Scholarlink

There is a significant detrimental impact that stress has on a person’s quality of life as well as their mental and physical health. Stress is a prevalent psychological illness that is brought on by challenging circumstances. This study provides an overview of the psychological and physiological effects of stress, while also highlighting the significance of early detection due to the potential adverse consequences that may arise. Five feature selection methods were tested to identify the optimal feature selection method: SelectKBest, Principal Component Analysis, Forward Selection, Recursive Feature Elimination, Backward Elimination. The purpose of this research paper is to understand the physiological metrics that are used to classify stress. There are a total of five feature selection methods and eight machine learning algorithms that are being explored for deployment. When it comes to Recursive Feature Elimination and forward feature selection methods, the Random Forest algorithm and the K-Nearest Neighbor algorithm are the ones that perform the best among all machine learning algorithms.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Stress Classification Using Machine Learning Techniques: Comparative Study

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Varad Sawant,
  • Santwana Gudadhe,
  • Anurag Karpe,
  • Lokesh Kad,
  • Pratham Kubetkar,
  • Anuradha Thakare

摘要

There is a significant detrimental impact that stress has on a person’s quality of life as well as their mental and physical health. Stress is a prevalent psychological illness that is brought on by challenging circumstances. This study provides an overview of the psychological and physiological effects of stress, while also highlighting the significance of early detection due to the potential adverse consequences that may arise. Five feature selection methods were tested to identify the optimal feature selection method: SelectKBest, Principal Component Analysis, Forward Selection, Recursive Feature Elimination, Backward Elimination. The purpose of this research paper is to understand the physiological metrics that are used to classify stress. There are a total of five feature selection methods and eight machine learning algorithms that are being explored for deployment. When it comes to Recursive Feature Elimination and forward feature selection methods, the Random Forest algorithm and the K-Nearest Neighbor algorithm are the ones that perform the best among all machine learning algorithms.