.

[0013] Further, in step S3, the athlete status evaluation is specifically to establish an athlete status evaluation model by designing a status evaluation activation function and a status evaluation loss function to obtain a target athlete evaluation status result, including the following steps:

[0014] Step S31: Establishing an athlete status assessment model, specifically including the following steps:

[0015] Step S311: Initialize model parameters, set the number of model hidden units, input vector dimension, output vector dimension and loss function threshold;

[0016] Step S312: Initialize the internal structure of the evaluation model ALM, which specifically includes the following steps:

[0017] Step S3121: Design a state evaluation activation function, the formula used is as follows:

[0018] ;

[0019] In the formula, x represents the input variable of the state evaluation activation function, represents the state evaluation activation function, represents the smoothing parameter of the activation function, Represents the scaling factor, which is used to adjust the range of the activation function output;

[0020] Step S3122: Forget gate calculation process, the formula used is as follows:

[0021] ;

[0022] In the formula, represents the output of the forget gate, It indicates input data. Indicates the hidden state at the last moment. and They represent the weight matrices corresponding to the forget gate processing input data and processing hidden states, respectively. represents the bias parameter of the forget gate, represents the Hadamard product;

[0023] Step S3123: Input gate calculation process, the formula used is as follows:

[0024] ;

[0025] ;

[0026] ;

[0027] In the formula, represents the output of the input gate, and Respectively represent the weight matrices corresponding to the input gate processing input data and processing hidden states, represents the bias parameter of the input gate, Indicates the current cell state. represents the candidate value of the new cell state, It indicates the cell state at the last moment. and Respectively represent the weight matrices corresponding to the cell state processing input data and processing hidden state, The bias parameter representing the cell state, represents element-wise multiplication;

[0028] Step S3124: Output gate calculation process, the formula used is as follows:

[0029] ;

[0030] In the formula, represents the output of the output gate, and They represent the weight matrices corresponding to the output gate processing input data and processing hidden states respectively; Represents the bias parameter of the output gate;

[0031] Step S3125: hidden state calculation process, the formula used is as follows:

[0032] ;

[0033] In the formula, Indicates the hidden state at the current moment;

[0034] Step S313: Calculate the bidirectional ALM using the following formula:

[0035] ;

[0036] ;

[0037] ;

[0038] In the formula, represents the hidden state of the combined forward and backward ALM, Indicates the hidden state in the positive direction at the current moment, Indicates the hidden state in the positive direction at the previous moment, Indicates the reverse hidden state at the current moment. Indicates the hidden state in the reverse direction at the next moment;

[0039] Step S314: Get the final output value, the formula used is as follows:

[0040] ;

[0041] In the formula, represents the output value of the model, represents the weight matrix corresponding to the output, Represents the bias parameter of the output layer;

[0042] Step S315: Design a state evaluation loss function, the formula used is as follows:

[0043] ;

[0044] In the formula, represents the state evaluation loss function value, Represents the steepness of the control loss function, Represents the parameter for accelerating convergence, which is used to adjust the extreme points of the derivative. Indicates the actual evaluation status of the athlete at the current time;

[0045] Step S316: Update parameters, specifically use the gradient descent method to update all weight matrices and bias items in the athlete status assessment model; the formula used is as follows:

[0046] ;

[0047] ;

[0048] In the formula, Denotes the loss function for the weight matrix The gradient of Denotes the loss function to represent the bias parameter The gradient of represents the updated weight matrix, represents the current weight matrix, represents the updated bias parameter, Represents the current bias parameter, represents the learning rate;

[0049] Step S317: Iteration determination, specifically monitoring the change of the loss function. If the loss function is lower than the set loss function threshold, the iterative training is stopped. Otherwise, the iterative training is continued until the maximum number of iterations is reached.

[0050] Step S32: training the athlete status assessment model, specifically using the reference athlete status assessment data in the athlete career management preliminary data as input data to train the athlete status assessment model to obtain a trained athlete status assessment model;

[0051] Step S33: Athlete status assessment, specifically, using the target athlete status assessment data in the preliminary athlete career management data as input data of the trained athlete status assessment model to obtain the target athlete status assessment result.

[0052] Further, in step S4, the athlete potential prediction is specifically performed by establishing an athlete potential prediction model and training the model, and performing target athlete potential prediction on the trained prediction model input data to obtain the target athlete's future potential prediction result; comprising the following steps:

[0053] Step S41: Establishing an athlete potential prediction model, specifically including the following steps:

[0054] Step S411: Calculate the hidden layer input using the following formula:

[0055] ;

[0056] In the formula, represents the hidden layer input value, represents the weight from the input layer to the hidden layer, Represents the input value of the i-th input neuron;

[0057] Step S412: Calculate the output of the hidden layer using the following formula:

[0058] ;

[0059] In the formula, represents the output value of the hidden layer neuron, Bias term of hidden layer;

[0060] Step S413: Calculate the output layer input using the following formula:

[0061] ;

[0062] In the formula, represents the output layer input value, Represents the weight from the hidden layer to the output layer;

[0063] Step S414: Get the final output value, the formula used is as follows:

[0064] ;

[0065] In the formula, represents the output value of the output layer neuron, The bias term of the output layer;

[0066] Step S42: training the athlete potential prediction model, specifically using the reference athlete potential prediction data and the reference athlete status in the preliminary athlete career management data as input data to train the athlete potential prediction model to obtain a trained athlete potential prediction model;

[0067] Step S43: predicting the future potential of athletes, specifically using the target athlete potential prediction data in the preliminary athlete career management data and the target athlete assessment status results as input data for training the trained athlete potential prediction model to obtain the target athlete future potential prediction results.

