skip to main content

Explore Research Products in the NSF-PAR

This content will become publicly available on June 12, 2024

Title: Semi-Supervised Learning for Wearable-based Momentary Stress Detection in the Wild
Physiological and behavioral data collected from wearable or mobile sensors have been used to estimate self-reported stress levels. Since stress annotation usually relies on self-reports during the study, a limited amount of labeled data can be an obstacle to 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 propose a two-stage semi-supervised learning framework that leverages wearable sensor data to help with stress detection. The proposed structure consists of an auto-encoder pre-training method for learning information from unlabeled data and the consistency regularization approach to enhance the robustness of the model. Besides, we propose a novel active sampling method for selecting unlabeled samples to avoid introducing redundant information to the model. We validate these methods using two datasets with physiological signals and stress labels collected in the wild, as well as four human activity recognition (HAR) datasets to evaluate the generality of the proposed method. Our approach demonstrated competitive results for stress detection, improving stress classification performance by approximately 7% to 10% on the stress detection datasets compared to the baseline supervised learning models. Furthermore, the ablation study we conducted for the HAR tasks supported the effectiveness of our methods. Our approach showed comparable performance to state-of-the-art semi-supervised learning methods for both stress detection and HAR tasks.  more » « less
Award ID(s):
1840167 2047296
NSF-PAR ID:
10454828
Author(s) / Creator(s):
;
Date Published:
Journal Name:
Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
Volume:
7
Issue:
2
ISSN:
2474-9567
Page Range / eLocation ID:
1 to 23
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ( , JMIR mHealth and uHealth)
    Background

    As mobile health (mHealth) studies become increasingly productive owing to the advancements in wearable and mobile sensor technology, our ability to monitor and model human behavior will be constrained by participant receptivity. Many health constructs are dependent on subjective responses, and without such responses, researchers are left with little to no ground truth to accompany our ever-growing biobehavioral data. This issue can significantly impact the quality of a study, particularly for populations known to exhibit lower compliance rates. To address this challenge, researchers have proposed innovative approaches that use machine learning (ML) and sensor data to modify the timing and delivery of surveys. However, an overarching concern is the potential introduction of biases or unintended influences on participants’ responses when implementing new survey delivery methods.

     Objective

    This study aims to demonstrate the potential impact of an ML-based ecological momentary assessment (EMA) delivery system (using receptivity as the predictor variable) on the participants’ reported emotional state. We examine the factors that affect participants’ receptivity to EMAs in a 10-day wearable and EMA–based emotional state–sensing mHealth study. We study the physiological relationships indicative of receptivity and affect while also analyzing the interaction between the 2 constructs.

     Methods

    We collected data from 45 healthy participants wearing 2 devices measuring electrodermal activity, accelerometer, electrocardiography, and skin temperature while answering 10 EMAs daily, containing questions about perceived mood. Owing to the nature of our constructs, we can only obtain ground truth measures for both affect and receptivity during responses. Therefore, we used unsupervised and supervised ML methods to infer affect when a participant did not respond. Our unsupervised method used k-means clustering to determine the relationship between physiology and receptivity and then inferred the emotional state during nonresponses. For the supervised learning method, we primarily used random forest and neural networks to predict the affect of unlabeled data points as well as receptivity.

     Results

    Our findings showed that using a receptivity model to trigger EMAs decreased the reported negative affect by >3 points or 0.29 SDs in our self-reported affect measure, scored between 13 and 91. The findings also showed a bimodal distribution of our predicted affect during nonresponses. This indicates that this system initiates EMAs more commonly during states of higher positive emotions.

     Conclusions

    Our results showed a clear relationship between affect and receptivity. This relationship can affect the efficacy of an mHealth study, particularly those that use an ML algorithm to trigger EMAs. Therefore, we propose that future work should focus on a smart trigger that promotes EMA receptivity without influencing affect during sampled time points.

