Athlete selection method and system based on physical test spatiotemporal big data

Abstract

The present invention discloses a method for selecting athletes based on physical test spatiotemporal big data, which includes the following steps: S1: collecting and storing the physical test spatiotemporal big data of athletes to obtain a physical test spatiotemporal big data set; S2: establishing an athlete selection model based on the physical test spatiotemporal big data set; S3: outputting the athlete selection analysis results based on the athlete selection model. Correspondingly, the present invention also discloses an athlete selection system, which is applied to the above-mentioned athlete selection method, and includes: a data acquisition module, a data storage module and a data processing module, wherein the data acquisition module is communicatively connected to the data storage module, and the data storage module is communicatively connected to the data processing module. The athlete selection method and system of the present invention can analyze and model the physical test spatiotemporal big data of athletes using machine learning algorithms and data mining techniques to establish an athlete selection model, thereby selecting athletes efficiently and accurately.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to an athlete selection method and system based on physical test spatiotemporal big data. Background Art

[0002] In recent years, with the continuous improvement of sports competition and the intensification of competition, it has become particularly important to test and select excellent athletes. Among them, the existing traditional selection methods often rely on the subjective judgment of professionals, which are easily affected by personal bias and experience; in addition, the traditional selection methods use limited data when selecting athletes and cannot fully understand the athletes' physical indicators and athletic ability.

[0003] To this end, researchers have tried and designed a selection method for athletes based on big data. However, existing selection methods based on big data usually only provide static, single-dimensional information, which cannot fully reflect the temporal and spatial characteristics of events and phenomena, and ignore the temporal and spatial correlation and dynamic evolution, resulting in inaccurate evaluation results and low evaluation efficiency. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an athlete selection method and system based on physical test spatiotemporal big data, so as to solve the problems of inaccurate evaluation results and low evaluation efficiency caused by the existing athlete selection method due to ignoring spatiotemporal correlation and dynamic evolution.

[0005] In order to solve the above technical problems, the technical solution of the present invention is: a method for selecting athletes based on physical measurement spatiotemporal big data, which includes the steps of: S1: collecting and storing the physical measurement spatiotemporal big data of athletes to obtain a physical measurement spatiotemporal big data set; S2: establishing an athlete selection model based on the physical measurement spatiotemporal big data set; S3: outputting the athlete selection analysis results based on the athlete selection model.

[0006] In view of the defects of the existing traditional technical solutions for evaluating and selecting athletes, the athlete selection method designed by the present invention first measures and collects the physical test data of athletes in terms of physical indicators, physiological data and athletic ability for a long time, and forms a huge physical test spatiotemporal data set based on the time information and spatial information during the physical test. These data not only cover the athlete's height, weight, muscle mass, explosive power, speed, flexibility and other indicators, but also cover the time point and location information of the athlete when the test is carried out. It can cover the athlete's training cycle, season cycle or personal development process, so it has the characteristics of being objective, comprehensive and quantifiable.

[0007] Accordingly, after obtaining the above-mentioned physical test spatiotemporal big data set, machine learning algorithms and data mining techniques can also be used to analyze and model the athletes' physical test spatiotemporal big data to establish an athlete selection model, thereby mining and discovering the relationship between physical test data and athlete performance, and providing reliable program support for subsequent selection agencies for athlete selection.

[0008] Furthermore, in the athlete selection method described in the present invention, in step S1, specifically: collecting the athlete's physical measurement data, and forming the physical measurement spatiotemporal big data based on the time information and location information of the athlete during the physical measurement; storing the physical measurement spatiotemporal big data, and obtaining a physical measurement spatiotemporal big data set based on a number of the physical measurement spatiotemporal big data.

[0009] Furthermore, in the athlete selection method described in the present invention, in step S1, the physical test data includes: body shape data, physical fitness data, sports performance data and coach scoring data obtained by the athlete through physical test.