[0068] Furthermore, in step S5, the hyperparameter optimization specifically includes the following steps:

[0069] Step S51: Initialize parameters, specifically by constructing initial algorithm parameters; the initial algorithm parameters include population size N, maximum number of iterations t max , search for the lower limit lb and the upper limit ub;

[0070] Step S52: Initialize the population, specifically, initialize the search individuals at random positions in the feasible search space, and the formula used is as follows:

[0071] ;

[0072] In the formula, represents the position of the i-th individual, Indicates between A random number uniformly distributed in the range;

[0073] Step S53: Population division: Calculate the individual fitness value f in the population i The performance of the mobilization potential prediction model based on individual position is used as the individual fitness value, and the individuals are sorted from best to worst according to their fitness values. The individual fa with the best fitness and its position X are defined. a , the individual fb with the second best fitness and its position X b , the third best individual fc and its position X c , classify fa, fb and fc as hunting individuals;

[0074] Step S54: Obtain the search adaptation factor, the formula used is as follows:

[0075] ;

[0076] In the formula, It indicates the acquisition of search adaptation factor, which is used to control the exploration and development capabilities of search individuals; t indicates the number of iterations; Represents a nonlinear parameter, which is used to control the change rate of the convergence factor;

[0077] Step S55: Update the search individual position; the formula used is as follows:

[0078] ;

[0079] ;

[0080] ;

[0081] ;

[0082] In the formula, represents the position of the candidate search individual after moving to fa, represents the position of the candidate search individual after moving to fb, represents the position of the candidate search individual after moving to fc, represents the position of the candidate search individual in the t-th generation population, represents the position of the candidate search individual in the t+1 generation population, , and represents the randomness parameter in the range [0,1];

[0083] Step S56: Adaptive neighborhood search, specifically, dynamically determining its own search direction through interaction with other individuals in the neighborhood; the formula used is as follows:

[0084] ;

[0085] In the formula, represents the position of the updated candidate search individual in the t-th generation population, , and represents a randomness parameter in the range [0,1], represents the position of other individuals in the neighborhood of the candidate search individual in the t-th generation population, represents the position of the global optimal individual, It represents the fitness function;

[0086] Step S57: greedily select and update the individual, the formula used is as follows:

[0087] ;

[0088] In the formula, Represents the position of the updated candidate search individual in the t+1 generation population;

[0089] Step S58: search determination, specifically, by constructing a search termination condition, searching and determining the optimal individual position, and obtaining the optimal individual position data setting;

[0090] The search termination conditions include threshold termination and iteration termination;

[0091] The threshold termination is specifically to set the fitness threshold, when the individual fitness value f i When it is above the fitness threshold, the hyperparameter search is completed;

[0092] The iteration termination specifically refers to terminating the iteration and obtaining the optimal individual position when the maximum number of iterations is reached;

[0093] The optimal individual position is specifically a parameter of an athlete potential prediction model.

[0094] Furthermore, in step S6, the comprehensive career management is specifically based on the target athlete's assessment status results and the target athlete's future potential prediction results, comprehensively assessing the athlete's career development in both short-term and long-term dimensions, and formulating a detailed career plan to help athletes ensure the maximum value of their careers at different stages of development. The career plan includes short-term training plans, competition arrangement plans, long-term peak development strategies and career transformation plans.

[0095] The technical solution adopted by the present invention is as follows: an athlete career management system based on artificial intelligence provided by the present invention includes a data acquisition module, a data preprocessing module, an athlete status evaluation module, an athlete potential prediction module, a hyperparameter optimization module and a comprehensive career management module;

[0096] The data acquisition module obtains the original data of athlete career management by collecting reference athlete status evaluation data, reference athlete potential prediction data, target athlete status evaluation data and target athlete potential prediction data from the sports event platform and the athlete personal information system, and sends the original data of athlete career management to the data preprocessing module;

[0097] The data preprocessing module receives the data sent by the data acquisition module, and performs data cleaning, standardization processing and feature selection on the original data of athlete career management to obtain preliminary data of athlete career management, and sends the preliminary data of athlete career management to the athlete status evaluation module and the athlete potential prediction module;

[0098] The athlete status evaluation module receives the preliminary athlete career management data sent by the data preprocessing module, establishes an athlete status evaluation model to evaluate the athlete's current status, obtains a target athlete evaluation status result, and sends the target athlete evaluation status result to the athlete status evaluation module and the comprehensive career management module;

[0099] The athlete potential prediction module receives the preliminary athlete career management data sent by the data preprocessing module and the target athlete evaluation status result sent by the athlete status evaluation module, establishes an athlete potential prediction model and performs model training on the model, inputs the target athlete evaluation status result and the target athlete potential prediction data into the trained prediction model to perform target athlete potential prediction, obtains the target athlete future potential prediction result, and sends the target athlete future potential prediction result to the comprehensive career management module;

[0100] The hyperparameter optimization module obtains the optimal solution combination of the athlete potential prediction model parameters by establishing a search algorithm, and sends the optimal solution combination of the athlete potential prediction model parameters to the athlete potential prediction module;

[0101] The comprehensive career management module receives the output results sent by the athlete status assessment module and the athlete potential prediction module. Through the two output results, it can comprehensively assess the athlete's career development in both short-term and long-term dimensions and formulate a career plan.

