Web-Semantic-Driven Machine Learning and Blockchain for Transformative Change in the Future of Physical Education

Machine learning is playing an increasingly important role in education. This article examines its potential to bring about transformative change in this field. By using machine learning algorithms, physical education teachers can gather and analyze data on student performance and behavior. This enables them to create personalized learning experiences that cater to the unique needs of each student. Machine learning can also track and assess student progress, providing educators with valuable insights into the effectiveness of their teaching strategies. Furthermore, it can optimize the design of physical education curricula and assessments, making them more efficient and effective. Additionally, machine learning offers a more objective and accurate approach to evaluating and grading students. This paper discusses the challenges and opportunities associated with integrating machine learning into physical education, including ethical considerations and potential limitations.


WEB-SEMANTIC-DRIVEN MACHINE LEARNING AND BLoCKCHAIN FoR TRANSFORMATIVE CHANGE IN THE FUTURE oF PHySICAL EDUCATIoN
Physical education is an important component of a well-rounded education because it gives students the knowledge, skills, and habits they need to live a healthy and active lifestyle.Traditional physical education techniques, on the other hand, are typically one-size-fits-all and fail to take into account each student's particular needs, interests, and abilities.As a result, proponents have called for a more effective and personalized approach to physical education-one that makes use of technology to improve student performance.Physical and emotional well-being is both dependent on physical education (PE), according to R. Trigueros et al. (2019).Furthermore, research has demonstrated that physical education improves a person's psychosocial qualities (Kapsal et al., 2019).Physical education offers several health benefits, including maintaining the body's energy balance and managing the body mass index (BMI) to treat chronic noncommunicable illnesses like as obesity (Bednar & Rouse, 2020).Exercise is also necessary for the development of the cardiovascular, metabolic, musculoskeletal, and overall bodily systems.It has several advantages, and there is a strong relationship between physical exercise and health (Yiran, 2021).Incorporating physical activity into a curriculum can assist students in honing their athletic ability, developing regular physical exercise routines, and maintaining good mental health throughout time.This, in turn, improves their academic performance and prepares them for labor-market demands (Zheng Keqiang, 2022;Baena-Morales & Gonzá-Víllora, 2022).Physical education is crucial in higher education institutions, notably colleges.Despite its numerous advantages, many colleges fail to adequately expose their students to these advantages due to inadequate delivery methods.To really grasp the ideas and abilities of physical education, a modernized education system with interactive teaching methods is essential.To maximize the promise of machine learning in physical education, blockchain technology can be leveraged to offer secure and immutable storage of student data while ensuring data integrity and privacy.Physical education can adopt innovative and dependable platforms by leveraging the transparency and decentralized nature of blockchain, empowering students to take charge of their educational journey and promoting a more inclusive and collaborative educational ecosystem, as shown in Figure 1 (Liu & Li, 2023;Dziatkovskii, 2023).The potential of machine learning has resulted in a dramatic shift in the field of physical education, and teachers can now develop individualized learning experiences that meet the specific needs of each student by using machine learning algorithms to evaluate data on behavior and performance.Additionally, machine learning can be utilized to monitor and assess student progress, providing teachers with crucial data regarding the efficacy of their instructional tactics (You, 2010).Machine learning may help make physical education programming and assessments more effective and efficient.
Machine learning can assist teachers to better understand each student's strengths and weaknesses because it provides a more precise and fair method of evaluating students.Although machine learning has considerable promise for physical education, there are a number of issues and limitations to consider.The research looks into how machine learning might change traditional classroom instruction.It examines the ethical issues of collecting and analyzing big student data sets, as well as the significance of professional development and training for educators in order to effectively employ machine learning techniques (Cambria & White, 2014;LeCun et al., 2015).It analyses the advantages and disadvantages of using machine learning in physical education, as well as potential constraints, ethical problems, and the importance of professional development and teacher training.This study's