SCALE@NTU Invited Talk: Spatiotemporal Representation Learning for Urban Data Analytics

This research seminar is organized by Singtel Cognitive and Artificial Intelligence Lab for Enterprises (SCALE@NTU). Please find below the registration information:

To attend the seminar physically (limited seats available), click here

To attend the seminar on Teams, click here

Abstract: The ubiquity of geo-positioning technologies has accelerated the digitalization of urban cities, capturing not only static data such as points of interest but also the dynamic mobility of vehicles and individuals. With such details of urban cities, recent research has focused on generating a better understanding of the spatial and temporal context of geographical elements, including points of interest, trajectories and regions.

This presentation will start with an introductory overview of our research in geospatial data. Subsequently, I will present our recent studies of representation learning techniques for encoding the spatiotemporal context and mobility patterns. Our approaches are instrumental in addressing various research problems, including traffic prediction, location-based time series modelling and location recommendations.

Speaker: Dr. Kaiqi Zhao is a Senior Lecturer (British university system) in the School of Computer Science, the University of Auckland, New Zealand. He received his Ph.D. in Computer Science from Nanyang Technological University in 2018. Before he joined the University of Auckland, he was a postdoctoral research fellow at the Singtel Cognitive and Artificial Intelligence Lab for Enterprise at NTU, Singapore. His research focuses on geospatial data mining, representation learning and probabilistic modelling. His research work has been published at top-ranked conferences and journals in database and data mining, including KDD, SIGMOD, VLDB, ICDE, SIGIR, TKDE, etc. He also serves as a programme committee member of international conferences.