My Ph.D. work uses wearable sensing and machine learning to understand safety and well-being in construction. Earlier work at Missouri State focused on context-aware language systems for smart homes.
My core Ph.D. research studies fatigue, safety and well-being during physically demanding construction tasks using wearable biosignals. It combines experimental design, signal processing and machine learning under real-world constraints.
Biosignal datasets from real workers are small and noisy. I use generative adversarial networks to augment them, improving how well models generalize in safety-critical, human-centered settings. This work led to our Construction Research Congress 2026 paper.
Exoskeletons can reduce physical strain, but what do they do to attention and mental workload? I evaluate their physical and cognitive effects during construction postures using wearable sensors and structured protocols.
Identified contextual intent from voice commands and conversations so home automation systems can adapt to new environments.
SBERT, attention, clustering, zero-shot learning
A lightweight framework for extracting context from short spoken text with low computational cost.
Transfer learning, SBERT, attention
Designed ten metamorphic relations to verify the robustness of context recognition models.
Software testing, model validation
Classified attention states from facial landmarks on the DAiSEE dataset.
MediaPipe FaceMesh, Xception
A neural classifier that categorizes papers from their titles and abstracts.
TF-IDF, feature selection
An end-to-end text analysis tool built for non-technical users.
NLP, usability
A hybrid computer vision tracker balancing speed and accuracy.
Mean Shift, particle filter, sliding window
Time-series forecasting on historical prices from 1987 to 2019.
LSTM
PHP, MySQL, JavaScript
WordPress, PHP, MySQL
PHP, MySQL
Web application
Google location APIs
Usage data, AI techniques
Computer graphics, C++
Expression parsing