We proudly presented our work on forecasting urban dynamics (daily people flow in a city) with mobility logs at The 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing (UbiComp 2015) held at Osaka, Japan September 2015.
We propose a low-rank bilinear model, which enables us to predict urban dynamics using external factors e.g. national holiday, day of week, and weather. Experimental results show that our model performs better predictive performance than existing models. In addition, we introduce various applications of our model such as event detection, and urban dynamics simulation.
9/10 Late Morning 11:30 – 12:40, Room 1+2
Forecasting Urban Dynamics with Mobility Logs by Bilinear Poisson Regression
Masamichi Shimosaka*, Keisuke Maeda**, Takeshi Tsukiji**, Kota Tsubouchi***
(*Tokyo Institute of Technology, **The University of Tokyo, ***Yahoo! JAPAN Research, Japan)
The (open accessible) paper is available on acm.org.