2023_A Multivariate and Machine Learning Classification Model for Team and Individual Sports Based on Anthropometric and Fitness Components Among Terengganu Sukma Athletes
| Format: | General Document |
|---|
| _version_ | 1860798074698334208 |
|---|---|
| building | INTELEK Repository |
| collection | Online Access |
| collectionurl | https://intelek.unisza.edu.my/intelek/pages/search.php?search=!collection3 |
| copyright | Copyright©PWB2025 |
| country | Malaysia |
| date | 2023-12-26 |
| format | General Document |
| id | 15776 |
| institution | UniSZA |
| internalnotes | Sila masukkan subject wajib Dissertations, Academic. Terima kasih... |
| originalfilename | AMULTI~1.PDF |
| person | Noor Aishah Binti Kamarudin |
| recordtype | oai_dc |
| resourceurl | https://intelek.unisza.edu.my/intelek/pages/view.php?ref=15776 |
| sourcemedia | Server storage Scanned document |
| spelling | 15776 https://intelek.unisza.edu.my/intelek/pages/view.php?ref=15776 https://intelek.unisza.edu.my/intelek/pages/search.php?search=!collection3 General Document Malaysia Library Staff (Top Management) Library Staff (Management) Library Staff (Support) Terengganu Faculty of Health Sciences English application/pdf 1.5 Server storage Scanned document Universiti Sultan Zainal Abidin UniSZA Private Access UNIVERSITI SULTAN ZAINAL ABIDIN SAMBox 2.3.4; modified using iTextSharp™ 5.5.10 ©2000-2016 iText Group NV (AGPL-version) Copyright©PWB2025 164 AMULTI~1.PDF 2023_A Multivariate and Machine Learning Classification Model for Team and Individual Sports Based on Anthropometric and Fitness Components Among Terengganu Sukma Athletes Noor Aishah Binti Kamarudin 2023-12-26 Machine learning—Applications in sports science Introduction: Performance monitoring has been used for many years in various fields, including sports, regardless of whether it is an individual sport or a team sport, to evaluate and monitor the performance of athletes. The inconsistent performance of the Terengganu contingent in the Malaysian Games over the past three years has created opportunities for athletes from other states because lack of a scientific modelling approach in determining the ability of an athlete in individual or team sports. The purpose of this study is to look into the relationship between physical ability measurements and sports participation. A machine learning classification model approach was used to tell the difference between individual sports athletes and team sports athletes based on their anthropometry, health, and fitness. Methodology: Involved a quantitative method with 218 male athletes from individual and team sports (mean age of individuals = 18.44 (1.43); team athletes = 17.31 (2.19)). Five individual sports and three group sports categories were studied using an experimental research design and an opportunistic sampling method. Athletes underwent anthropometric tests (weight, height, sitting height, and front length) and physical fitness tests (sit and reach, 1 minute sit up, push up, hand grip, VO2max, medicine ball throw, 20 m speed, vertical jump, standing broad jump, stork stand, T-test), which were then recorded into the data collection form. To predict and categorise individual and team athletes, the Artificial Neural Networks (ANN), k-nearest Neighbour (k-NN), Support Vector Machines (SVM), Decision Trees, Random Forests (RF), Naive Bayes (NB), and Logistic Regression (LR) were used. Based on performance in the most important variables, K means clustering (k-MC) divides athletes into individual and team sports. Result: Analysis of k-MC for anthropometric test data showed that the LR model outperformed the RF and NB in determining 119 athletes grouped into team sports and 99 athletes in the individual sports categories. For the health-related fitness component HRFC test, ANN showed the highest percentage of accuracy compared to k-NN and Decision Tree, with 161 athletes grouped into team sports and another 56 athletes in individual sports categories. In the skill related fitness components SRFC test, the SVM model overcame the RF and k-NN by grouping 153 team sports athletes and 65 individual sports athletes’ categories. In this study, LR (anthropometry), ANN (health component), and SVM (skill component) models were used to predict the sport type of each athlete. These models had the highest accuracy for each component compared to the other models and all of the classification models had a classification accuracy of between 0.9 and 1.0. Conclusion: This study revealed the value of employing multivariate analysis and machine learning in identifying high-potential athletes in individual and group sports by taking into consideration a variety of important performance characteristics. It also saves money and time in the long-term during talent enhancement and development programmes, in line with the needs of a specific sport. Consequently, athletes can assess their abilities in sports in which they excel. With these results, coaches can figure out how to make the best training plan for each sport Dissertations, Academic Sila masukkan subject wajib Dissertations, Academic. Terima kasih... Machine Learning in Sports Performance Analysis Anthropometric Profiling Of Athletes Fitness Components in Team VS. Individual Sports Thesis |
