Multimodal Workplace Monitoring for Human Activity Recognition

The aim of this work is to design and implement a system that deals with the recognition of human behavior on issues of mental nature. Since the "invasion" of technology in human daily life is continuously growing, a plethora of novel applications have emerged. Therefore, there exists a growing interest for applications that will serve as an aid for people experiencing problems with work fatigue, stress and anxiety. In essence, the goal of this work is to combine ideas, theories and methods from the research areas of psychology and information technologies, towards the automatic recognition of activities that occur within a work environment. To this goal, we use a series of observations including actions, behaviors and situations that constitute sources of fatigue, anxiety and stress, based on measurements coming from several peripheral parts of a computer and also from force sensitive resistor sensors placed on the users chairs. These observations take place during working time. For behaviour recognition we use supervised machine learning techniques. We demonstrate the proposed approach using a real-life scenario.