ultimate goal is to inspire discussion about how technology is changing physical education, as well as to provide recommendations and insights to educators and policymakers who want to use machine learning to improve student outcomes (Hirschberg & Manning, 2015;Lee & Lee., 2020).Previous studies have also used AI and physical education to improve the health (Li et al., 2023;Gupta et al., 2023;Singh & Jaiswal, 2022;Leghari & Ali, 2023).This study serves to complement the precedent research.Alsareii et al. (2022) uses AI to generate personalized suggestions for physical activity among older adults based on their distinct talents, preferences, and health status.The study uses AI calculations to assess children's actual work levels, providing substantial insights into the factors that influence active work in this segment (Gülü et al., 2023).This review investigates actual training performance with the use of AI, using the internet of Things (IoT) and AI to develop an astute actual schooling system.It achieves this by providing educators with insights about student behavior and execution by utilizing AI to organically break down actual schooling execution (Ullmann, 2019;Ruzmatovich & G'ayratjon o'g'li, 2023).The study recommends an insightful actual school system that offers individualized and enrapturing actual instruction encounters by using web of things and AI innovation (Wang & Du, 2022).In this work, information on understudy lead and execution are examined, utilizing AI to upgrade the plan of the actual schooling educational program (Mendoza Torralba, 2020).This review sheds light on the social elements impacting understudy execution by anticipating actual schooling grades utilizing AI and informal organization investigation (Taş et al., 2021).This research provides an AI-based framework to assist actual schooling instructors with sorting out their classes to assist actual schooling instructors with making appealing and effective classes (Lee & Lee, 2021;Thornton et al., 2023;Silva et al., 2007).Man-made awareness (simulated intelligence)-enabled items and programs are now ubiquitous and frequently cooperate with customers (Göksel & Bozkurt, 2019).Housman distinguishes two primary applications of artificial intelligence: anticipating the outcomes of human-ordered information to accomplish mundane tasks and using human-planned calculations to make decisions that are nearly equal to those of people.Computer-based intelligence consistently completes tasks and gives options to customers, learning from inputs and determining appropriate paths for them.Scientific and technical improvements have an impact on educational models, systems, and organizations in ways that go beyond changes in teaching methods and resources (McArthur et al., 2005).Despite the fact that cutting edge instructive innovation is widely used, there is still uncertainty about how to effectively deal with its establishment, expand the use of instructive asset applications, and improve educational viability through extensive recreation (Baker, 2007).As a result, instructional technology research is badly needed.Education is a critical social pillar because it provides people with the tools they need to prepare for the future and discover personal pleasure.The rise of AI technology brings both benefits and challenges for the educational industry.Chatbots that use natural language processing, digital textbooks, big data analysis for personalized learning, speech recognition and synthesis, and learning management systems are all being developed (Roll & Wylie, 2016).The majority of AI technology can be applied to education and educational policy.Predictive analytic technology can assist struggling students by assessing their learning levels, estimating future academic performance, and forecasting academic outcomes (McCabe & Trevathan, 2008).Future lesson plans must therefore incorporate this technology.AI has had an impact on all aspects of life, including education, and it has the potential to improve educational resources, transform attitudes towards learning, and revolutionize time-honored teaching practices.Physical education technology is a multidisciplinary field that includes computer science, education, sports, and other disciplines.AI solutions to real-world problems may be beneficial (Xian, 2010).Despite its importance in preparing for future educational systems, there has been relatively little study on the application of AI to physical education.Machine learning has the ability to improve physical education by delivering individualized and exciting learning experiences, improving curriculum design, and providing insights into student performance and behavior (Han, 2011;Lim et al., 2013).To improve physical education technology, specialized study objectives and themes for exploring computer networks, the internet, and information and communication technology must be devised.In general, AI has the potential to alter traditional classroom instruction and promote lifelong learning.We tackled this issue in this study because there is currently no system that can anticipate blockchain technology and physical education.

