Kewen Wang is a Professor in the School of Information and Communication Technology at Griffith University. He has been actively working in computational logic (a.k.a. knowledge representation) and its applications in artificial intelligence (AI) and programming for over 30 years. The results of his studies led to the introduction of powerful new logics, computer languages and systems for knowledge representation and reasoning. He has published over 100 papers in high quality journals and conferences in his area, such as Artificial Intelligence, Journal of Artificial Intelligence Research, ACM Transactions on Computational Logic, AAAI, IJCAI and KR, including over 35 CORE A* and 30 CORE A conference/journal papers. His research results have been widely cited by active/leading researchers in his area. He was awarded six ARC Grants (5 Discoveries and 1 Linkage) as well as some other small grants.

He is regularly a Senior Program Committee Member (SPC) and Program Committee Member (PC) of major conferences in his areas including AAAI, IJCAI, TheWebConf (WWW) and KR. Especially, he was/is an Area Chair of AAAI-2022 and AAAI-2024 (AAAI is a premier conference in AI). He is also an Area Editor/Associator Editor for Journal of Web Semantics and Associate Editor for Transactions on Graph Data and Knowledge (TGDK).

In the past few years, he has been actively working in Knowledge Graphs and Explainable AI. More recently, with his team he has achieved several important research results and state-of-the-art prototype systems:

• RLvLR: the first embedding-based rule learner for KGs that is scalable to large KG such as Yago, Wikidata, Wikipedia and Freebase. We are also among the first to advocate the application of rule-based systems in link prediction [IJCAI-2018, TDKE-2021].
• TyRuLe: a state-of-the-art rule learner, which can learn typed rules over KGs and is scalable for large KGs (most scalable rule learners cannot learn such logic rules) [KR-2022].
• Drewer: query engine for existential rules (Datalog+-) that outperforms major baselines [IJCAI-2020, TKDE-2021, ICLP-2022].
• ALBERT+CCR: a question answering model for CommonsenseQA. Our model outperforms other comparable models for CommonsenseQA.