Machine Learning Approach for Recognizing Human Activity: An Improved Support Vector Machine over K-Nearest Neighbor Method

Abstract

Aim: The main aim of the research is to enhance Human Activity Recognition using Support Vector Machine (SVM) over K-Nearest neighbor (KNN) Materials and Methods: SVM and KNN are implemented in this research work. Sample size is calculated using G power software and determined as 10 per group with pretest power 80%, threshold 0.05% and CI 95%. Result: SVM provides a higher of 93.32% compared to KNN algorithm with 92.02% in predicting lack of accuracy in Human Activity Recognition. The significance value of p=0.009 (p<0.05) for SVM is statistically insignificant. Conclusion: SVM algorithm predicts human activity recognition better than K-Nearest Neighbor Algorithm.