METHoDoLoGy Data Set (College Basketball)
The Center for Global Development collects time sequence data to investigate the relationship between college basketball and the championship adjusted offensive efficiency (AdjO) and adjusted defensive efficiency (AdjD) outputs.AdjO calculates a team's offensive efficiency (i.e., points scored per 100 possessions) against an average Division I defense.Meanwhile, AdjD estimates a team's defensive efficiency (i.e., points allowed per 100 possessions) against an average offense.We collected this dataset from Kaggle with the name "college basketball dataset."

Linear Regression Model (LR)
Linear regression (LR) is a statistical strategy that forecasts the outcome of a dependent variable by using a large number of independent variables.The goal of multiple linear regression (MLR) is to develop a model that characterizes the linear connection between the dependent variable and the independent components.It is an extension of the ordinary least squares (OLS) regression, which uses a single explanatory variable.In the context of college basketball, (1) represents both the dependent and independent components.

Logistic Regression
Logistic regression is a statistical method for binary classification tasks in which the outcome or target variable has two classes.It is a classification strategy, not a regression technique, despite its name.This calculation determines the likelihood that a given piece of information belongs to a specified class.The computed capability, also known as the sigmoid capability, is used to determine the chance that the information belongs to a specific class.The computed capability takes any actual number and turns it to a value between 0 and 1. Calculated relapse evaluates the relationship between the dependent variable and at least one autonomous element by computing probability using the strategic capability.This straight model predicts using the weighted sum of the information highlights; the calculated capability is then used to change the information and provide the final likelihood score.This strategy is frequently used due to its productivity, interpretability, and ease of use especially when the model's interpretability is simple, or the classes are immediately detachable.Logistic regression is a linear technique, but it can be extended to deal with more complex correlations by incorporating polynomial features or interaction factors.The logistic regression statistical approach is used in this article to model a discrete response variable in two-dimensional independent variable problems such as yes/no, success/failure, and offensive/defensive.The general logistic regression model is as follows.See ( 2), (3), and (4).
The logistic regression model forecasts the likelihood of a financial crisis event occurring (Y = 1) or not (Y = 0) using an observed response variable, 'Y,' and an unobserved latent variable, Y.The explanatory variable matrix 'x' signifies the explanatory variable error, as well as the explanatory variable parameter matrix, and it contains covariates of interest.This model represents the probability of an observation belonging to either group, which may be described using Hosmer and Lemeshow's formula, which can be converted into a linear model in the parameters.The maximum likelihood estimate calculates the probability that a fresh observation (x) belongs to one of two groups.The observation is given to the probability category with the highest value.

Support Vector Machines (SVM)
The support vector machines (SVM) model distinguishes itself by using a separating hyperplane to project input vectors into a high-dimensional eigenspace via linear or nonlinear projection.Tibshirani, Friedman, Moguerza, and Muoz present a kernel function that converts data into a linearly separable characteristic space.Each data point in this new space is an abstract point in a p-dimensional space, where p is the number of variables in the dataset.The updated data (xi, yi n i=1) consists of abstract dots with yi representing one of two possible classes (one or one).The equidistant hyperplane to the closest point of each class in the new space is represented by w T(x) + b = 0. See ( 6) and ( 7).
SVM is a machine learning method that maps input vectors into a high-dimensional space using linear or nonlinear core functions.It makes use of an isolating hyperplane to distinguish between various material types.The SVM determines the optimal edge hyperplane to increase the distance between the two classes.Support vectors, or the points closest to the hyperplane, are used to identify it.Following that, the discriminant capacity is extended once more into the underlying space.In contrast to neural networks, SVM is based on the mathematically sound structural risk reduction principle, which makes machine learning easier to understand (Kolb & Kolb, 2005).SVM is a modularized AI method with a simple design, curved streamlining, and sparse presentation.It is made up of two modules that can be used as the foundation of a measured design: a chunk capability and a learning machine.The SVM classifier employs a kernel function; radial basis function (RBF), programmable logic (PL), signature (SIG), and link (LN) kernels can be used.Because the RBF component is less vulnerable to anomalies, it is commonly used in landslip powerlessness planning.The accuracy of the SVM solution may improve if kernel parameters such as g, d, and r are correctly defined.SVM has been used successfully in a variety of assessments to demonstrate landslip weakness.

Convolutional Neural Networks (CNNs)
Convolutional Neural Networks (CNNs) are primarily intended for the processing and analysis of visual data, such as images or movies.However, they are rarely used directly for numerical data in formats such as CSV spreadsheets, which lack the spatial correlations and pixel-level information that CNNs are designed to extract.A CN''s architecture is based on the visual cortex of the human brain and is comprised of many layers.Convolutional layers extract features from input data such as edges, textures, and forms using filters or kernels convolution methods are used in these layers to keep the spatial link between pixels.Pooling layers sample the information obtained by convolutional layers, lowering spatial dimensionality, and controlling overfitting by keeping the most critical input.The network's fully connected layers perform classification or regression tasks based on features learned at previous levels.These layers connect every neuron in the previous and subsequent levels, allowing the network to recognize complex patterns and correlations in the input.

