Short and Long-term Pattern Discovery Over Large-Scale Geo-Spatiotemporal Data

Published in The 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (SIGKDD 2019). (Anchorage, AK), 2019

This paper is a new framework to discovery short and long-term patterns over geo-spatiotemporal data (e.g. traffic, weather, etc.). Short-term patterns show propagation of geo-spatiotemporal entities which cause other entities to happen (e.g. rain –> accident –> traffic jam). Long-term patterns show the influence of a temporally long-term entity on its spatiotemporal neighborhood (e.g. major construction –> more traffic issue).

Download here

Characterizing driving context from driver behavior

Published in 25th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (Los Angeles, CA), 2017

A new framework to characterize driving context based on driving behavior as prerequisite for driving risk prediction.

Download here