[0102] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0103] (1) In view of the risk problem in existing methods for athlete career management, which is to over-focus on the current status of athletes and ignore their future potential, thus leading to unbalanced development of athletes' careers and loss of overall benefits, this program creatively combines athlete status assessment with potential prediction to provide comprehensive and accurate support for athletes' career management, improve athletes' short-term competitive performance, provide scientific guidance for athletes' long-term development, help athletes smoothly achieve career transformation and development after their peak period, and ensure the continuity and stability of athletes' careers.

[0104] (2) In order to address the technical problem that the athlete status evaluation results are not accurate enough due to the limitations of the activation function and loss function in the existing models suitable for athlete status evaluation, this solution innovatively designs the status evaluation activation function and status evaluation loss function to improve the stability of the evaluation model, thereby solving the problem of inaccurate athlete status evaluation results.

[0105] (3) In view of the technical problem that the existing models suitable for predicting athlete potential have improper built-in parameter settings, which leads to inaccurate output results of the prediction model, this scheme adopts the search adaptation factor and neighborhood adaptive search strategy to improve the algorithm for obtaining the optimal parameters of the prediction model, obtain the optimal parameter combination, and improve the accuracy of the model output results, thereby making the prediction of athlete potential more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0106] Figure 1 A schematic diagram of a process flow of an athlete career management method based on artificial intelligence provided by the present invention;

[0107] Figure 2 A schematic diagram of a module of an athlete career management system based on artificial intelligence provided by the present invention;

[0108] Figure 3 This is a flow chart of athlete status assessment in step S3;

[0109] Figure 4 This is a schematic diagram of the process of predicting athlete potential in step S4;

[0110] Figure 5 A schematic diagram of the process of hyperparameter optimization in step S5;

[0111] Figure 6 A schematic diagram of the process of establishing an athlete status assessment model in step S31;

[0112] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0113] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0114] In the description of the present invention, it is necessary to understand that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred system or element must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0115] Example 1, see Figure 1 The present invention provides an athlete career management method based on artificial intelligence, the method comprising the following steps:

[0116] Step S1: data collection, specifically, obtaining athlete career management raw data by collecting reference athlete status evaluation data, reference athlete potential prediction data, target athlete status evaluation data, and target athlete potential prediction data from the sports event platform and the athlete personal information system;

[0117] Step S2: data preprocessing, specifically, data cleaning, standardization and feature selection of the original data of athlete career management to obtain preliminary data of athlete career management;

[0118] Step S3: Athlete status assessment, specifically, assessing the athlete's current status by establishing an athlete status assessment model to obtain a target athlete's assessment status result;

[0119] Step S4: predicting the potential of athletes, specifically, by establishing an athlete potential prediction model and training the model, inputting the target athlete assessment status result and the target athlete potential prediction data into the trained prediction model to predict the potential of the target athlete, and obtaining the prediction result of the future potential of the target athlete;

[0120] Step S5: Hyperparameter optimization, specifically, obtaining the optimal solution combination of athlete potential prediction model parameters by establishing a search algorithm;

[0121] Step S6: Comprehensive career management, specifically, comprehensively evaluating the athlete's career development in both short-term and long-term dimensions based on the target athlete's assessment status results and the target athlete's future potential prediction results, and formulating a career plan.

[0122] Example 2, see Figure 1 and Figure 2, this embodiment is based on the above embodiment. In step S1, the data collection is specifically to obtain the original data of athlete career management from the sports event platform and the athlete personal information system by collecting; the original data of athlete career management includes reference athlete status evaluation data, reference athlete potential prediction data, target athlete status evaluation data and target athlete potential prediction data; the reference athlete status evaluation data and the target athlete status evaluation data both contain physiological data, training data and psychological state data; the reference athlete status evaluation data also includes reference athlete status; the reference athlete potential prediction data and the target athlete potential prediction data both contain historical game data, historical training data and historical injury records; the physiological data includes heart rate, blood oxygen saturation, body temperature, fatigue and blood pressure; the training data includes training time, training intensity, training type and training effect feedback; the historical game data includes game results, game type and individual performance in the game; the historical training data includes long-term training data and training effects in different training cycles; the historical injury records include injury records, health files and recovery status data.

[0123] Example 3, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. In step S2, the data preprocessing is used to perform data cleaning, standardization processing and feature selection on the original data of athlete career management to obtain preliminary data of athlete career management; the data cleaning is to process missing values, abnormal values ​​and duplicate values; the standardization processing is to standardize the data based on the maximum and minimum normalization method; the feature selection is to use the correlation analysis method to screen out the relevant features for athlete status evaluation and athlete potential prediction from the original data of athlete career management.

