Recognition of human activity based on data from sensors in wearable devices

Abstract

In this work, I researched wearable computing and human activity recognition. I analyzed the data from wearable sensors. In the first chapter, I defined wearable computing, sensors, and human activities. I provided an overview of commercial wearable devices. In the second chapter, I explored the tools and methods of data analysis in multi-class classification problems and time series analysis. I described the stages of the analytical process and the dataset I used. In the third chapter, I evaluated my analytical process. I analyzed the data from an electrocardiogram, thoracic electrical bioimpedance, and electrodermal activity signals. The activities analyzed were: neutral, emotional, mental, and physical. The data comes from the open-access dataset. At the end of the chapter, I summarised the findings. In the last chapter, I concluded my research. My interest lies in commercial and popular wearable devices. That is the group I focused on throughout this work.