Physical Education Prediction Model
It is also critical for national to increase literacy rates.Increased access to physical activity leads to better outcomes and national development.Physical exercise prices can be predicted using two methods: technical data analysis and fundamental analysis, which takes into account growth factors that influence physical exercise.These algorithms forecast physical activity based on current and historical data.Machine learning techniques such as linear regression, logistic regression (LR), SVM, and CNN are used to predict physical exercise results for students.Machine learning has been extensively researched for its effectiveness in boosting physical activity development.The physical activity model depicted in Figure 2 presents the proposed physical exercise model and subsequent following processes in the prediction model with the following parameters: We install several libraries, Anaconda and Jupiter Notebook; in the second step, we have a CSV file as an input file; in the third step, we preprocess the data; and in the fourth step, we analyze physical exercise using machine learning models (linear regression, LR, SVM, and CNN).

RESULTS
In post-season analysis for championship teams, championship AdjO: AdjO was the stat with the highest correlation with championship teams.These teams had a median AdjO of around 120.This median was higher than the 7 5t h percentile of all other teams in all other rounds.Even championship teams between the 2 5t h and 5 0t h percentiles had a higher AdjO than the median AdjO of all other teams in other rounds, as shown in Figures 3 and 4. Figure 5 shows the confusion matrix of offence and defense.Looking at the p-value between the different rounds, championship teams had a significantly higher AdjO than R68, R64, and R32.However, there wasn't a significant difference among teams in rounds 16 and 17.Championship AdjD: The Adj of championship teams, while superior to teams in other rounds, was less significant than AdjO.Because AdjD is a defensive stat, the median AdjD of championship teams is 94, as shown in Figure 5. Championship teams have a statistically significant better AdjD than R68 and R64 teams, but no statistical advantage over R32+ teams.However, the interquartile range is tightly grouped between 90 and 95, which may signify that championship-caliber teams have high AdjD consistency.Championship 2-and 3-point percentage: Other stats were explored but were not found to have a high correlation with championship teams.2PO and 2PD didn't show a lot of distinction between championship-caliber teams and teams from other rounds.In fact, the median 2PO was in line with the R16+ teams.For 2PD, the champion only showed a significant difference from the R68 teams.While these stats are important, they do not have a high individual correlation with winning a championship.Correlation between stats and AdjO/AdjD: Because AdjO and AdjD both seem to be the most significant stats when determining championship teams, it can be helpful to take a closer look at the correlation between other stats and AdjO/AdjD.Below is a heat map that shows the relative correlation, with darker being more correlated and lighter being less correlated, as shown in Figures 6 and 7.The stats that have a weak-to-strong correlation between 0.30 and 1, with AdjO, REF, GO, 2PO, 3PO, and TOR.The stats that have a weak-to-strong correlation with AdjO are EF GD, 2PD, and 3PD.The discussion section of an article on machine learning tools and physical education is a critical component that enables researchers to explore the implications of their findings in greater detail.In this section, it is important to highlight the strengths and limitations of machine learning tools in physical education, address ethical concerns related to the use of student  1, which illustrates machine learning technologies have the power to completely transform the physical education sector by giving instructors individualized, data-driven methods for instruction.Educators can give individualized instruction that tends to every understudy's particular prerequisites by analyzing information on conduct, execution, and inclinations of individual understudies.Furthermore, solid areas and shortcoming can be recognized, understudy execution can be consequently dissected and input on, and the plan of actual schooling educational programs can be streamlined with AI procedures.AI can likewise be utilized to distinguish and classify different proactive tasks, giving educators valuable data about how their students take part in active work and how AI can assist educators in creating really fascinating and useful classes (Perl, 2004).

Limitations of Machine Learning in Physical Education
The use of AI processes in practical training is subject to a number of constraints, notwithstanding the predicted benefits.One of the key reasons for concern is the possibility of bias in AI computations, which could result in unfair or prejudiced outcomes for select ethnic groups.Furthermore, an overreliance on AI methods may result in a one-size-fits-all approach to training and a reduction in the role that educators play in the growing experience.Finally, there are concerns about student information security and protection, as well as the need to ensure that information is gathered and handled morally and dependably (McCullagh, 2010;Ghasemzadeh et al., 2009).
Moral Issues.The collecting and processing of student data is one of the most serious ethical challenges linked with the use of machine learning technology in physical education.It is vital to ensure the moral and thoughtful collection and utilization of understudy information, as well as the execution of acceptable safeguards to ensure understudy security and individual data privacy.Furthermore, as Ghasemzadeh and Jafari (2010) and Lamb et al. (2010) point out, it is critical to address concerns about the possibility of bias in AI computations and ensure that decisions based on these outcomes are simple and equivalent for each understudy.
Methods and Ideas for Further Research.To address the requirements and moral quandaries associated with the employment of AI developments in the sphere of actual training, it is critical to consistently investigate fresh strategies and establish optimal methodology for gathering, assessing, and simply selecting.This includes the development of moral frameworks to guide the appropriate use of student data and the continuation of evaluation to select the accepted ways for integrating artificial intelligence tools into real-world preparation.To ensure that teachers have the capacities and expertise to apply AI methods in their work, it is likewise, basic to give them the help and preparing they need.In light of everything, the utilization of AI strategies in actual training can possibly reform the discipline by working with individualized, information-driven techniques for guidance and learning.Notwithstanding, it is basic to address the downsides and moral problems connected with their application, as well as to continue to investigate new thoughts and industry best practices for integrating AI into the act of actual training (Bartlett & Lamb, 2011).

