Avatar

Real-Time Sensing of Traffic Information in Twitter Messages

21 Setembro 2010
Sem comentários

S. Carvalho, L. Sarmento and R. Rossetti
4th Workshop on Artificial Transportation Systems and Simulation (ATSS @ IEEE ITSC 2010), September 19–22, 2010, Madeira, Portugal

Abstract:
This paper presents an initial attempt to use micro-blogging messages posted on Twitter (by users in transit) to perform real-time sensing of traffic-related information. We propose a text classification approach to the problem: we wish to automatically identify traffic-related messages posted on Twitter, among the millions of unrelated messages posted by users. Given that the fraction of relevant traffic messages on Twitter is extremely low (< 0.05%), the main challenges involved at this stage are (i) creating a suitable training set for setting up the classifier, and (ii) driving the classifier to a reasonable level of precision in identifying relevant messages. We opted for a dual-stage bootstrapping strategy for tackling both these problems simultaneously. First, we used short message reports that are automatically posted on Twitter …



Sem comentários