Human Stress Detection Based on Sleeping Habits Using Machine Learning Algorithms

Abstract

Stress is a natural response to challenging or unavoidable situations, often referred to as stressors. Understanding stress levels is crucial to preventing negative impacts on daily life. Sleep disturbances, in particular, are closely linked to various physical, mental, and social issues. This study explores how machine learning can be used to detect stress based on sleep-related behaviors. The dataset includes different sleep patterns and corresponding stress levels. Six machine learning models—Multilayer Perceptron (MLP), Random Forest, Support Vector Machine (SVM), Decision Trees, Naïve Bayes, and Logistic Regression—were used to classify the data after preprocessing. By comparing their performance, we identified the most accurate model. The results show that the Naïve Bayes algorithm performs best, achieving an accuracy of 91.27%, along with high precision, recall, and F-measure values. It also recorded the lowest mean absolute error (MAE) and root mean squared error (RMSE). These findings suggest that machine learning can effectively estimate stress levels, allowing for early intervention and betterstress management.