[0124] Example 4, see Figure 1 , Figure 2 , Figure 3 and Figure 6 This embodiment is based on the above embodiment. In step S3, the athlete status evaluation is specifically to establish an athlete status evaluation model by designing a status evaluation activation function and a status evaluation loss function to obtain a target athlete evaluation status result, including the following steps:

[0125] Step S31: Establishing an athlete status assessment model, specifically including the following steps:

[0126] Step S311: Initialize model parameters, set the number of model hidden units, input vector dimension, output vector dimension and loss function threshold;

[0127] Step S312: Initialize the internal structure of the evaluation model ALM, which specifically includes the following steps:

[0128] Step S3121: Design a state evaluation activation function, the formula used is as follows:

[0129] ;

[0130] In the formula, x represents the input variable of the state evaluation activation function, represents the state evaluation activation function, represents the smoothing parameter of the activation function, Represents the scaling factor, which is used to adjust the range of the activation function output;

[0131] Step S3122: Forget gate calculation process, the formula used is as follows:

[0132] ;

[0133] In the formula, represents the output of the forget gate, It indicates input data. Indicates the hidden state at the last moment. and They represent the weight matrices corresponding to the forget gate processing input data and processing hidden states, respectively. represents the bias parameter of the forget gate, represents the Hadamard product;

[0134] Step S3123: Input gate calculation process, the formula used is as follows:

[0135] ;

[0136] ;

[0137] ;

[0138] In the formula, represents the output of the input gate, and Respectively represent the weight matrices corresponding to the input gate processing input data and processing hidden states, represents the bias parameter of the input gate, Indicates the current cell state. represents the candidate value of the new cell state, It indicates the cell state at the last moment. and Respectively represent the weight matrices corresponding to the cell state processing input data and processing hidden state, The bias parameter representing the cell state, represents element-wise multiplication;

[0139] Step S3124: Output gate calculation process, the formula used is as follows:

[0140] ;

[0141] In the formula, represents the output of the output gate, and They represent the weight matrices corresponding to the output gate processing input data and processing hidden states respectively; Represents the bias parameter of the output gate;

[0142] Step S3125: hidden state calculation process, the formula used is as follows:

[0143] ;

[0144] In the formula, Indicates the hidden state at the current moment;

[0145] Step S313: Calculate the bidirectional ALM using the following formula:

[0146] ;

[0147] ;

[0148] ;

[0149] In the formula, represents the hidden state of the combined forward and backward ALM, Indicates the hidden state in the positive direction at the current moment, Indicates the hidden state in the positive direction at the previous moment, Indicates the reverse hidden state at the current moment. Indicates the hidden state in the reverse direction at the next moment;

[0150] Step S314: Get the final output value, the formula used is as follows:

[0151] ;

[0152] In the formula, represents the output value of the model, represents the weight matrix corresponding to the output, Represents the bias parameter of the output layer;

[0153] Step S315: Design a state evaluation loss function, the formula used is as follows:

[0154] ;

[0155] In the formula, represents the state evaluation loss function value, Represents the steepness of the control loss function, Represents the parameter for accelerating convergence, which is used to adjust the extreme points of the derivative. Indicates the actual evaluation status of the athlete at the current time;

[0156] Step S316: Update parameters, specifically use the gradient descent method to update all weight matrices and bias items in the athlete status assessment model; the formula used is as follows:

[0157] ;

[0158] ;

[0159] In the formula, Denotes the loss function for the weight matrix The gradient of Denotes the loss function to represent the bias parameter The gradient of represents the updated weight matrix, represents the current weight matrix, represents the updated bias parameter, Represents the current bias parameter, represents the learning rate;

[0160] Step S317: Iteration determination, specifically monitoring the change of the loss function and the current number of iterations. If the loss function is lower than the set threshold, the iterative training is stopped, otherwise the iterative training is continued. If the maximum number of iterations is reached, the iterative training is stopped, otherwise the iterative training is continued;

[0161] Step S32: training the athlete status assessment model, specifically using the reference athlete status assessment data in the athlete career management preliminary data as input data to train the athlete status assessment model to obtain a trained athlete status assessment model;

[0162] Step S33: Athlete status assessment, specifically, using the target athlete status assessment data in the preliminary athlete career management data as input data of the trained athlete status assessment model to obtain the target athlete status assessment result.

[0163] By performing the above operations, this solution innovatively designs the state evaluation activation function and the state evaluation loss function to improve the stability of the evaluation model, thereby solving the problem of inaccurate athlete state evaluation results, which exists in the limitations of the activation function and the loss function in the existing model suitable for athlete state evaluation.

