Knowledge Graph Forum at DataFunCon – Speakers, Topics, and Registration Details

Knowledge graphs, as a key driver of artificial intelligence, have attracted extensive attention in both industry and academia. With the maturation of large‑scale knowledge graph construction driven by pre‑training technologies, they now deliver significant value across various sectors.

On December 18, from 9:00 to 12:45, the DataFunCon conference will host a Knowledge Graph Forum organized by Meituan senior algorithm expert Wang Sirui. The forum will feature five speakers from Peking University, Southeast University, Baidu, Xiaomi, and Meituan, who will share the latest advances in large‑scale knowledge graph storage, general and domain‑specific graph construction, enterprise applications, and related challenges.

Speaker 1 – Wang Sirui (Meituan Senior Algorithm Expert) Topic: Large‑scale Knowledge Graph Storage and Applications. Audience Benefit: Understanding the fundamentals and practical uses of knowledge graphs.

Speaker 2 – Zou Lei (Peking University Professor) Topic: Knowledge Graphs and Graph Databases. Description: Introduction to knowledge graph and graph database concepts, overview of the gStore graph database system developed by his team, and discussion of research in RDF data management. Audience Benefit: Grasp basic concepts and applications of knowledge graphs and graph databases.

Speaker 3 – Qi Guilin (Southeast University Professor) Topic: Challenges and Technologies for Enterprise‑Level Knowledge Graph Construction. Description: Explores the difficulties of building domain‑specific knowledge graphs, especially in low‑resource scenarios such as automotive intelligent maintenance, and presents techniques for efficient construction using active and continual learning. Audience Benefit: Learn the latest techniques for efficient knowledge graph construction and discover new research directions.

Speaker 4 – Zhao Min (Baidu Researcher) Topic: "Jieyu" – Building a Chinese Knowledge Base from Zero Samples. Description: Introduces the open‑source "Jieyu" framework, which provides a full‑lexicon Chinese knowledge base, annotation tools, and pipelines for text mining and model improvement. Audience Benefit: Acquire practical steps to create a custom knowledge base, including taxonomy construction, annotation, mining, and sample optimization.

Speaker 5 – Dai Wen (Xiaomi Senior Algorithm Engineer) Topic: Graph‑Based QA in the XiaoAi Assistant – Practice and Outlook. Description: Presents the overall framework and key technologies of graph‑based question answering, shares the champion solution from CCKS2021, and discusses event extraction techniques and future directions. Audience Benefit: Understand the system architecture, key techniques, and practical experiences of deploying graph QA in a commercial voice assistant.

Speaker 6 – Zhang Hongzhi (Meituan Algorithm Expert) Topic: Meituan Brain – Food‑Domain Knowledge Graph Construction and Applications. Description: Details the construction pipeline using massive review data, multi‑modal pre‑training for dish graphs, and the deployment of the resulting graph across Meituan’s food‑related services. Audience Benefit: Gain practical experience in building and applying a domain‑specific knowledge graph from zero to production.

The forum will be livestreamed for free, and interested participants can register online to join the event.