SVM Based Classification Approach to Identify Suitable Beginners for Tennis Based on Anthropometric Measurements

Abstract

Anthropometric measurements are generally used to determine and predict achievement in different sports. An athlete’s anthropometric and physical characteristics may be an important precondition for successful participation in any given sport. Further, anthropometric profiles indicate whether the player would be suitable for the competition at the highest level in a specific sport. The main aim of this study is to measure accuracy of the given classification algorithms. Each and every data mining algorithm provides separate prediction accuracy details. This study investigates an integration of four data mining algorithms (Naïve Bayes, Decision Trees, Random Forest, and Support Vector Machine) and an Ensemble approach (bagging, boosting, and stacking). In this research, Push up, Sit-up, Cardiac respect endurance, flexibility, agility, Height, Weight, Upper Arm Relaxed Girth, Fore Arm Girth, Chest Girth, Wrist Girth, Waist Girth, Thigh Girth, Calf Girth, Angle Girth, Acromial Radial Length, Radials lion Dactyl ion, Foot Length, body composition, fore arm length, hand length, Explosive power, Breath hold in time, resting heart rate, volume of oxygen, Force vital capacity, force explotaty volume in 1 second, Upper arm radius, and Leg Length are used. In this paper, classification performance of these models has been evaluated using Accuracy, Precision, Recall, F-Measure, MCC, ROC Area, PRC Area, Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) etc. The results of this study indicated that SVM was the most suitable algorithm to classify Anthropometric measurements of tennis players.