Decoding Human Stress Levels using GRU-LSTM Models

Stress is a major concern affecting health and productivity worldwide. Traditional methods for detecting stress often fall short in accuracy and accessibility. To tackle this, we developed a model using Long Short- Term Memory (LSTM) networks for effective stress detection. Our study utilized a stress dataset featuring seven key attributes: Nervousness, Unable to Control, Worry, Trouble in Relaxation, Restlessness, Irritability, and Fear. After comprehensive preprocessing, our LSTM model was trained to accurately classify stress levels. The results were impressive, with the model achieving an accuracy of 91.73% and a precision of 0.9%, demonstrating its robustness and potential for real-world application.