Revolutionise Ontology Engineering with Deep Learning and Language Models学术报告

发布者:曹玲玲发布时间:2023-08-09浏览次数:35

应吴天星老师邀请,823日(周三)上午9:30-12:00,英国曼彻斯特大学 Jiaoyan Chen 教授,给本实验室师生做学术报告并展开交流研讨,具体情况如下:

 

报告题目:Revolutionise Ontology Engineering with Deep Learning and Language Models

报告人:英国曼彻斯特大学 Jiaoyan Chen 教授

报告时间:2023823日(周三)上午9:30-12:00 

报告地点:东南大学九龙湖校区计算机楼413会议室

 

报告摘要:Ontology engineering is critical in AI but heavily relies on human efforts. With the development of deep learning especially large language models in recent years, it becomes a promising direction to utilize these techniques to automate ontology engineering, during which new and general AI techniques on neural-symbolic integration with parametric and none-parametric knowledge could be developed. In this talk, Dr. Jiaoyan Chen will present a series of works using language models for address ontology engineering tasks including ontology alignment and ontology completion, as well as a work using ontological knowledge for benchmarking pre-trained language models. All these works are implemented as opensource APIs in a Python package named DeepOnto.

 

报告人简介:Dr. Jiaoyan Chen is a tenured Lecturer (Assistant Professor) in Department of Computer Science, University of Manchester, and a part-time Senior Researcher in Department of Computer Science, University of Oxford. Before that, Dr. Chen was a full-time Senior Researcher at Department of Computer Science, University of Oxford, reporting to Prof. Ian Horrocks, and got his Ph.D. in Computer Science and Technology in Zhejiang University in 2016, supervised by Prof. Huajun Chen. Dr. Chen mainly works on Knowledge Graph, Ontology, Machine Learning and Neural-symbolic AI. He has published over 60 papers in top computer science conferences and journals such as AAAI, IJCAI, ACL, KDD, WWW, Proceedings of the IEEE, etc., and acts as PC or Senior PC of several conferences such as AAAI, IJCAI, ISWC and CIKM, as well as Senior Editor of Transactions of Graph Data and Knowledge (TGDK) and Editor of Data Intelligence.  Homepage: https://chenjiaoyan.github.io/.