Random Forest for Human Daily Activity Recognition

Abstract

Abstract Machine learning classifiers are often used to evaluate the predicting accuracy of human activity recognition. This study aimed to evaluate the performance of random forest (RF) compared to other classifiers with considering the time taken to build the models. Human activity daily living data, namely walking, walking upstairs, walking downstairs, sitting, standing, and lying down were collected from smartphone-based accelerometer with sampling frequency of 50Hz. The dataset was evaluated using artificial neural network (ANN), k-nearest neighbors (KNN), linear discriminant analysis (LDA), naïve Bayes (NB), support vector machine (SVM), and random forest (RF). The results of the study showed that RF indeed predicted the activities with the highest accuracy. However, the time taken to build the models using RF was the second-longest after ANN.