[0010] In the above technical scheme of the present invention, physical fitness test spatiotemporal big data refers to large-scale data related to time and space collected, stored and analyzed in the physical fitness test of athletes. It covers the physical fitness test data of athletes at different time points and places, as well as the time series and spatial distribution characteristics of these data. Body morphology data refers to the external appearance and body shape characteristics of human individuals, including characteristics in terms of height, weight, body proportion, muscle development, bone structure, etc. The body shape of the human body is determined by multiple factors such as genetic factors, physiological development, lifestyle and environment. Physical fitness data: data that characterizes a person's overall physical quality and ability, which involves multiple aspects of physical characteristics and abilities. It is a comprehensive indicator for evaluating a person's physical health and athletic ability. Sports performance data: represents the results of actual sports physical tests. Coach scoring data: represents the coach's personal subjective evaluation of the athlete.

[0011] Furthermore, in the athlete selection method described in the present invention, in step S2, specifically: statistics, learning and data mining are performed on the physical measurement spatiotemporal big data in the physical measurement spatiotemporal big data set to extract key features and patterns from the physical measurement spatiotemporal big data and establish an athlete selection model.

[0012] Furthermore, in the athlete selection method described in the present invention, in step S2, key features and patterns are extracted and an athlete selection model is constructed through statistical analysis, clustering algorithm, regression analysis, association mining, anomaly detection and time series analysis methods.

[0013] Accordingly, another object of the present invention is to disclose and design an athlete selection system based on physical test spatiotemporal big data, which is applied to the above-mentioned athlete selection method of the present invention, and comprises: a data acquisition module, a data storage module and a data processing module, wherein the data acquisition module is communicatively connected to the data storage module, and the data storage module is communicatively connected to the data processing module; Among them, the data acquisition module is used to collect the physical measurement spatiotemporal big data of athletes, the data storage module is used to store the physical measurement spatiotemporal big data, and obtain the physical measurement spatiotemporal big data set; the data processing module is used to establish an athlete selection model according to the physical measurement spatiotemporal big data set, and based on the athlete selection model, output the selection analysis results of candidate athletes.

[0014] Furthermore, the athlete selection system described in the present invention also includes a visualization module, which is communicatively connected to the data processing module. The visualization module receives the selection analysis results and displays the selection analysis results in a visual manner.

[0015] In the above technical solution of the present invention, by setting up a visualization module, the analysis results can be displayed in a visual manner, and its visualization interface is used to present the athlete's physical test data, evaluation indicators, development trajectory and other information in time and space dimensions, thereby helping users to more intuitively understand and compare the potential and adaptability of candidate athletes.

[0016] Furthermore, in the athlete selection system described in the present invention, the storage data in the data storage module includes athlete ID information and time information, geographic information and physical measurement data of the athlete during the physical measurement; wherein the physical measurement data includes: body shape data, physical fitness data, athletic performance data and coach scoring data obtained by the athlete through the physical measurement.

[0017] Furthermore, the athlete selection system described in the present invention also includes a preprocessing module, which is connected to the data acquisition module and the data storage module respectively, and the preprocessing module preprocesses the collected initial data, and the preprocessing includes: data cleaning, outlier detection and data normalization.

[0018] Furthermore, in the athlete selection system described in the present invention, the data processing module includes: a modeling unit and an evaluation unit; wherein the modeling unit analyzes and models the physical measurement spatiotemporal big data in the data storage module, specifically through statistical analysis, clustering algorithm, regression analysis, association mining, anomaly detection and time series analysis methods, to extract key features and patterns, and construct an athlete selection model; the evaluation unit evaluates and obtains the athlete selection analysis results based on the established athlete selection model.

[0019] The beneficial effect of the present invention is that the present invention aims to solve the problem of efficiency and accuracy in athlete selection. Specifically, the inventor has optimized and designed an athlete selection method based on physical test spatiotemporal big data. The method automatically completes the collection and uploading of physical test data of multiple physical fitness and athletic ability indicators of athletes by networking physical test equipment, and then stores these physical test data according to time and space dimensions, and then selects athletes efficiently and accurately through data analysis in time and space dimensions.