DISCUSSIoN AND RECoMMENDATIoNS
Human growth depends on physical education and integrating technology into the classroom can enhance the curriculum, promote fresh perspectives on PE, and have other useful effects.Artificial intelligence has the potential to play a significant role in the future of physical education due to its ability to offer a more thorough educational experience (Silva et al., 2007).By identifying and maximizing each student's athletic potential, AI in physical education may provide individualized instruction and individualized learning experiences.AI can also foster human potential and creative thinking.AI also offers a range of educational resources to maintain students' interests and motivation in their studies.Three categories apply to AI-based physical education: learner, instructor, and educational work.Students have access to high-level physical exercises, both virtual and actual, data collection and analysis, and communication with teachers.AI helps teachers by providing multiple answers to students' problems, real-time class status information, and efficient assessment and learning management.Finally, AI helps teachers by reducing administrative tasks, so they may spend more time improving the quality of their classroom instruction.Physical education teachers should actively develop and implement information resources, as well as choose physical education technology that is multimedia and network based.They must also accept new concepts and theories in order to thoroughly rethink their teaching approaches (Baca & Kornfeind, 2012).In order to modernize sports instruction, technology must be integrated into physical education programs.This necessitates a transformation in views towards knowledge, talent, education, and the viewpoints of physical education teachers exhibit to students, research concepts, class concepts and assessments, and educational philosophy.Physical education teachers must have a thorough understanding of educational technology, including fundamental knowledge and skills, problem-solving ability, physical educational technology recognition, and social responsibility.Physical educational technology is crucial in the field of physical education, and its implementation can facilitate physical education innovation.The advancement of AI and physical educational technology has stimulated countless investigations, but a lack of academic communication between study domains disperses research capacity.As a result, it is required to build an integrated research workforce, provide a stable environment for the development of current physical educational technology, and develop overall AI-related applications for comprehensive sports educational innovation.This study investigates the use of AI in physical education, concentrating on customized physical education lessons, knowledge provision, learner evaluation, learner counselling methods, and future physical education teachers' skills and roles (Baca et al., 2009;Novatchkov & Baca, 2013).To construct physical education classes utilizing AI technology, however, requires more comprehensive, useful, and in-depth research for every region-a topic not covered in this study.Additionally, experimental studies are needed to assess how well AI algorithms match empirical verification outcomes in real-world applications.The importance of AI applications in physical educational technology and their potential applicability in other educational domains are highlighted by the study's consequences for practice and policy.Combining AI and blockchain technology in conventional classroom instruction may yield additional benefits in the future.Coordinating the changelessness and straightforwardness of blockchain innovation with manmade consciousness' modified learning capacities can assist us with establishing a protected and solid climate for putting away and surveying understudy execution information.This decentralized methodology tends to cause worry about understudy security and information control while saving information trustworthiness and propelling moral information on the board.Additionally, scholars, policymakers, and educational institutions may be able to collaborate more easily with one another thanks to the integration of blockchain technology.By embracing the combination of blockchain and simulated intelligence, we can enable actual training with information driven bits of knowledge, tweaked growth opportunities, and a more comprehensive and progressive instructive scene.Blockchain innovation will be incorporated into our cycles to guarantee the respectability and protection of the gathered volleyball spike execution information.Before being added to a distributed and immutable blockchain ledger, each data item will be timestamped and cryptographically hashed.This circulated approach will safeguard the information from unlawful modification, saving its dependability and legitimacy.Reception of blockchain innovation will likewise work with straightforward information trade and participation across researchers, partners, and different volleyball crews.Shrewd agreements can be utilized to characterize information access privileges, which save information proprietorship and protection while empowering explicit gatherings to add to and access explicit areas of the dataset.To make a decentralized and trustless information framework and work on the convenience and validity of our AI models for volleyball spike investigation, we intend to incorporate blockchain innovation into our review.The volleyball local area in general will acquire from a more careful and helpful way to deal with sports execution checking made conceivable by the utilization of blockchain innovation and AI.The two fundamental pieces of the spike in volleyball are the methodology and the leap.The methodology assists the player with preparing for a higher jump by assisting them with moving on a level plane toward the ball.The player can strike the ball in the right spiking position by jumping vertically toward it.Maintaining proper alignment is crucial for generating velocity in the spike, and executing the run-up and take-off with precision can yield a quick shot.To prevent harm and joint strain, athletes should also be able to land well.Figure 1 depicts action diagrams of the technical structure of the volleyball spike, which has been demonstrated in tests to improve volleyball performance.Machine learning is a field that allows algorithms to learn from data and experience, allowing them to perform tasks such as prediction, decision making, and model construction.The data offers useful information on variable relationships, but the set of possible patterns may be too large for the training data to cover.To develop effective models for fresh data, machine learning algorithms must generalize successfully.Many machine learning algorithms have been applied successfully to scientific and technical challenges, yielding quantifiable results in a variety of settings (Göksel & Bozkurt, 2019).However, the method chosen is determined by the task at hand and the data being examined.Decision trees, random forests (RF), neural networks (NNs), SVMs, and k-nearest neighbors (k-NN) are among the most frequently used algorithms.Choosing the right algorithm for a given task can lead to accurate and efficient results.
In the future, physical education could use web-semantic-driven machine learning to tailor individual training regimens, while also using blockchain for safe data administration and promoting decentralized fitness networks through incentivized collaboration and knowledge exchange.Actual schooling (PE) innovation coordination has a promising future, particularly with the normal expansion of blockchain and man-made consciousness (computer-based intelligence).Simulated intelligence's custom-made learning capacities joined with blockchain's changelessness and straightforwardness hold extraordinary potential.This connection works with a decentralized procedure that puts an accentuation on information honesty and mindful organization by offering a reliable and secure climate for the capacity and investigation of understudy execution information.The security of delicate information and the goal of issues with information control and understudy protection make this mechanical mix a distinct advantage in instructive information across the board.Besides, the blend of blockchain innovation and man-made consciousness (artificial intelligence) may work with smooth information trade and participation between scholarly organizations, analysts, and policymakers, breaking beforehand unrealistic hindrances, all the while.Tolerating this association prompts the production of a more imaginative and comprehensive instructive climate, as well as information driven bits of knowledge and modified opportunities for growth that lay the foundation for an imaginative future in actual schooling.

