ON DEEP LEARNING FOR SPATIOTEMPORAL RASTER DATA MINING

PhD Qualifying Examination


Title: "ON DEEP LEARNING FOR SPATIOTEMPORAL RASTER DATA MINING"

by

Mr. Jiachuan WANG


Abstract:

Spatiotemporal data mining has received significant attention for decades. In
recent years, deep learning methods have proven to be effective in mining
spatiotemporal data, outperforming traditional methods for various tasks, such
as prediction and data enhancement. Although much effort has been devoted to
developing deep learning methods for spatiotemporal data mining and the area
has been hot for years, its popularity has increased even more due to the huge
demand for high accuracy in handling extremely large-scale data in the big data
era. Additionally, complex application scenarios require powerful method
designs that ensure both professionalism and generality. Therefore, a targeted
and comprehensive survey is needed to consider the vast amount of existing
work. In this survey, we focus on one class of data with a wide range of
application prospects, raster spatiotemporal data, and aim to provide an
intuitive taxonomy and comprehensive summary of various techniques by
introducing the mainstream applications and problems of raster spatiotemporal
data, and reviewing the state-of-the-art (SOTA) deep learning methods. We also
summarize the widely used datasets for evaluation.


Date:                   Friday, 15 September 2023

Time:                   2:00pm - 4:00pm

Venue:                  Room 5510
                        Lifts 25/26

Committee Members:      Prof. Lei Chen (Supervisor)
                        Prof. Xiaofang Zhou (Chairperson)
                        Dr. Wilfred Ng
                        Prof. Ke Yi


**** ALL are Welcome ****