AUTHOR=Javed Abdul Rehman , Khan Habib Ullah , Alomari Mohammad Kamel Bader , Sarwar Muhammad Usman , Asim Muhammad , Almadhor Ahmad S. , Khan Muhammad Zahid TITLE=Toward explainable AI-empowered cognitive health assessment JOURNAL=Frontiers in Public Health VOLUME=Volume 11 - 2023 YEAR=2023 URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2023.1024195 DOI=10.3389/fpubh.2023.1024195 ISSN=2296-2565 ABSTRACT=Explainable Artificial Intelligence (XAI) is of paramount importance to various domains, including healthcare, fitness, skill assessment, and personal assistants, to understand and explain the decision-making process of the Machine Learning (ML) model. Smart homes embedded with smart devices and sensors enable many context-aware applications to recognize physical activities. This paper presents \textit{XAI-HAR}, a novel XAI empowered Human Activity Recognition (HAR) and cognitive health assessment approach based on key features identified from the data collected from smart sensors located at different places of a sustainable smart home. \textit{XAI-HAR} identifies a set of new features (i.e., the total number of sensors used in a specific activity) as \textit{Physical Key Features Selection (PKFS)} based on weighting criteria. Next, it presents \textit{Statistical Key Features Selection (SKFS)} (i.e., mean, standard deviation) to handle the outliers and higher-class variance. A set of extensive experiments demonstrates the superior performance of \textit{XAI-HAR} over existing techniques. For explainability, \textit{XAI-HAR} uses local interpretable model agnostic (LIME) with Random Forest (RF) classifier. \textit{XAI-HAR} achieves 0.96\% f-score for health and dementia classification, 0.95\%, 0.97\% for activity recognition of dementia and healthy individuals, respectively.