Ye DING, Ph.D., is a professor and master’s supervisor at the School of Computer Science and Technology, Dongguan University of Technology. His work mainly focuses on data analytics and artificial intelligence. He has published over 40 papers in well-known international conferences and journals, filed more than 80 patents, and has been involved in multiple major research projects, including those funded by the National Natural Science Foundation of China and the National Key R&D Program. He has also received top provincial and national awards for technological innovation. For more information, visit: https://unicorn.org.cn/valency/ |
Data analytics & machine learning: covering classical and modern algorithms for classification, clustering, graph learning, knowledge graph reasoning, multi-task learning, few-shot learning, as well as deep learning methods for structured and unstructured data.
Spatiotemporal / trajectory data mining: how to process, compress, analyze, and extract meaningful insights from trajectory, GPS / map, traffic, and mobility datasets under road network constraints or urban settings.
AI in medicine and medical imaging: exploring how computer vision and deep learning can assist in medical diagnosis and measurement — for example, automatic measurement of fetal head circumference, or computer-aided diagnosis of eye diseases.
Multi-task / multi-domain learning & domain adaptation: developing models that can work across tasks or domains — for example, transferring learned knowledge across domains, or building models that are robust under data imbalance and cross-domain shifts.
Privacy-aware, federated, and distributed learning: designing frameworks that enable learning under privacy constraints, dealing with real-world issues like data distribution imbalance, and enabling collaborative learning in distributed environments.
Smart city / transportation / urban computing: applying data mining, prediction, and AI to problems like traffic forecasting, map-matching, mobility analysis, and urban mobility planning — areas which relate to my interest in spatiotemporal machine learning and smart-city applications. |
I have authored or co-authored more than 50 papers (spanning conferences and journals), covering a wide spectrum of areas mentioned above. Some recent representative works:
A multi-task stance detection model published in a top journal: “MG-SIN: Multigraph Sparse Interaction Network for Multitask Stance Detection” (2025, IEEE TNNLS) — a network that handles multiple stance detection tasks via sparse interactions among multiple graphs. (unicorn.org.cn)
“Dynamic Knowledge Path Learning for Few-Shot Learning” (2025, Big Data Mining and Analytics) — tackling few-shot generalization by dynamically learning knowledge paths to guide knowledge transfer. (unicorn.org.cn)
Contributions to spatial-temporal modeling for urban mobility: e.g. MUSE-Net, a model for traffic-flow forecasting (ICDE 2024). (unicorn.org.cn)
Research on federated learning under practical constraints: FedGR: Federated Learning with Gravitation Regulation for Double Imbalance Distribution (DASFAA 2023), addressing imbalance in distributed data settings. (unicorn.org.cn)
Fundamental work on map-matching, trajectory compression, and mobility data: over many years I studied how to compress trajectory data under road-network constraints, infer road types, detect and analyze how weather impacts transport, and more — all contributing to urban mobility and big location data analytics. (unicorn.org.cn)
|