[0020] The athlete selection method of the invention reduces the errors and biases of subjective judgment by collecting a large amount of physical test data related to time and space and scientific model analysis. It can comprehensively evaluate the advantages and improvement directions of athletes and provide guidance for the personalized training of athletes; and it is also scalable and real-time, and can continuously learn and optimize the model as the data accumulates and updates, so as to always maintain accuracy and adaptability. Therefore, the athlete selection method of the present invention provides a more objective and accurate evaluation basis for the selection organization, reduces the interference of subjective factors, and improves the efficiency and scientificity of the selection, so as to better provide the selection organization with ranking, scoring or recommendation suggestions to assist its decision-making process.

[0021] Correspondingly, in the present invention, an athlete selection system based on physical test spatiotemporal big data is also designed. The athlete selection system is used to implement the above-mentioned athlete selection method of the present invention, which also has the above-mentioned advantages and beneficial effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The figure schematically shows the system architecture diagram of the athlete selection system based on physical test spatiotemporal big data in one embodiment of the present invention.

[0023] Figure 2 The figure is a schematic diagram of the steps of the athlete selection method based on physical test spatiotemporal big data in one implementation mode of the present invention.

[0024] Figure 3 A model diagram of an athlete selection model established by the athlete selection system based on physical measurement spatiotemporal big data in one implementation manner of the present invention. DETAILED DESCRIPTION

[0025] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in conjunction with the implementation modes and the accompanying drawings.

[0026] The study found that in the current existing technology, the selection methods designed by researchers for selecting athletes based on big data usually only provide static, single-dimensional information, which cannot fully reflect the temporal and spatial characteristics of events and phenomena, and ignores the temporal and spatial correlation and dynamic evolution, which will cause the evaluation results to be inaccurate and unreliable.

[0027] Therefore, in order to solve the problems existing in the above-mentioned prior art, different from the prior art, the present invention designs an athlete selection method based on physical test spatiotemporal big data, which makes full use of time and space information to evaluate and select athletes in a more comprehensive and scientific way.

[0028] The present invention designs a method for selecting athletes based on physical test spatiotemporal big data, which comprises the following steps: S1: Collecting the physical test data of athletes, and forming the physical test spatiotemporal big data according to the time information and location information of the athletes during the physical test; storing the physical test spatiotemporal big data, and obtaining a physical test spatiotemporal big data set according to a plurality of the physical test spatiotemporal big data.

[0029] S2: Conduct statistics, learning and data mining on the physical test spatiotemporal big data in the physical test spatiotemporal big data set to extract key features and patterns from the physical test spatiotemporal big data and establish an athlete selection model.

[0030] S3: Obtain and compare the physical test data of each candidate athlete with the standards and requirements in the athlete selection model, and provide a score, matching degree or potential index for each candidate athlete to output the selection analysis results of the candidate athletes.

[0031] Accordingly, if Figure 1 As shown, the present invention also discloses an athlete selection system based on physical measurement spatiotemporal big data, for implementing the above-mentioned athlete selection method of the present invention. The athlete selection system specifically includes: a preprocessing module, a data acquisition module, a data storage module, a data processing module and a visualization module; and, in actual application, the above-mentioned data acquisition module is communicatively connected to the data storage module, and the data storage module and the visualization module are both communicatively connected to the data processing module.

[0032] Among them, the data acquisition module is used to collect the physical measurement spatiotemporal big data of athletes, the data storage module is used to store the above physical measurement spatiotemporal big data, and obtain the physical measurement spatiotemporal big data set based on several individual physical measurement spatiotemporal big data; the above data processing module is used to establish an athlete selection model based on the above physical measurement spatiotemporal big data set, and output the selection analysis results of candidate athletes based on the athlete selection model; the above visualization module receives the selection analysis results and displays the selection analysis results in a visual manner.

