Friday, May 6, 3:00 p.m.
Center for Social Complexity Suite
3rd Floor, Research Hall
Matthias Renz, Associate Professor
Department of Computational and Data Sciences
George Mason University
Towards Reliable Spatial And Spatio-Temporal Pattern Analysis
ABSTRACT: Current technology trends, such as smart phones, mobile devices, stationary sensors and satellites, coupled with a new user mentality of voluntarily sharing information generates a huge volume of geo-spatial and geo-temporal data that might be useful for many applications. Indeed, the increasing volume of geo-spatial data from heterogeneous sources is an example of Big Data. In this talk I will present effective and efficient solutions to problems related to reliable spatial pattern analysis and mining of uncertain spatial and spatio-temporal data. These techniques can substantially improve the quality of decision making, minimize risk, and unearth valuable insights from data that would otherwise remain hidden. Use of uncertain data presents a two-fold challenge: 1) identifying potential solutions and assigning a probability to each solution such that the user is confident about the results, and 2) enabling fast computations such that the user obtains results in a reasonable time, even for large data sets.