Abstract

Stress, a prevalent aspect of modern life, significantly impacts mental and physical well-being. Traditional stress detection methods often rely on subjective self-reporting, limiting their reliability. This project proposes a novel approach to stress detection using machine learning algorithms, specifically the Random Forest Classifier, to predict human stress levels based on various physiological parameters. The system aims to enhance accuracy and classification granularity by classifying stress levels into five categories: low/normal, medium low, medium, medium high, and high. By incorporating features such as snoring range, respiration rate, body temperature, limb movement rate, blood oxygen levels, eye movement, number of hours of sleep, and heart rate, the system provides a comprehensive assessment of individuals' stress experiences. Additionally, users receive personalized recreational suggestions based on their stress levels, promoting holistic stress management strategies. The proposed system demonstrates feasibility in technical implementation, operational usage, economical viability, and behavioral acceptance, offering a promising solution for stress detection and management in today's society.

You can download abstract from here

Technologies Used

Modules

1. User

🧁 Login/Signup

🧁 Upload Food images

🧁 Classes - 101 Food Items

🧁 Voice recognition (user can say food name instead of image)

🧁 View History of previous predictions

🧁 Specify quantity and size of food Item

🧁 View nutritional value of detected food Item

🧁 Check if food is safe for users with allergies