[0033] The core content of the present invention is to use physical measurement spatiotemporal big data, through data analysis and modeling, to provide scientific basis and decision support for the selection and selection of athletes, so as to more objectively evaluate the potential and adaptability of athletes, and help decision makers make more accurate and scientific decisions when selecting athletes. In practical application, this athlete selection method and system designed by the present invention can be applied to athlete selection and potential discovery process, to select or discover athletes who are good at specific sports. Of course, the present invention is based on physical measurement spatiotemporal big data to assist athletes in the selection and selection process, so it can be applicable to various sports scenes, including but not limited to table tennis, football, basketball, badminton, swimming, weightlifting, track and field, etc.

[0034] See also Figure 1 It can be seen that, from bottom to top, the processes to be executed by the athlete selection system include: data collection, data storage, data analysis and modeling, evaluation and decision-making, data visualization and decision support.

[0035] In the present invention, when the above-mentioned athlete selection system is used to execute the above-mentioned "data collection" process, it can specifically use the data collection device to collect the athlete's physical measurement data; when collecting data, various physical measurement sensors, physical measurement equipment, etc. can be connected to the network; during the physical measurement, the athlete triggers the collection device, and by connecting the physical measurement wearable device or physical fitness tester to the network, the athlete's physical measurement data can be automatically collected and reported, and the collection and storage of the athlete's physical measurement data can be automatically completed.

[0036] In practical application, the data acquisition device provided by the present invention is responsible for connecting and communicating with the physical measurement equipment, and its communication form may include: wired connection communication, wireless Wi-Fi communication, RFID communication, and wireless Bluetooth communication. The physical measurement equipment obtains the physical measurement data of the athlete in real time, and reports the data to the data acquisition device, which is then stored by the data storage module.

[0037] Accordingly, when performing the "data storage" process, the data storage module will store: the athlete's ID information and the time information, geographic information and physical test data of the athlete. The physical test data includes: the athlete's body shape data, physical fitness data, sports performance data and coach scoring data obtained through the physical test.

[0038] In the present invention, the "body shape data" stored in the data storage module may include multiple feature data as shown in Table 1: Table 1

[0039] In the present invention, the "physical fitness data" stored in the data storage module may include multiple feature data shown in Table 2: Table 2

[0040] In the present invention, the "sports performance data" stored in the data storage module may include the characteristic data shown in Table 3: Table 3

[0041] In the present invention, the "sports performance data" stored in the data storage module may include multiple feature data as shown in Table 4: Table 4

[0042] It should be pointed out that in the present invention, in order to obtain better technical effects, a preprocessing module is also provided on the above-mentioned athlete selection system, and the preprocessing module is respectively connected to the data acquisition module and the data storage module, so that the original data collected by the data acquisition module can be input into the preprocessing module, and the preprocessing module preprocesses the collected original data, which includes: data cleaning, outlier detection and data normalization processing; after completing the above-mentioned preprocessing of the original data, the preprocessed data is stored in the data storage module, and the preprocessing module can effectively ensure the accuracy and uniformity of the data.

[0043] Accordingly, in the athlete selection system described in the present invention, the data processing module can perform a "data analysis and modeling" process on the data stored in the above-mentioned data storage module, and perform an "evaluation and screening" process on the athletes, and finally output the selection analysis results, so as to utilize the visualization interface of the visualization module to present the athlete's physical test data, evaluation indicators, development trajectory and other information in the time and space dimensions, so as to help users more intuitively understand and compare the potential and adaptability of candidate athletes.

[0044] In the present invention, the designed data processing module specifically includes: a modeling unit and an evaluation unit, the modeling unit is used to perform statistics, learning and data mining based on the physical test spatiotemporal big data in the data storage module to analyze and build a model for athlete selection; and the evaluation unit obtains the athlete selection analysis results based on the constructed athlete selection model. Among them, the evaluation unit specifically assists coaches, selectors or decision makers in selecting and making selection decisions by calculating the comprehensive score, matching degree and potential value of the athlete's physical test data, and assists decision makers in selecting athletes.

[0045] The modeling unit in the data processing module of the present invention specifically extracts key features and patterns in the physical test spatiotemporal big data through statistical analysis, clustering algorithm, time series analysis and other methods to construct an athlete selection model. The athlete selection model finally constructed is shown in FIG. Figure 3.

