Research

My research interests mainly focuse on affective computing, signal processing, and deep learning. My goal is to build reliable and interpretable deep learning systems that can be applied in real-world scenarios. I am / have been working on the following exciting research topics:

1. Design machine learning systems to address the varies challenges (e.g., label scarcity, domain shift, label base rate difference) for human-related data
K. Feng, J. B. Duong, K. Carta, S. Walters, G. Margolin, A. C. Timmons, T. Chaspari, “A Semi-supervised Few-shot Learning Approach With Domain Adaptation for Personalized Stress Detection Within Dating Couples,” submitted to ICASSP 2023.

2. Knoweledge driven deep learning models to potentially help the clinical mental disorder diagnose
K. Feng and T. Chaspari, “Toward Knowledge-driven Speech-Based Models of Depression: Leveraging Spectrotemporal Variations in Speech Vowels,” IEEE International Conference on Biomedical and Health Informatics (BHI 2022), Ioannina, Greece, September, 2022.

3. Data mining and analysis for social emotions
K. Feng, P. Zanwar, A. Behzadan, and T. Chaspari, “Exploring Speech Cues in Web-mined COVID-19 Conversational Vlogs,” ACM Multimedia-2020 workshop on Media Analytics for Societal Trends (MAST 2020), October 2020, DOI: 10.1145/3423268.3423584