[0164] Example 5, see Figure 1 , Figure 2 and Figure 4 This embodiment is based on the above embodiment. In step S4, the athlete potential prediction is specifically performed by establishing an athlete potential prediction model and training the model, and the target athlete potential is predicted for the trained prediction model input data to obtain the target athlete's future potential prediction result; the following steps are included:

[0165] Step S41: Establishing an athlete potential prediction model, specifically including the following steps:

[0166] Step S411: Calculate the hidden layer input using the following formula:

[0167] ;

[0168] In the formula, represents the hidden layer input value, represents the weight from the input layer to the hidden layer, Represents the input value of the i-th input neuron;

[0169] Step S412: Calculate the output of the hidden layer using the following formula:

[0170] ;

[0171] In the formula, represents the output value of the hidden layer neuron, Bias term of hidden layer;

[0172] Step S413: Calculate the output layer input using the following formula:

[0173] ;

[0174] In the formula, represents the output layer input value, Represents the weight from the hidden layer to the output layer;

[0175] Step S414: Get the final output value, the formula used is as follows:

[0176] ;

[0177] In the formula, represents the output value of the output layer neuron, The bias term of the output layer;

[0178] Step S42: training the athlete potential prediction model, specifically using the reference athlete potential prediction data and the reference athlete status in the preliminary athlete career management data as input data to train the athlete potential prediction model to obtain a trained athlete potential prediction model;

[0179] Step S43: predicting the future potential of athletes, specifically using the target athlete potential prediction data in the preliminary athlete career management data and the target athlete assessment status results as input data for training the trained athlete potential prediction model to obtain the target athlete future potential prediction results.

[0180] Example 6, see Figure 1 , Figure 2 and Figure 5 This embodiment is based on the above embodiment. In step S5, the hyperparameter optimization specifically includes the following steps:

[0181] Step S51: Initialize parameters, specifically by constructing initial algorithm parameters; the initial algorithm parameters include population size N, maximum number of iterations t max , search for the lower limit lb and upper limit ub;

[0182] Step S52: Initialize the population, specifically, initialize the search individuals at random positions in the feasible search space, and the formula used is as follows:

[0183] ;

[0184] In the formula, represents the position of the i-th individual, Indicates between A random number uniformly distributed in the range;

[0185] Step S53: Population division: Calculate the individual fitness value f in the population i The performance of the mobilization potential prediction model based on individual position is used as the individual fitness value, and the individuals are sorted from best to worst according to their fitness values. The individual fa with the best fitness and its position X are defined. a , the individual fb with the second best fitness and its position X b , the third best individual fc and its position X c , classify fa, fb and fc as hunting individuals;

[0186] Step S54: Obtain the search adaptation factor, the formula used is as follows:

[0187] ;

[0188] In the formula, It indicates the acquisition of search adaptation factor, which is used to control the exploration and development capabilities of search individuals; t indicates the number of iterations; Represents a nonlinear parameter, which is used to control the change rate of the convergence factor;

[0189] Step S55: Update the search individual position; the formula used is as follows:

[0190] ;

[0191] ;

[0192] ;

[0193] ;

[0194] In the formula, represents the position of the candidate search individual after moving to fa, represents the position of the candidate search individual after moving to fb, represents the position of the candidate search individual after moving to fc, represents the position of the candidate search individual in the t-th generation population, represents the position of the candidate search individual in the t+1 generation population, , and represents the randomness parameter in the range [0,1];

[0195] Step S56: Adaptive neighborhood search, specifically, dynamically determining its own search direction through interaction with other individuals in the neighborhood; the formula used is as follows:

[0196] ;

[0197] In the formula, represents the position of the updated candidate search individual in the t-th generation population, , and represents a randomness parameter in the range [0,1], represents the position of other individuals in the neighborhood of the candidate search individual in the t-th generation population, represents the position of the global optimal individual, It represents the fitness function;

[0198] Step S57: greedily select and update the individual, the formula used is as follows:

[0199] ;

[0200] In the formula, Represents the position of the updated candidate search individual in the t+1 generation population;

[0201] Step S58: search determination, specifically, by constructing a search termination condition, searching and determining the optimal individual position, and obtaining the optimal individual position data setting;

[0202] The search termination conditions include threshold termination and iteration termination;

[0203] The threshold termination is specifically to set the fitness threshold, when the individual fitness value f i When it is above the fitness threshold, the hyperparameter search is completed;

[0204] The iteration termination specifically refers to terminating the iteration and obtaining the optimal individual position when the maximum number of iterations is reached;

[0205] The optimal individual position is specifically a parameter of an athlete potential prediction model.

[0206] By performing the above operations, in view of the technical problem that the existing models suitable for athlete potential prediction models have improper built-in parameter settings, which leads to inaccurate output results of the prediction model, this solution adopts the search adaptation factor and neighborhood adaptive search strategy to improve the algorithm for obtaining the optimal parameters of the prediction model, obtain the optimal parameter combination, and improve the accuracy of the model output results, thereby making the prediction of athlete potential more accurate.

[0207] Embodiment 7, see Figure 1 and Figure 2 This embodiment is based on the above embodiment. In step S6, the comprehensive career management is specifically based on the target athlete assessment status results and the target athlete future potential prediction results, comprehensively assessing the athlete's career development in both short-term and long-term dimensions, and formulating a detailed career plan to help athletes ensure the maximum value of their careers at different development stages. The career plan includes short-term training plans, competition arrangement plans, long-term peak development strategies and career transformation plans.