[0046] Accordingly, the core of the present invention is to construct the above-mentioned athlete selection model based on the above-mentioned physical test spatiotemporal big data, and to combine the above-mentioned Figure 2 , the specific analysis and modeling process of this modeling unit is as follows: 1. Analysis of sports cycles, assessment of athlete status, and prediction of development trends in the time dimension: 1a. Exercise cycle analysis: by using the athlete’s physical test data at different time points , analyze its training cycle It records the cyclical changes in athletes' physical fitness and training effects during a specific period of time, as well as the impact of different training stages on physical fitness development.

[0047] Specifically, firstly, the physical test data in time period 0 to N are calculated according to the training cycle Slice it, expressed as: , and then use the moving average method to calculate the average value in each time period, that is, , this average value is used to reveal the physical fitness status of athletes within a cycle.

[0048] 1b. Athlete status assessment: Evaluate the status and stability of athletes according to defined time periods. By comparing physical test data from different time periods, the fluctuations in the athlete's performance, long-term stability and short-term change trends can be found. This is used to determine whether the athlete's performance is continuously stable, whether there are obvious fluctuations or declines, and whether the athlete can maintain competitive status at critical moments. By evaluating the athlete's status, their potential and adaptability can be judged more accurately.

[0049] Specifically, according to the average value in each training cycle calculated in step 1a above , calculate the standard deviation of each type of physical test data (for specific types of physical test data, see Table 1-3 above) for all exercise cycles in the entire time period, that is:

[0050] in express The average value of the physical test items corresponding to the exercise cycle is It reflects the fluctuation and stability of the corresponding athlete within a defined time period.

[0051] c. Development trend prediction: Use time series analysis methods to predict future physical test data performance, thereby evaluating the potential and growth space of athletes, so as to determine the feasibility and potential level of their future development. This trend prediction can help coaches and selectors better understand the development trajectory of athletes, formulate reasonable selection strategies, and select athletes more accurately.

[0052] Specifically, Figure 3 As shown, a neural network model based on an encoder and a predictor is specifically designed to make reasonable predictions of athletes' future physical test data: First, the average value of the physical fitness data of the last ten exercise cycles in all exercise cycles obtained in step 1b is extracted as the input data of the model, denoted as .in A total of eight cycles of data are input to the encoder's input layer 1 for encoding. After encoding, the output is an embedded feature with a length of 128, which is then bitwise added to the position code. The result of the addition enters the multi-head self-attention layer. Each self-attention layer contains 8 attention heads and the output dimension is 128, so the encoder's output dimension is the number of features of 128. The data of the cycle is input to the input layer 2 of the predictor for embedding feature encoding, and then input to the predictor together with the output feature matrix of the encoder, and the output feature number of the multi-head attention layer 3 and 4 is 128. Finally, the average value of the body measurement data of the next exercise cycle is output through the feedforward network layer of the predictor, which is recorded as , here the predicted value This predicted value indicates the athlete's performance in a certain physical test item in the future sports cycle and indicates the athlete's development potential.

[0053] 2. In the spatial dimension, assess the athletes’ adaptability by analyzing the impact of the geographical environment and the differences in venues: 2a. Analysis of the impact of geographical environment: The climate, altitude and environmental conditions in different locations have a significant impact on the physical performance and adaptability of athletes. By recording the geographical information of the physical test site, including longitude and latitude values, the performance of athletes in different geographical environments can be analyzed and their ability to adapt to different geographical conditions can be evaluated. For example, for endurance athletes, training and testing in high altitude areas may improve their oxygen utilization capacity; while for sprinters, high temperatures and humidity in tropical areas may pose challenges to their performance. Therefore, understanding the impact of the geographical environment on athletes is crucial for selecting appropriate athletes and formulating appropriate training plans.

