Publication Date

Spring 2025

Degree Type

Master's Project

Degree Name

Master of Science in Computer Science (MSCS)

Department

Computer Science

First Advisor

Sayma Akther

Second Advisor

William Andreopoulos

Third Advisor

Nandhini Gounder

Keywords

Augmentation, Class imbalance, Deep learning, DeepSense, Evaluation, GAN-PCA, Generalization, Human Activity Recognition (HAR), Imputation, Missing sensor values, Opportunity dataset, Wearable sensor data

Abstract

Applications of ubiquitous computing, including health monitoring, sports analytics, and ambient-assisted living, rely on Human Activity Recognition (HAR) using wearable sensors. However, model robustness is challenged by missing sensor values, class imbalance, inter-subject variability, and temporal noise. This work proposes a complete HAR pipeline that addresses these challenges through sampling, time-series augmentation, dynamic feature handling, and GAN-PCA-based imputation. Built on the DeepSense architecture, the model integrates convolutional feature extraction with bi-GRUs for temporal modeling. The system is evaluated using 5-fold cross-validation, subject-aware holdout, and LOSEO strategies on the Opportunity dataset. Results demonstrate consistent accuracy across folds and strong generalization to unseen sessions and subjects, significantly outperforming baseline models. This study highlights the effectiveness of hybrid deep learning approaches in real-world HAR scenarios

Download

Available for download on Wednesday, May 20, 2026

Share

COinS