Quantum Enhanced Machine Learning for Unobtrusive Stress Monitoring

Abstract

Prolonged stress can negatively impact a person's mental health leading to multiple diseases. Stress monitoring can be efficiently done with the help of artificial intelligence technology combined with the benefits of quantum computing. The main aim of the paper is to analyze quantum enhanced machine learning techniques in predicting the stress of knowledge workers at office through multiple modalities. A general overview of popular quantum enhanced machine learning methods such as Quantum Support Vector Machine (QSVM), Variational Quantum Classifier (VQC) and Quantum K-Nearest Neighbor (QKNN) methods has been presented after studying the literatures of past 10 years. Besides, these models have been implemented on multimodal SWELL-KW dataset, which contains knowledge worker's computer interaction, facial expressions, body postures, heart rate variability and skin conductance data recorded in various working conditions. Further, the impacts of Quantum Principal Component Analysis based feature reduction on their performances have been analyzed. Experimental results show that for the current dataset, QSVM model with PCA on heart rate variability and skin conductance data results in highest accuracy of 0.8.