We design AI-powered models that extract actionable insights from complex, high-dimensional, and often unstructured data. This line of research spans:
Healthcare AI: Developing interpretable deep learning models for predicting dementia and other cognitive disorders using clinical notes, consultation logs, and behavioral signals.
Social Network Mining: Analyzing signed networks, user behavior, and content propagation to detect spam, evaluate trust, and design robust recommender systems.
Recommender Systems: Building robust, trustworthy recommendation models that are resilient to manipulation and sensitive to user trust and preference dynamics, with applications in social platforms and information services.
Our goal is to ensure that the insights derived are not only accurate but also explainable and practically useful in fields such as medicine and online platforms.
Dementia Prediction Using Hierarchical Attention and Evaluation of Context Quality
Kyu-haeng Lee, Seokbeom Lim, Ilju Lee, Ok Kim, Hyun Woo Jung, Sehwan Kim, Hee Jung Kim, Keunsoo Kang, and Jung Jae Lee
Alzheimer’s Association International Conference (AAIC), 2025
link
Sihyun Jeong, and Kyu-haeng Lee
MDPI Applied Sciences, 2020
Toward Trustworthy Social Network Services: A Robust Design of Recommender Systems
Giseop Noh, Hayoung Oh, Kyu-haeng Lee, Chong-kwon Kim
Journal of Communications and Networks, 2015