[0054] Specifically, the present invention classifies the physical test scores of each athlete (Table 3 above) according to the six major geographical regions of China, namely North China, Northeast China, Northwest China, East China, South Central China, and Southwest China, based on the longitude and latitude information. The physical test scores of athletes in corresponding geographical regions are recorded as , and then take the average score of each athlete's physical test scores in these six areas:

[0055] 2b. Regional difference analysis: Different regions or areas have different sports culture, training resources and competition levels. Combine physical test data and geographical information to compare the physical fitness level and potential of athletes from different regions, and evaluate their competitiveness and development prospects in specific regions. Understanding regional differences helps to identify potential athlete resources and explore the advantages of specific regions. For example, a region may perform well in a specific event, so the selection and selection of athletes from that region can be given priority consideration. 2c. Venue Characteristics Analysis: For athlete selection for a specific event, it is important to understand the athlete's performance and adaptability under different venue conditions. By recording the venue characteristics and facility conditions of the test site, the athlete's technical performance and adaptability under a specific venue environment can be evaluated. For example, for football players, the performance on grass and artificial grass may be different; for basketball players, the height of the basket and the hardness of the floor of different indoor venues will also affect their performance. Therefore, for the selection of athletes for a specific event, considering the venue characteristics and conditions can more accurately evaluate the athlete's technical potential and adaptability.

[0056] Specifically, we divide different venues into seven categories according to the venue materials, namely cement polished surface material, PU, ​​acrylic, synthetic rubber, wooden floor, grass, artificial grass (including what). The physical test score of each athlete corresponding to the venue (Table 3 above) is recorded as Then, each athlete's physical test scores were divided according to the seven venues, and the average score was taken:

[0057] 3. Exploring the relationship between time and space dimensions: 3a. Spatiotemporal data mining, combining the time and space dimensions, using the spatiotemporal clustering method to explore the spatiotemporal patterns and association rules in the data. Specifically, here we use the DBSCAN algorithm (this algorithm is well known to those skilled in the art and will not be described in detail here) to cluster the athletes' body shape data (Table 1 above) and physical fitness test data (Table 2 above) in the time (different physical test cycles) and space (scores and coach scores in different regions) dimensions. The spatiotemporal clustering results are recorded as According to the clustering results, the physical fitness data of athletes at specific time and space are recorded to find the time and space when they perform best in physical tests.

[0058] 3b. Use visualization methods such as heat maps and scatter plots to display the data in relation to time and space, so as to intuitively display the relationship between physical test data in time and space. Specifically, the matplotlib tool is used here to draw heat maps and scatter plots of each athlete's physical data, sports performance, and coach ratings with time as the X-axis and space as the Y-axis, respectively. and . This allows for intuitive observation of athletes’ physical performance at a specific time and space.

[0059] 4. Decision makers select athletes: 4a. According to the above analysis, we get the result that , and then compare the physical fitness levels of different athletes according to the selection criteria of different competitive events, and rank the sports performance and coach evaluation. According to the pre-set selection criteria, set thresholds or scoring systems to evaluate the degree of match between athletes and target sports.

[0060] 4b. Visualize the athlete's rating ranking and development potential , spatiotemporal clustering results , spatiotemporal correlation graph and , rank the candidate athletes and compare their indicators so that decision makers can select the athletes.

[0061] Therefore, by comprehensively considering time and space information, the above method uses physical test spatiotemporal big data to more comprehensively, objectively and scientifically evaluate and select athletes. This method uses the integration and analysis of large-scale data to help determine training cycles, predict development trends, evaluate athlete status and stability, and examine geographical environment and venue characteristics to provide more efficient and accurate selection and selection decisions.

[0062] Based on the above content, it is not difficult to see that the system first automatically collects data by connecting the physical test wearable device and the physical fitness tester to the network, obtains the athlete's physical test data, time information and spatial information in real time, and reports it to the data acquisition module. The data acquisition module stores the above data after preprocessing by the preprocessing module to obtain a physical test spatiotemporal big data set; at this time, the system's data processing module can use data analysis and modeling technology to analyze the physical test spatiotemporal big data, thereby extracting key features and patterns from the data, and constructing an athlete selection model. The extracted key features and patterns will be used for athlete selection evaluation and decision-making to obtain selection analysis results; finally, the system presents the selection analysis results to decision makers through the visualization module to assist in understanding and comparing the potential and adaptability of candidate athletes.

