Semi-Supervised Learning and Data Augmentation in Wearable-based Momentary Stress Detection in the Wild

TL;DR

This paper explores semi-supervised learning and data augmentation techniques to improve wearable-based stress detection models using unlabeled data, achieving significant performance gains in real-world conditions.

Contribution

It introduces a semi-supervised framework combining data augmentation and consistency regularization for stress detection from wearable sensor data, leveraging unlabeled data effectively.

Findings

Improved stress classification accuracy by 7.7% to 13.8%.

Effective use of unlabeled data enhances model robustness.

Validated on three real-world wearable sensor datasets.

Abstract

Physiological and behavioral data collected from wearable or mobile sensors have been used to estimate self-reported stress levels. Since the stress annotation usually relies on self-reports during the study, a limited amount of labeled data can be an obstacle in developing accurate and generalized stress predicting models. On the other hand, the sensors can continuously capture signals without annotations. This work investigates leveraging unlabeled wearable sensor data for stress detection in the wild. We first applied data augmentation techniques on the physiological and behavioral data to improve the robustness of supervised stress detection models. Using an auto-encoder with actively selected unlabeled sequences, we pre-trained the supervised model structure to leverage the information learned from unlabeled samples. Then, we developed a semi-supervised learning framework to leverage the unlabeled data sequences. We combined data augmentation techniques with consistency regularization, which enforces the consistency of prediction output based on augmented and original unlabeled data. We validated these methods using three wearable/mobile sensor datasets collected in the wild. Our results showed that combining the proposed methods improved stress classification performance by 7.7% to 13.8% on the evaluated datasets, compared to the baseline supervised learning models.