Database Seminar

Friday, October 14, 2016 - 10:30am to 11:30am

Location:

Traffic21 Classroom 6501 Gates & Hillman Centers

Speaker:

JESSE LI, Assistant Professor https://faculty.ist.psu.edu/jessieli/Site/index.html

How could we harness the increasingly available big data to understand our dynamic ecosystem? For example, why people or animals move in the space in certain ways and how do their movements respond to surrounding environments? Why are crimes more frequent in certain regions and can we explain it using heterogeneous urban data? Is shale gas development contaminating our environment and how to mine the correlations between environment and all potential factors? Our research aims to develop data mining techniques for geospatial data collected from different sources to semantically understand trajectories, urban dynamics, and environment, by closely collaborating with domain experts. In this talk, I will focus on data mining techniques to understand spatial trajectories. I will first discuss why existing methods often make trivial discoveries when contexts are not considered. I will then present our recent results in semantic understanding of trajectories with rich spatial-temporal contexts. I will also show that using cross-domain big data is critical to understand crimes and environment. Throughout the talk, I would like to share my experiences in exciting interdisciplinary collaborations. About the Speaker

Event Website:

http://db.cs.cmu.edu/events/jessie-li-penn-state

For More Information, Contact:

hyunahs@cs.cmu.edu

Keywords:

Seminar Series