- This event has passed.
Seminar Reboan, in collaboration with the APTIKOM Lecturer Competency Division, presents an engaging topic with speaker Prof. Dr. Xue Li. 🤩✨
Multimodal Data Learning and its Applications
Speaker’s Brief Biography:
Xue Li holds a Bachelor of Computer Software Engineering, a Master of Knowledge Engineering and a PhD in Information Systems. Currently a tenured professor at the School of Electrical Engineering and Computer Science, The University of Queensland, Australia, he has supervised over 40 doctoral students to graduation.
Professor Li’s research centres on data science systems theory, multimodal data fusion and mining and big data privacy protection. He has long taught required courses in databases and cloud computing for Computer Science students. He has led over 20 projects including Australian Research Council (ARC) projects and published over 200 research papers in leading international journals and conferences. He founded the ADMA International Conference on Data Mining, published with Springer. His research team has won technology achievement awards from Google, Microsoft and the Australian government. In 2015, he was named one of the 50 most influential people in Australia by the Australian Financial Review.
Abstract:
Data-driven artificial intelligence research spans the entire data lifecycle, encompassing diverse data sources and formats. Multimodal data processing and learning present numerous challenges, including metadata modelling, data collection, quality management, and fusion. This talk delves into multimodal data collection analysis and processing within data engineering, highlighting our team’s contributions. As multimodal data learning becomes widespread, particularly with OpenAI and large language models (LLMs), AI applications face significant hurdles. Multimodal data processing demands innovative algorithm design and optimisation, as well as semantic understanding. Addressing these challenges involves sequence alignment of semantic units across modalities, handling many-to-many relationships, and managing feature engineering complexity and consistency at various granularities. To tackle these issues, we’ve developed optimisation strategies and algorithmic frameworks like multi-granularity, multi-level, and multi-time interval feature alignment, knowledge graph relationship improvement, and comparative learning using tabular data. Starting with data representation, the talk introduces the problem and explores methods for synthetic and simulated data in current data engineering.
Moderator: Prof. Heru Suhartanto, Ph.D.
Join the discussion with Prof. Dr. Xue Li at Seminar Reboan 2026 Episode 8, which will be held on:
📅 Day/Date: Wednesday, 24 June, 2026
⏰ Time: 13:00 – 15:00 WIB
Venue:
📌 On Site: Ruang Sidang (A.409), 4th Floor, New Building, Fasilkom UI, Depok
— Limited to the first 40 participants
💻 Online: The link will be provided after completing the confirmation form.
📝 Confirmation Link:
Thank you for your participation. We look forward to seeing you at Seminar Reboan Episode 8.