[0208] By performing the above operations, this solution creatively combines athlete status assessment with potential prediction to provide comprehensive and accurate support for athlete career management, improve athletes' short-term competitive performance, provide scientific guidance for athletes' long-term development, help athletes smoothly achieve career transformation and development after their peak period, and ensure the continuity and stability of athletes' careers.

[0209] Embodiment 8, see Figure 1 and Figure 2 , this embodiment is based on the above embodiment, and the present invention provides an athlete career management system based on artificial intelligence, including a data acquisition module, a data preprocessing module, an athlete status evaluation module, an athlete potential prediction module, a hyperparameter optimization module and a comprehensive career management module;

[0210] The data acquisition module obtains the original data of athlete career management by collecting reference athlete status evaluation data, reference athlete potential prediction data, target athlete status evaluation data and target athlete potential prediction data from the sports event platform and the athlete personal information system, and sends the original data of athlete career management to the data preprocessing module;

[0211] The data preprocessing module receives the data sent by the data acquisition module, and performs data cleaning, standardization processing and feature selection on the original data of athlete career management to obtain preliminary data of athlete career management, and sends the preliminary data of athlete career management to the athlete status evaluation module and the athlete potential prediction module;

[0212] The athlete status evaluation module receives the preliminary athlete career management data sent by the data preprocessing module, establishes an athlete status evaluation model to evaluate the athlete's current status, obtains a target athlete evaluation status result, and sends the target athlete evaluation status result to the athlete status evaluation module and the comprehensive career management module;

[0213] The athlete potential prediction module receives the preliminary athlete career management data sent by the data preprocessing module and the target athlete evaluation status result sent by the athlete status evaluation module, establishes an athlete potential prediction model and performs model training on the model, performs target athlete potential prediction on the trained prediction model input data, obtains the target athlete future potential prediction result, and sends the target athlete future potential prediction result to the comprehensive career management module;

[0214] The hyperparameter optimization module obtains the optimal solution combination of the athlete potential prediction model parameters by establishing a search algorithm, and sends the optimal solution combination of the athlete potential prediction model parameters to the athlete potential prediction module;

[0215] The comprehensive career management module receives the output results sent by the athlete status assessment module and the athlete potential prediction module. Through the two output results, it can comprehensively assess the athlete's career development in both short-term and long-term dimensions and formulate a career plan.

[0216] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0217] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0218] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. An athlete career management method based on artificial intelligence, characterized by: The method comprises the following steps: Step S1: data collection, specifically, obtaining athlete career management raw data by collecting reference athlete status evaluation data, reference athlete potential prediction data, target athlete status evaluation data, and target athlete potential prediction data from the sports event platform and the athlete personal information system; Step S2: data preprocessing, specifically, data cleaning, standardization and feature selection of the original data of athlete career management to obtain preliminary data of athlete career management; Step S3: Athlete status evaluation, specifically, establishing an athlete status evaluation model by designing a status evaluation activation function and a status evaluation loss function to evaluate the athlete's current status and obtain the target athlete's evaluation status result; Step S4: predicting the potential of athletes, specifically, by establishing an athlete potential prediction model and training the model, inputting the target athlete assessment status result and the target athlete potential prediction data into the trained prediction model to predict the potential of the target athlete, and obtaining the prediction result of the future potential of the target athlete; Step S5: Hyperparameter optimization, specifically using a search adaptation factor and a neighborhood adaptive search strategy to improve the algorithm for obtaining the optimal parameters of the prediction model, and obtaining the optimal solution combination of the athlete potential prediction model parameters; Step S6: Comprehensive career management, specifically, comprehensively evaluating the athlete's career development in both short-term and long-term dimensions based on the target athlete's assessment status results and the target athlete's future potential prediction results, and formulating a career plan.