     
    more » « less
  2. ; ; ; ; ( , Winter Conference on Applications of Computer Vision (WACV), 2021.)
    null (Ed.)
    We propose a semi-supervised learning approach for video classification, VideoSSL, using convolutional neural networks (CNN). Like other computer vision tasks, existing supervised video classification methods demand a large amount of labeled data to attain good performance. However, annotation of a large dataset is expensive and time consuming. To minimize the dependence on a large annotated dataset, our proposed semi-supervised method trains from a small number of labeled examples and exploits two regulatory signals from unlabeled data. The first signal is the pseudo-labels of unlabeled examples computed from the confidences of the CNN being trained. The other is the normalized probabilities, as predicted by an image classifier CNN, that captures the information about appearances of the interesting objects in the video. We show that, under the supervision of these guiding signals from unlabeled examples, a video classification CNN can achieve impressive performances utilizing a small fraction of annotated examples on three publicly available datasets: UCF101, HMDB51, and Kinetics. 
    more » « less
  3. ; ; ; ; ( , 2022 IEEE International Conference on Smart Computing (SMARTCOMP))
    We explore the effect of auxiliary labels in improving the classification accuracy of wearable sensor-based human activity recognition (HAR) systems, which are primarily trained with the supervision of the activity labels (e.g. running, walking, jumping). Supplemental meta-data are often available during the data collection process such as body positions of the wearable sensors, subjects' demographic information (e.g. gender, age), and the type of wearable used (e.g. smartphone, smart-watch). This information, while not directly related to the activity classification task, can nonetheless provide auxiliary supervision and has the potential to significantly improve the HAR accuracy by providing extra guidance on how to handle the introduced sample heterogeneity from the change in domains (i.e positions, persons, or sensors), especially in the presence of limited activity labels. However, integrating such meta-data information in the classification pipeline is non-trivial - (i) the complex interaction between the activity and domain label space is hard to capture with a simple multi-task and/or adversarial learning setup, (ii) meta-data and activity labels might not be simultaneously available for all collected samples. To address these issues, we propose a novel framework Conditional Domain Embeddings (CoDEm). From the available unlabeled raw samples and their domain meta-data, we first learn a set of domain embeddings using a contrastive learning methodology to handle inter-domain variability and inter-domain similarity. To classify the activities, CoDEm then learns the label embeddings in a contrastive fashion, conditioned on domain embeddings with a novel attention mechanism, enforcing the model to learn the complex domain-activity relationships. We extensively evaluate CoDEm in three benchmark datasets against a number of multi-task and adversarial learning baselines and achieve state-of-the-art performance in each avenue. 
    more » « less
  4. ; ; ; ( , 16th ACM/EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services (MobiQuitous), 2019)
    Human activity recognition (HAR) from wearable sensor data has recently gained widespread adoption in a number of fields. However, recognizing complex human activities, postural and rhythmic body movements (e.g., dance, sports) is challenging due to the lack of domain-specific labeling information, the perpetual variability in human movement kinematics profiles due to age, sex, dexterity and the level of professional training. In this paper, we propose a deep activity recognition model to work with limited labeled data, both for simple and complex human activities. To mitigate the intra- and inter-user spatio-temporal variability of movements, we posit novel data augmentation and domain normalization techniques. We depict a semi-supervised technique that learns noise and transformation invariant feature representation from sparsely labeled data to accommodate intra-personal and inter-user variations of human movement kinematics. We also postulate a transfer learning approach to learn domain invariant feature representations by minimizing the feature distribution distance between the source and target domains. We showcase the improved performance of our proposed framework, AugToAct, using a public HAR dataset. We also design our own data collection, annotation and experimental setup on complex dance activity recognition steps and kinematics movements where we achieved higher performance metrics with limited label data compared to simple activity recognition tasks. 
    more » « less
  5. ; ; ; ( , 2022 18th International Conference on Distributed Computing in Sensor Systems (DCOSS))
    The scarcity of labeled data has traditionally been the primary hindrance in building scalable supervised deep learning models that can retain adequate performance in the presence of various heterogeneities in sample distributions. Domain adaptation tries to address this issue by adapting features learned from a smaller set of labeled samples to that of the incoming unlabeled samples. The traditional domain adaptation approaches normally consider only a single source of labeled samples, but in real world use cases, labeled samples can originate from multiple-sources – providing motivation for multi-source domain adaptation (MSDA). Several MSDA approaches have been investigated for wearable sensor-based human activity recognition (HAR) in recent times, but their performance improvement compared to single source counterpart remained marginal. To remedy this performance gap that, we explore multiple avenues to align the conditional distributions in addition to the usual alignment of marginal ones. In our investigation, we extend an existing multi-source domain adaptation approach under semi-supervised settings. We assume the availability of partially labeled target domain data and further explore the pseudo labeling usage with a goal to achieve a performance similar to the former. In our experiments on three publicly available datasets, we find that a limited labeled target domain data and pseudo label data boost the performance over the unsupervised approach by 10-35% and 2-6%, respectively, in various domain adaptation scenarios. 
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email