Previous [ 1] [ 2] [ 3] [ 4] [ 5] [ 6] [ 7] [ 8] [ 9] [ 10] [ 11] [ 12] [ 13]


Journal of Information Science and Engineering, Vol. 32 No. 1, pp. 113-131 (January 2016)

Locating Traffic Hot Routes from Massive Taxi Tracks in Clusters

School of Computer Science
Beijing University of Technology
Beijing, 100124 P.R. China

The increasing availability of location-acquisition technologies has resulted in huge volumes of trajectories. The sheer volume of these data sets prevents their processing by traditional centralized technologies. In this paper, we propose a MapReduce-based extraction- and-group framework to locate traffic hot routes from taxis track. In the proposed framework, massive trajectory data are partitioned into data chunks so that they can be processed in parallel on multiple machines. Then the low speed parts from each trajectory are extracted by a speed based clustering. Finally, a MapReduce inner-function based grouping method is used to locate traffic hot routes from all low speed parts. Based on this extraction-and-group framework, we develop a traffic hot route locating algorithm. The algorithm was evaluated through experiments on real life data sets, and was shown to have considerable potential to promptly and accurately locate traffic hot routes from massive trajectory data through various analyses on the experimental results.

Keywords: trajectory clustering, traffic hot route analysis, map-reduce, cloud computing, DBSCAN

Full Text () Retrieve PDF document (201601_07.pdf)

Received August 22, 2014; revised December 26, 2014; accepted March 3, 2015.
Communicated by Meng Chang Chen.