Stress Classification and Personalization: Getting the most out of the least

TL;DR

This paper introduces a CNN-based stress detection framework that simplifies data processing by eliminating feature extraction, achieves high accuracy with a single sensor modality, and emphasizes the importance of personalizing stress models.

Contribution

A novel CNN approach for stress classification using only one sensor modality without feature engineering, outperforming existing methods and highlighting personalization.

Findings

Achieved 92.85% classification accuracy.

Outperformed state-of-the-art techniques.

Showed the importance of personalized stress models.

Abstract

Stress detection and monitoring is an active area of research with important implications for the personal, professional, and social health of an individual. Current approaches for affective state classification use traditional machine learning algorithms with features computed from multiple sensor modalities. These methods are data-intensive and rely on hand-crafted features which impede the practical applicability of these sensor systems in daily lives. To overcome these shortcomings, we propose a novel Convolutional Neural Network (CNN) based stress detection and classification framework without any feature computation using data from only one sensor modality. Our method is competitive and outperforms current state-of-the-art techniques and achieves a classification accuracy of $92.85\%$ and an $f1$ score of $0.89$. Through our leave-one-subject-out analysis, we also show the importance of personalizing stress models.