Ensemble Filter Based Feature Selection Technique for Classification of Human Activity Recognition

Through the advancement of wearable sensors, wireless communication, and machine learning techniques, Assistive Technologies (AT) which endorse autonomous, active, and healthy lifestyles are emerging in recent years. Among these advances, Human Activity Recognition (HAR) is one of the most innovative means to support or monitor human activities. However, misclassifications such as intra-class variation and inter-class overlap in similar activities degrade classification accuracy in HAR. To improve the recognition of daily human activities, handcrafted features of time-domain and frequency-domain are combined. However, several extracted features may not be significant in describing the activities. Therefore, this research aims to propose a feature selection technique for optimal human activities recognition. The methodology proposed for this research is the Ensemble Filter (Relief-F and mRMR) to select the most relevant and less redundant features. Although a filter feature ranking approach is commonly used in related studies, most works fail to consider the threshold limit to exclude unnecessary and redundant features. An ensemble Random Forest (RF) was used as the base classifier to evaluate the performance of the hybrid algorithm. The results demonstrate that the proposed ensemble filter selection was beneficial in reducing the total number of features while improving overall classification accuracy.