June, 2022. VOL.14. ISSUE NO.2 https://hrdc.gujaratuniversity.ac.in/Publication/article?id=10103 E-journal Page | 1033 ENSEMBLE-BASED HUMAN ACTIVITY RECOGNITION FOR MULTI RESIDENTS IN SMART HOME ENVIRONMENT John W. Kasubi* The Local Government Training Institute, Dodoma, Tanzania Manjaiah D. Huchaiah Department of Computer Science, Mangalore University, India Ibrahim Gad Faculty of Science, Tanta University, Tanta, Egypt Mohammad Kazim Hooshmand Department of Computer Science, Mangalore University, India Abstract The ensemble methods play a vital role in machine learning for obtaining a high-performing model for the study dataset, and combining multiple classifiers to build a best-predictive model. On the other hand, Feature selection helps to remove irrelevant variables in the dataset in order to construct better predictive models. Therefore this research aimed to develop a robust model for activity recognition for multi-residents in smart homes using the ARAS dataset. The study employed Tree-based feature selection to cater to feature selection; two ensemble approaches, hard and soft voting, in line with five base learner classifiers: Logistic Regression (LR), Linear Discriminant Analysis (LDA), Naïve Bayes (NB), Random Forest (RF), and K-nearest neighbor (KNN), were applied to build the human activity recognition (HAR) model. The experimental results show that RF performed best compared to the rest of the classifiers, with an accuracy of 99.1%, and 99.2% in houses A and B, respectively. In comparison to prior findings, Feature Selection and ensemble methods enhanced prediction accuracy in the ARAS dataset. Keywords: Human Activity Recognition, Ensemble Methods, Feature Selection, ARAS, Smart Home, Multi Residents 1.0 INTRODUCTION Human activity recognition (HAR) is a broad field of research that focuses on detecting a person's activities of daily living (ADL) collected by the sensor data. The HAR is a fascinating research topic that may be used to detect human activity in various domains, including surveillance, healthcare, energy, and water resource management [1]. Ensemble methods are machine learning approaches that integrate many classifiers intending to obtain a high-performing classifier for the research dataset; they merge multiple classifiers to construct the best-predictive model. To generate a robust predictive model, the ensemble trains numerous weak learners, creates a collection of learners, and combines them [2, 3]. This study used two methods: hard and soft voting ensembles with five base learning classifiers were used: LR, LDA, NB, RF, and KNN [4, 5]. Feature selection (FS) is an integral part of model development because it affects the model's efficiency, helps understand the data for easy interpretations, improves model accuracy, reduces model overfitting, and shortens