Speaker: Yiqun Xie, Ph.D. candidate in Computer Science at the University of Minnesota – Twin Cities
Date: Thursday, March 5
Time: 10-11am
Location: HBL 1947 room
Title: Spatial Data Science: Challenges and New Techniques
Abstract: Our world has been witnessing a revolution brought by spatial technologies (e.g., Google Maps, Waze, Uber, Lyft, Lime, autonomous driving). According to a McKinsey Global Institute report, location data will generate about $600 billion annual revenue by 2020 with applications in energy, health, retail, etc. However, one size data science does not fit for all. Spatial data pose unique challenges to traditional data science techniques when applied to important societal problems. First, many spatial applications (e.g., public safety, transportation) have very low tolerance to spurious patterns due to their high economic and social costs, which are often ignored by traditional pattern mining approaches. Second, spatial data possess intrinsic interdependencies, introducing new challenges to traditional learning and optimization techniques which often neglect such dependencies among data samples or decision variables. Third, there is very limited training data for many geospatial problems and expert knowledge is required to generate such data, limiting the use of deep learning techniques.
In this talk, I will discuss novel formulations and computational techniques to bridge these research gaps. Specifically, I will focus on: (1) Statistically robust spatial clustering; and (2) Spatial-dependency-aware optimization. I will conclude my talk with near-term visions (e.g., theory-guided spatial data science), as well as a long-term goal to grow the impact of spatial data science on societal priorities.
Bio: Yiqun Xie is a Ph.D. candidate in Computer Science at the University of Minnesota – Twin Cities, working with Dr. Shashi Shekhar. His research focuses on developing novel spatial data science techniques to address important societal problems. His work has received a Best Paper Award from SSTD’19 and a Best Vision Paper Award from ACM SIGSPATIAL’19, which was highlighted by the Great Innovative Ideas program of CRA’s Computing Community Consortium. More information about Yiqun is available through his website: https://www-users.cs.umn.edu/~xiexx347/.