2. The method for athlete career management based on artificial intelligence according to claim 1, characterized in that: In step S3, the athlete status assessment is specifically to establish an athlete status assessment model by designing a status assessment activation function and a status assessment loss function to obtain a target athlete assessment status result, including the following steps: Step S31: Establishing an athlete status assessment model, specifically including the following steps: Step S311: Initialize model parameters, set the number of model hidden units, input vector dimension, output vector dimension and loss function threshold; Step S312: Initialize the internal structure of the evaluation model ALM, which specifically includes the following steps: Step S3121: Design a state evaluation activation function, the formula used is as follows: ； In the formula, x represents the input variable of the state evaluation activation function, represents the state evaluation activation function, represents the smoothing parameter of the activation function, Represents the scaling factor, which is used to adjust the range of the activation function output; Step S3122: Forget gate calculation process, the formula used is as follows: ； In the formula, represents the output of the forget gate, It indicates input data. Indicates the hidden state at the last moment. and They represent the weight matrices corresponding to the forget gate processing input data and processing hidden states, respectively. represents the bias parameter of the forget gate, represents the Hadamard product; Step S3123: Input gate calculation process, the formula used is as follows: ； ； ； In the formula, represents the output of the input gate, and Respectively represent the weight matrices corresponding to the input gate processing input data and processing hidden states, represents the bias parameter of the input gate, Indicates the current cell state. represents the candidate value of the new cell state, It indicates the cell state at the last moment. and Respectively represent the weight matrices corresponding to the cell state processing input data and processing hidden state, The bias parameter representing the cell state, represents element-wise multiplication; Step S3124: Output gate calculation process, the formula used is as follows: ； In the formula, represents the output of the output gate, and They represent the weight matrices corresponding to the output gate processing input data and processing hidden states respectively; Represents the bias parameter of the output gate; Step S3125: Hidden state calculation process, the formula used is as follows: ； In the formula, Indicates the hidden state at the current moment; Step S313: Calculate the bidirectional ALM using the following formula: ； ； ； In the formula, represents the hidden state of the combined forward and backward ALM, Indicates the hidden state in the positive direction at the current moment, Indicates the hidden state in the positive direction at the previous moment, Indicates the reverse hidden state at the current moment. Indicates the hidden state in the reverse direction at the next moment; Step S314: Get the final output value, the formula used is as follows: ； In the formula, represents the output value of the model, represents the weight matrix corresponding to the output, Represents the bias parameter of the output layer; Step S315: Design a state evaluation loss function, the formula used is as follows: ； In the formula, represents the state evaluation loss function value, Represents the steepness of the control loss function, Represents the parameter for accelerating convergence, which is used to adjust the extreme points of the derivative. Indicates the actual evaluation status of the athlete at the current time; Step S316: Update parameters, specifically use the gradient descent method to update all weight matrices and bias items in the athlete status assessment model; the formula used is as follows: ； ； In the formula, Denotes the loss function for the weight matrix The gradient of Denotes the loss function to represent the bias parameter The gradient of represents the updated weight matrix, represents the current weight matrix, represents the updated bias parameter, Represents the current bias parameter, represents the learning rate; Step S317: Iteration determination, specifically monitoring the change of the loss function. If the loss function is lower than the set loss function threshold, the iterative training is stopped. Otherwise, the iterative training is continued until the maximum number of iterations is reached. Step S32: training the athlete status assessment model, specifically using the reference athlete status assessment data in the athlete career management preliminary data as input data to train the athlete status assessment model to obtain a trained athlete status assessment model; Step S33: Athlete status assessment, specifically, using the target athlete status assessment data in the preliminary athlete career management data as input data of the trained athlete status assessment model to obtain the target athlete status assessment result.

3. The method for athlete career management based on artificial intelligence according to claim 1, characterized in that: In step S4, the athlete potential prediction is specifically performed by establishing an athlete potential prediction model and training the model, and predicting the target athlete's potential based on the trained prediction model input data to obtain the target athlete's future potential prediction result; the steps include: Step S41: Establishing an athlete potential prediction model, specifically including the following steps: Step S411: Calculate the hidden layer input using the following formula: ； In the formula, represents the hidden layer input value, represents the weight from the input layer to the hidden layer, Represents the input value of the i-th input neuron; Step S412: Calculate the output of the hidden layer using the following formula: ； In the formula, represents the output value of the hidden layer neuron, Bias term of hidden layer; Step S413: Calculate the output layer input using the following formula: ； In the formula, represents the output layer input value, Represents the weight from the hidden layer to the output layer; Step S414: Get the final output value, the formula used is as follows: ； In the formula, represents the output value of the output layer neuron, The bias term of the output layer; Step S42: training the athlete potential prediction model, specifically using the reference athlete potential prediction data and the reference athlete status in the preliminary athlete career management data as input data to train the athlete potential prediction model to obtain a trained athlete potential prediction model; Step S43: predicting the future potential of athletes, specifically using the target athlete potential prediction data in the preliminary athlete career management data and the target athlete assessment status results as input data for training the trained athlete potential prediction model to obtain the target athlete future potential prediction results.

4. The method for athlete career management based on artificial intelligence according to claim 1, characterized in that: The hyperparameter optimization specifically includes the following steps: Step S51: Initialize parameters, specifically by constructing initial algorithm parameters; the initial algorithm parameters include population size N, maximum number of iterations t max , search for the lower limit lb and the upper limit ub; Step S52: Initialize the population, specifically, initialize the search individuals at random positions in the feasible search space, and the formula used is as follows: ； In the formula, represents the position of the i-th individual, Indicates between A random number uniformly distributed in the range; Step S53: Population division: Calculate the individual fitness value f in the population i The performance of the mobilization potential prediction model based on individual position is used as the individual fitness value, and the individuals are sorted from best to worst according to their fitness values. The individual fa with the best fitness and its position X are defined. a , the individual fb with the second best fitness and its position X b , the third best individual fc and its position X c , classify fa, fb and fc as hunting individuals; Step S54: Obtain the search adaptation factor, the formula used is as follows: ； In the formula, It indicates the acquisition of search adaptation factor, which is used to control the exploration and development capabilities of search individuals; t indicates the number of iterations; Represents a nonlinear parameter, which is used to control the change rate of the convergence factor; Step S55: Update the search individual position; the formula used is as follows: ； ； ； ； In the formula, represents the position of the candidate search individual after moving to fa, represents the position of the candidate search individual after moving to fb, represents the position of the candidate search individual after moving to fc, represents the position of the candidate search individual in the t-th generation population, represents the position of the candidate search individual in the t+1 generation population, , and represents the randomness parameter in the range [0,1]; Step S56: Adaptive neighborhood search, specifically, dynamically determining its own search direction through interaction with other individuals in the neighborhood; the formula used is as follows: ； In the formula, represents the position of the updated candidate search individual in the t-th generation population, , and represents a randomness parameter in the range [0,1], represents the position of other individuals in the neighborhood of the candidate search individual in the t-th generation population, represents the position of the global optimal individual, It represents the fitness function; Step S57: greedily select and update the individual, the formula used is as follows: ； In the formula, Represents the position of the updated candidate search individual in the t+1 generation population; Step S58: search determination, specifically, by constructing a search termination condition, searching and determining the optimal individual position, and obtaining the optimal individual position data setting; The search termination conditions include threshold termination and iteration termination; The threshold termination is specifically to set the fitness threshold, when the individual fitness value f i When it is above the fitness threshold, the hyperparameter search is completed; The iteration termination specifically refers to terminating the iteration and obtaining the optimal individual position when the maximum number of iterations is reached; The optimal individual position is specifically a parameter of an athlete potential prediction model.

