Research Seminar by Dr. Meng-Fen Chiang | Urban Transportation Data Analytics

Large cities today are facing increasing challenges in managing high commuting demand using taxis and public transportation. A good understanding of urban mobility patterns is thus critical to the formulation of effective transportation policies. Specifically, we focus on three research topics by using the data that has been generated in cities, including identifying traffic congestion cascades, modelling taxi demands, and modeling lifestyles of users in a city. First, the knowledge of traffic health status is essential to the general public and urban traffic management. To identify congestion cascades, an important phenomenon of traffic health, we propose a Bus Trajectory based Congestion Identification (BTCI) framework that explores the anomalous traffc health status and structure properties of congestion cascades using bus trajectory data. Extensive evaluations on 11.8 million bus trajectory data show that BTCI is effective in high- lighting congested segments and identifying congestion cascades. Second, as the demand for taxis changes over space and time, we thus propose a Grid-based Gaussian Mixture Model (GGMM) to model spatio-temporal dynamics of taxi bookings to capture the demand for taxis. We demonstrated the effectiveness of our proposed model through anomaly detection and successfully verified the anomalies with real-world events. Lastly, we focus on the analysis of human mobility from bus and subway transaction records for modeling lifestyles of users in a city. We developed two models to learn the activity centers for passengers in Singapore. We also propose to automatically predict HOME,WORK and OTHERS labels of locations visited by each user. With the location labels assigned, we further derive interesting insights of urban lifestyles at both individual and population levels. Further, I shall continue to identify and work on important urban and human talent research problems using my past research experience in urban computing.

About the Speaker