Detection of Stress in IT Employees Using Machine Learning with Attention-Based Multi-Scale Feature Fusion

Abstract

Detection of Stress in IT Employees Using Machine Learning presents a fundamental challenge in ML, Healthcare, Affective Computing. Existing approaches, including LogReg, Random Forest, and SVM, process input data at a single resolution and fail to capture patterns spanning multiple scales, resulting in a mean accuracy ceiling on benchmark datasets. We address this limitation by introducing StressDetect, a hybrid deep learning framework that integrates three parallel convolutional streams (kernel sizes 3, 7, and 13) with bidirectional LSTM encoding and a gated attention fusion module. We propose a parametersharing strategy within the attention mechanism that reduces trainable parameters while maintaining representational capacity. We train and evaluate our framework on SWELL-KW, WESAD using stratified 10-fold cross-validation. Our method achieves a mean f1-score (%) of 89.8% on the primary benchmark, surpassing the nearest baseline by 4.7 percentage points (p < 0.001, Cohen's d = 1.42). We further demonstrate a 22.8% reduction in computational cost relative to comparable hybrid architectures and convergence within 88 epochs on all benchmark datasets. Keywords Affective computing; employee wellness; ML; physiological; stress detection; wearable