Role of ML and DL for IoT based Societal Applications

Abstract

Due to the emergence of IoT-cloud based infrastructure and the wide availability of smart handheld devices, a novel group of societal applications is coming up. Most of these applications are data intensive and consumer driven. Machine learning and deep learning techniques fuelled the success of such applications in making a better-connected society. The recent pandemic caused by COVID-19 further emphasizes the need for such applications not only to the healthcare domain but also for fitness tracking, entertainment and many other domains. These applications heavily rely on the cloud infrastructure for providing sophisticated user experience. So, the presentation will cover a comprehensive view of this group of societal applications and the research challenges associated to it along with the implementation aspects. The discussion will revolve around three representative research directions-overview of machine learning and deep learning techniques used in this field, challenges and solutions of smartphone and wearable sensing data, and case studies on health monitoring and indoor localization. The presentation will conclude with a discussion on the open research issues in this area.