• Version
• Download 5
• File Size 307.65 KB
• File Count 1
• Create Date 16 May 2025
• Last Updated 16 May 2025

HUMAN SLEEP STRESS DETECTION USING MACHINE LEARNING ALGORITHMS

Authors:

Palani Bharathi V

Abstract

Machine learning (ML), a new technique, allows computers to simulate human behaviour. It may provide significant benefits for a variety of industries, including industry, agriculture, and medicine. This article's main focus is the healthcare sector, namely the identification of human stress when we sleep. Stress may be classified as either eustress or distress. Persistent discomfort may lead to serious health problems. The hormones cortisol and adrenaline are involved in the body's stress response. Reliable detection methods are necessary for stress management to be successful. The objective of our study is to assess stress levels based on human criteria using a machine-learning algorithm. By comparing the outcomes of machine learning models such as Random Forest Classifier, Naive Bayes, and K-Nearest Neighbour with those of individual variables such as age, sex, cp, trest bps, FBS, respect, thali, old peak, slope, ca, thal, and target, we shall ascertain if stress is present in people. If we want to achieve even greater benefits, future research may concentrate on feature engineering and ensemble techniques.

Keywords: Machine learning, human stress, Random Forest Classifier, Naive Bayes, and K- Nearest Neighbour