Smart Algorithm as a Machine Learning Application for Elderly People Safety Using Mobile Platforms

This chapter introduces a smart algorithm in Machine Learning (ML) applications that aims to enhance elderly people safety in daily life. The main focus was the idea of creating a fall detecting mobile application for people who are paralyzed from waist down (Paraplegic) or have injured their legs. The algorithm uses the training data collected from real-world scenarios to closely learn the best calibration for system parameters. The smart phone fall detection system detects falls using sensors available in the phone itself. The chapter starts off by explaining the acronym of the application PFAST, then discusses the functions, necessity for building such an app. It also discusses the problems that prior applications have faced and how exactly will it work, which algorithm and sensors from Android devices will be used. Using various UML diagrams, it displays how the user will interact with the system. We address the issue “sudden falls” which is very common in elder people and has led to serious health injuries.