Mining Urban Traffic Condition from Crowd-Sourced Data

Traffic congestion is an inherent and hard issue to be tackled in huge urban areas, particularly in developing countries where transportation infrastructures have not been grown well to fulfill speedy developing request demands. This paper proposes novel solutions to these issues by devising mobile crowd-sourcing based approaches to traffic estimation. A framework for effective collecting, integrating and analyzing traffic-related data shared by mobile crowds has been devised. Besides, essential issues on predicting traffic conditions at streets where real-time data is missed are also resolved by applying data mining techniques to historical data. A prototype system has been developed to validate the proposed solutions. The experimental results show the feasibility and the effectiveness of the proposed methods revealing that they are ready to be applied in the practice.