11th February 2016 at 11:00AM
Abstract: Nowadays, an ever increasing number of digital traces describing urban mobility is generated on a daily basis: check-ins and geo-tagged messages on social networks, trajectories collected from smartphones and GPS devices, etc. Although these devices were not initially designed for the analysis of mobility, there usefulness is obvious. In the context of public transportation, passenger traces depicted by ticketing data collected through automated fare collection (AFC) systems can be leveraged not only to measure the quality of service but also to understand passenger behavior, characterize the travel demand, and adapt the transport offering accordingly. This seminar will focus on research works carried out by "Mobility and Data" team at Ifsttar. Two generative models to clustering smartcard data will be considered from two complementary standpoints: a station-oriented, operational point of view and a passenger-focused one. The first approach clusters stations based on when their activity occurs (i.e. how trips made at the stations are distributed over time). The second approach makes it possible to identify groups of passengers that have similar boarding times. By applying our approaches to a real dataset issued from the metropolitan area of Rennes (France) we illustrate how they can help reveal valuable insights about urban mobility.