A review of physiological signal processing via Machine Learning (ML) for personal stress detection

Abstract

Personal stress is maintained and measured by Machine learning. The device which is wearable has been used for the monitoring of personal self stress and data collection. In this research, it has been talked about the factors by which the physiological signal of the stress has been assessed. On the other hand, different type of technology has been used for the detection of the personal stress such as Electrocardiography (ECG) and many other devices. The observation and difficulties has been seen in this research by using this device and the technology. Stress disorder or ailment is one of the most common ailments in all individuals around the world. Stress and anxiety can greatly influence the life, emotion, behavioural pattern and thinking attributes of individuals. It is important to address this issue sooner or later. Psychological signal processing through machine learning effectively assists to detect the stress disorder at an early stage. The general system often considers some variables to detect stress. They are electrocardiogram, galvanic response, heart rate, respiration and many other elements. The ML tend to use algorithms to compare and contrast data to fetch effective e results. The paper has also carried out a statistical analysis based on three variables to fetch a proper result that provided the study group to comprehend a better understanding of the scenario. The researchers have taken the 'percentage of stress rate' cases' are considered independent variables whereas 'usage of a machine learning system' is considered a dependant variable. The study group has fetched and collected numerous data related to these three variables to get a better understanding.