CoNCLUSIoN
Machine learning technologies that deliver personalized and data-driven educational tactics have the potential to transform physical education.Be that as it may, the gamble of calculation inclination and moral issues concerning understudy security should be considered while carrying them out.The advantages of applying AI strategies in actual training are evident in spite of these disadvantages, and further review and application are expected to make ideal methods and helpful methodologies.To fully realize the capabilities of AI advancements in practical training, scientists, instructors, and politicians must work together.This partnership can aid in the disclosure of effective strategies as well as the moral and conscientious collection and use of information.Furthermore, it is critical to provide teachers with continuous training and support to ensure they have the abilities and understanding required to effectively incorporate AI systems into their professions.In light of this, incorporating AI improvements into traditional homeroom guidance could benefit the profession by providing understudies with recreated and interesting learning opportunities.However, in order to fully realize this technology's promise, more study, collaboration, and ethical application are required.The immutability and transparency of blockchain have the potential to enhance confidence among children, parents, and instructors by maintaining the integrity of student progress data and improving physical education accountability.By merging AI and blockchain breakthroughs, we can create a powerful and trustworthy environment that encourages participation, development, and information-driven experiences in order to advance genuine training into a more significant and customized future.Using a collegiate basketball dataset and a machine learning technique, we obtain 78% accuracy in our proposed physical education solution for CNN.This study improves both the use of blockchain technology for data storage and human physical welfare.This research is beneficial to human physical health as well as the use of blockchain to save data.

Figure 1 .
Figure 1.Block chain and physical exercise information

Figure 2 .
Figure 2. Physical education prediction model

Figure 6 .
Figure 6.Prediction of college basketball and AdjO