5. The method for athlete career management based on artificial intelligence according to claim 1, characterized in that: In step S6, the comprehensive career management is specifically based on the target athlete's assessment status results and the target athlete's future potential prediction results, comprehensively assessing the athlete's career development in both short-term and long-term dimensions, and formulating a detailed career plan to help athletes ensure the maximum value of their careers at different stages of development. The career plan includes short-term training plans, competition arrangement plans, long-term peak development strategies and career transformation plans.

6. The method for athlete career management based on artificial intelligence according to claim 1, characterized in that: In step S1, the data collection is specifically to obtain the original data of athlete career management from the sports event platform and the athlete personal information system; the original data of athlete career management includes reference athlete status evaluation data, reference athlete potential prediction data, target athlete status evaluation data and target athlete potential prediction data; The reference athlete status assessment data and the target athlete status assessment data both include physiological data, training data and psychological status data; the reference athlete status assessment data also includes the reference athlete status; the reference athlete potential prediction data and the target athlete potential prediction data both include historical game data, historical training data and historical injury records; the physiological data include heart rate, blood oxygen saturation, body temperature, fatigue and blood pressure; the training data include training duration, training intensity, training type and training effect feedback; the historical game data include game results, game type and individual performance in the game; the historical training data include long-term training data and training effects in different training cycles; the historical injury records include injury records, health files and recovery status data.

7. The method for athlete career management based on artificial intelligence according to claim 1, characterized in that: In step S2, the data preprocessing is used to perform data cleaning, standardization and feature selection on the original data of athlete career management to obtain preliminary data of athlete career management; the data cleaning is to process missing values, abnormal values ​​and duplicate values; the standardization is to standardize the data based on the maximum and minimum normalization method; The feature selection is to use the correlation analysis method to screen out the relevant features for athlete status evaluation and athlete potential prediction from the original data of athlete career management.

8. An athlete career management system based on artificial intelligence, used to implement an athlete career management method based on artificial intelligence as described in any one of claims 1-7, characterized in that: It includes data collection module, data preprocessing module, athlete status assessment module, athlete potential prediction module, hyperparameter optimization module and comprehensive career management module.

9. An artificial intelligence-based athlete career management system according to claim 8, characterized in that: The data acquisition module obtains the original data of athlete career management by collecting reference athlete status evaluation data, reference athlete potential prediction data, target athlete status evaluation data and target athlete potential prediction data from the sports event platform and the athlete personal information system, and sends the original data of athlete career management to the data preprocessing module; The data preprocessing module receives the data sent by the data acquisition module, and performs data cleaning, standardization processing and feature selection on the original data of athlete career management to obtain preliminary data of athlete career management, and sends the preliminary data of athlete career management to the athlete status evaluation module and the athlete potential prediction module; The athlete status evaluation module receives the preliminary athlete career management data sent by the data preprocessing module, establishes an athlete status evaluation model to evaluate the athlete's current status, obtains a target athlete evaluation status result, and sends the target athlete evaluation status result to the athlete status evaluation module and the comprehensive career management module; The athlete potential prediction module receives the preliminary athlete career management data sent by the data preprocessing module and the target athlete evaluation status result sent by the athlete status evaluation module, establishes an athlete potential prediction model and performs model training on the model, inputs the target athlete evaluation status result and the target athlete potential prediction data into the trained prediction model to perform target athlete potential prediction, obtains the target athlete future potential prediction result, and sends the target athlete future potential prediction result to the comprehensive career management module; The hyperparameter optimization module obtains the optimal solution combination of the athlete potential prediction model parameters by establishing a search algorithm, and sends the optimal solution combination of the athlete potential prediction model parameters to the athlete potential prediction module; The comprehensive career management module receives the output results sent by the athlete status assessment module and the athlete potential prediction module. Through the two output results, it can comprehensively assess the athlete's career development in both short-term and long-term dimensions and formulate a career plan.

Images