Talk Abstract: Big data sources such as smartphones, mobile devices, social media and ubiquitous sensors allow us to collect data with details and coverage unimaginable before. These datasets show us the importance of considering dynamics in urban modeling. Transportation systems research can exploit these vast data sources to a great extent. The overarching goal of this work is to develop data analytics to understand urban dynamics and user behavior using geo-located social media data. Novel statistical estimation techniques are developed to understand the spatiotemporal patterns of urban activities based on more than half a million Foursquare check-ins of about 20,000 users from New York City. A novel method is developed to infer activity type, its duration and location, and the sequence of the activities from incomplete trajectory data. When aggregated these activity-location sequences indicate the travel demand within a region. The potential of geo-location data to derive dynamic traffic patterns is demonstrated by building an agent-based simulation model. Potential applications of these techniques to a number of urban systems science challenges are also outlined.
Dr. Meead Saberi, lecturer in transportation engineering, data guru, and urban scientist