Colloquium: Simulating Geo-Social Society through Geospatial Data Lifecycle
Abstract: Data has been the force driving many areas’ research in the data science era. Yet, there exist diverse factors to hinder from exploiting such data, including low quality of data (e.g., missing, noise, sparsity, etc.) due to data collection cost and privacy concern. Location-based social network (LBSN) data are one of the cases. While a company that has data ownership of LBSN access all information privately, others can utilize publicly accessible data only. Researchers that rely on publicly available LBSN, such as LBSN analysis, location recommendation, social link prediction, community detection, and social influence, encounter problems caused by lack or the quality of data.
In this talk, I will discuss the geospatial data lifecycle (i.e., acquisition, management, sharing, and analysis), with spatial extents and three geospatial types: maps, location, and geo-enriched data. I will share my experience to contribute to the geospatial data ecosystem to flourish indoor GIS. Unlike the outdoor environment, where sensors (such as GPS, satellite) covering broad areas have collected map data and location data thanks to openness, indoor geospatial data require more effort to cover indoor spaces in terms of collection. On the one hand, data has been acquired by sensors or has been post-processed by authoring and converting tools. On the other hand, synthetic data has been generated by plausible simulation.
I will introduce generating geo-enriched LSBN data that provide ground truths based on plausible geo-social simulation. This process eventually leads us to understanding our complex society, predicting uncertain future events, and prescribe a remedy via experiencing “What-If” scenarios. I will exhibit examples of geospatial simulations, including epidemic simulations (e.g., influenza) and ride-haling (e.g., Uber, Lyft).
Bio: Joon-Seok Kim is a Postdoctoral Research Fellow at the Department of Geography and Geoinformation Science, George Mason University. He received his Ph.D. in Computer Science from Pusan National University. During his Ph.D., he participated in the MediaQ project at the Integrated Media System Center, the University of Southern California. He worked in the United Nations Global Service Centre as a Geospatial Expert. His research topics span agent-based modeling, spatiotemporal databases, urban mobility analysis, location privacy, and indoor/3D GIS. He won the first place in ACM SPATIAL Cup 2019.
*To sign up for an appointment to meet with Joon-Seok Kim, please pay a visit to the following link: Appointment Sign Up Sheet.