Graph Neural Network-Based Knowledge Graph Reasoning – Talk by Zhang Yongqi at DataFun Summit 2023

On March 18, 2023, from 09:00 to 17:00, the DataFun Summit 2023 will host an online forum titled “Unified Knowledge Representation and Complex Reasoning,” featuring Dr. Zhang Yongqi, an algorithm scientist at Fourth Paradigm, who will present the talk “Knowledge Graph Reasoning Based on Graph Neural Networks.”

Dr. Zhang joined Fourth Paradigm’s scientific research department in 2020 and focuses on machine‑learning and knowledge‑graph research. He has achieved internationally leading results in link prediction and entity alignment, secured first place in the large‑scale biological KG prediction task ogbl‑biokg, and authored nearly ten high‑impact papers in top AI conferences and journals. He earned his Ph.D. from Hong Kong University of Science and Technology in 2020 and his bachelor’s degree from Shanghai Jiao‑Tong University in 2015.

The presentation will explore how knowledge graphs, as a special form of structured data, drive downstream applications in domains such as search, healthcare, and finance, and why they are crucial for data mining and machine learning. It will review the evolution of graph neural network (GNN) techniques for KG reasoning—from information aggregation to subgraph learning—highlight key applications, and discuss the advantages and current bottlenecks of subgraph‑based methods.

Audience takeaways include: (1) the unique aspects of GNN modeling on knowledge graphs; (2) important real‑world applications of KG reasoning; and (3) the main challenges faced by GNN‑based KG inference.

Participants can join the live online session for free by scanning the QR code provided in the announcement.

About DataFun: Founded in 2017, DataFun focuses on big‑data and artificial‑intelligence technology sharing and exchange. It has organized more than 100 offline and 100 online salons, forums, and summits across major Chinese cities, involving over 2,000 experts and scholars, publishing over 900 original articles with millions of reads and a highly engaged follower base.