[0063] In summary, the athlete selection method and system based on physical test spatiotemporal big data designed by the present invention can evaluate and select athletes more comprehensively, objectively and scientifically by comprehensively considering time and space information. This method utilizes the integration and analysis of large-scale data, which can help determine the training cycle, predict development trends, evaluate the athlete's state and stability, and examine the geographical environment and venue characteristics to provide more accurate selection and selection decisions, which has good promotion prospects and application value.

[0064] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for selecting athletes based on physical test spatiotemporal big data, characterized in that: The method comprises the following steps: S1: collecting and storing the big spatiotemporal data of the physical tests of athletes to obtain a big spatiotemporal data set of the physical tests; S2: establishing an athlete selection model according to the big spatiotemporal data set of the physical tests; and S3: outputting the athlete selection analysis results according to the athlete selection model.

2. The method for selecting athletes according to claim 1, characterized in that: In step S1, specifically: collecting the athlete's physical measurement data, and forming the physical measurement spatiotemporal big data according to the time information and location information of the athlete during the physical measurement; storing the physical measurement spatiotemporal big data, and obtaining a physical measurement spatiotemporal big data set according to a number of the physical measurement spatiotemporal big data.

3. The method for selecting athletes according to claim 2, characterized in that: In step S1, the physical test data includes: the athlete's body shape data, physical fitness data, sports performance data and coach scoring data obtained through physical test.

4. The method for selecting athletes according to claim 1, characterized in that: In step S2, specifically: statistics, learning and data mining are performed on the physical measurement spatiotemporal big data in the physical measurement spatiotemporal big data set to extract key features and patterns from the physical measurement spatiotemporal big data and establish an athlete selection model.

5. The method for selecting athletes according to claim 4, characterized in that: In step S2, key features and patterns are extracted and an athlete selection model is constructed through statistical analysis, clustering algorithm, regression analysis, association mining, anomaly detection and time series analysis methods.

6. An athlete selection system based on physical test spatiotemporal big data, which is applied to the athlete selection method according to any one of claims 1 to 5, characterized in that: include: A data acquisition module, a data storage module and a data processing module, wherein the data acquisition module is communicatively connected to the data storage module, and the data storage module is communicatively connected to the data processing module; Among them, the data acquisition module is used to collect the physical measurement spatiotemporal big data of athletes, the data storage module is used to store the physical measurement spatiotemporal big data, and obtain the physical measurement spatiotemporal big data set; the data processing module is used to establish an athlete selection model according to the physical measurement spatiotemporal big data set, and based on the athlete selection model, output the selection analysis results of candidate athletes.

7. The athlete selection system according to claim 6, characterized in that: It also includes a visualization module, which is in communication with the data processing module. The visualization module receives the material selection analysis result and displays the material selection analysis result in a visual manner.

8. The athlete selection system according to claim 6, characterized in that: The stored data in the data storage module includes the athlete's ID information and the time information, geographic information and physical test data of the athlete; wherein the physical test data includes: the athlete's body shape data, physical fitness data, sports performance data and coach scoring data obtained through the physical test.

9. The athlete selection system according to claim 6, characterized in that: It also includes a preprocessing module, which is connected to the data acquisition module and the data storage module respectively, and the preprocessing module preprocesses the collected initial data, and the preprocessing includes: data cleaning, outlier detection and data normalization.

10. The athlete selection system according to claim 7, characterized in that: The data processing module includes: a modeling unit and an evaluation unit; wherein the modeling unit analyzes and models the physical measurement spatiotemporal big data in the data storage module, specifically through statistical analysis, clustering algorithm, regression analysis, association mining, anomaly detection and time series analysis methods, to extract key features and patterns, and build an athlete selection model; the evaluation unit evaluates and obtains the athlete selection analysis results based on the established athlete selection model.

Images