Stacking Ensemble Model for Stress Detection using Sleeping Habits

This Stress is defined as a state of emotional response or mental tension triggered by problematic circumstances or challenging situations, called as stressors. It is a normal response of human beings whenever threats and emotions are addressed. The degree of stress levels experienced by humans varies based on various negative outcomes in life. To mitigate stress levels, it is essential to know the various stress levels. This research explores how sleep disturbance is related to various physical, mental, and social issues. The main aim of this research is in detection of the levels of stress through machine learning algorithms in sleep-related behaviours. The multiclass dataset used comprises information about the numerous sleep habits and the corresponding stress levels. The pre-processed data are applied to Machine Learning (ML) techniques like Naive Bayes (NB), Decision Tree (DT), Gradient Boosted Tree (GBT), Random Forest (RF) and Random Tree (RT) for classification. The ensemble stacking method is proposed to obtain better results than the existing results. The findings of the study can be computed accurately, facilitating in addressing sleep related issues in a timely manner.