【海韵讲座】2024年第41期- Machine Learning-Based Spatio-Temporal Data Management and Pattern Discovery
报告题目：Machine Learning-Based Spatio-Temporal Data Management and Pattern Discovery
主讲人：Jianzhong Qi (Associate Professor, The University of Melbourne)
报告时间：2024年9月3日（周二）10:30-11:30
报告地点：厦门大学翔安校区bwin必赢6号楼209会议室
报告摘要：
Spatio-temporal data (i.e., data with information in location and time) is being generated at an unprecedented scale due to the prevalence of mobile devices, the Internet of Things, and 5G networks. Traditional retrieval methods become relatively expensive and difficult to use for data of such large scale and level of complexity, creating significant overheads in energy costs and human efforts. This talk presents our team’s recent research on machine learning-based techniques to achieve high query efficiency and accessibility for spatio-temporal data. The talk will cover machine learning-based methods to (1) optimise database indices and improve the efficiency of retrieving spatial data records, (2) learn representations for spatio-temporal data, to enable automatic feature extraction, and (3) discover patterns from such data, for downstream analytical tasks such as traffic forecasting.
报告人简介：
Jianzhong Qi is an Associate Professor in the School of Computing and Information Systems, The University of Melbourne, Australia. His general research area is data management, and his research concerns fundamental algorithms for spatial, temporal, and geo-textual data, including data indexing, query and update processing, and machine learning. He publishes in top database and machine learning venues such as SIGMOD, VLDB, ICML, NeurIPS, TODS, and TPAMI. He has won several awards including the prestigious Future Fellowship of the Australian Research Council.
邀请人：计算机科学与技术系 范晓亮高级工程师、王程教授