| spellingShingle | 2023_A Multivariate and Machine Learning Classification Model for Team and Individual Sports Based on Anthropometric and Fitness Components Among Terengganu Sukma Athletes |
| state | Terengganu |
| subject | Machine learning—Applications in sports science Dissertations, Academic |
| summary | Introduction: Performance monitoring has been used for many years in various fields, including sports, regardless of whether it is an individual sport or a team sport, to evaluate and monitor the performance of athletes. The inconsistent performance of the Terengganu contingent in the Malaysian Games over the past three years has created opportunities for athletes from other states because lack of a scientific modelling approach in determining the ability of an athlete in individual or team sports. The purpose of this study is to look into the relationship between physical ability measurements and sports participation. A machine learning classification model approach was used to tell the difference between individual sports athletes and team sports athletes based on their anthropometry, health, and fitness. Methodology: Involved a quantitative method with 218 male athletes from individual and team sports (mean age of individuals = 18.44 (1.43); team athletes = 17.31 (2.19)). Five individual sports and three group sports categories were studied using an experimental research design and an opportunistic sampling method. Athletes underwent anthropometric tests (weight, height, sitting height, and front length) and physical fitness tests (sit and reach, 1 minute sit up, push up, hand grip, VO2max, medicine ball throw, 20 m speed, vertical jump, standing broad jump, stork stand, T-test), which were then recorded into the data collection form. To predict and categorise individual and team athletes, the Artificial Neural Networks (ANN), k-nearest Neighbour (k-NN), Support Vector Machines (SVM), Decision Trees, Random Forests (RF), Naive Bayes (NB), and Logistic Regression (LR) were used. Based on performance in the most important variables, K means clustering (k-MC) divides athletes into individual and team sports. Result: Analysis of k-MC for anthropometric test data showed that the LR model outperformed the RF and NB in determining 119 athletes grouped into team sports and 99 athletes in the individual sports categories. For the health-related fitness component HRFC test, ANN showed the highest percentage of accuracy compared to k-NN and Decision Tree, with 161 athletes grouped into team sports and another 56 athletes in individual sports categories. In the skill related fitness components SRFC test, the SVM model overcame the RF and k-NN by grouping 153 team sports athletes and 65 individual sports athletes’ categories. In this study, LR (anthropometry), ANN (health component), and SVM (skill component) models were used to predict the sport type of each athlete. These models had the highest accuracy for each component compared to the other models and all of the classification models had a classification accuracy of between 0.9 and 1.0. Conclusion: This study revealed the value of employing multivariate analysis and machine learning in identifying high-potential athletes in individual and group sports by taking into consideration a variety of important performance characteristics. It also saves money and time in the long-term during talent enhancement and development programmes, in line with the needs of a specific sport. Consequently, athletes can assess their abilities in sports in which they excel. With these results, coaches can figure out how to make the best training plan for each sport |
| title | 2023_A Multivariate and Machine Learning Classification Model for Team and Individual Sports Based on Anthropometric and Fitness Components Among Terengganu Sukma Athletes |
| title_full | 2023_A Multivariate and Machine Learning Classification Model for Team and Individual Sports Based on Anthropometric and Fitness Components Among Terengganu Sukma Athletes |
| title_fullStr | 2023_A Multivariate and Machine Learning Classification Model for Team and Individual Sports Based on Anthropometric and Fitness Components Among Terengganu Sukma Athletes |
| title_full_unstemmed | 2023_A Multivariate and Machine Learning Classification Model for Team and Individual Sports Based on Anthropometric and Fitness Components Among Terengganu Sukma Athletes |
| title_short | 2023_A Multivariate and Machine Learning Classification Model for Team and Individual Sports Based on Anthropometric and Fitness Components Among Terengganu Sukma Athletes |
| title_sort | 2023_a multivariate and machine learning classification model for team and individual sports based on anthropometric and fitness components among terengganu sukma athletes |