A machine learning approach to stress prediction and its accuracy based on sleep and heart rate metrics

In an era especially since stress plays a huge part in overall health now more than ever, learning the physiological factors in stress such as quality sleep, amount of sleep, and heart rate is important. This paper aims at establishing the correlation between these variables and stress levels within various employees categories. Using random forest, Decision trees, SVM, GBM, and elastic net regression algorithm we propose to predict stress using health data. This work aims to determine which of the models in this study has the highest prediction accuracy , and showing which of the models are most effective in such data.