Table of Content

Open Access iconOpen Access

ARTICLE

crossmark

Tensor-Based User Trajectory Mining

Chen Yu, Qinmin Hong, Dezhong Yao, Hai Jin

Cluster and Grid Computing Lab
Big Data Technology and System Lab
Services Computing Technology and System Lab
School of Computer Science and Technology
Huazhong University of Science and Technology, Wuhan, 430074, China

* Corresponding Author: E-mail: email

Computer Systems Science and Engineering 2018, 33(2), 87-94. https://doi.org/10.32604/csse.2018.33.087

Abstract

The rapid expansion of GPS-embedded devices has showed the emerging new look of location-based services, enabling such offerings as travel guide services and location-based social networks. One consequence is the accumulation of a rich supply of GPS trajectories, indicating individuals’ historical position. Based on these data, we aim to mine the hot route by using a collaborative tensor calculation method. We present an efficient trajectory data processing model for mining the hot route. In this paper, we rst model the individual’s trajectory log, extract sources and destinations, use map matching to get the corresponding road segments, and nally apply the source-destination-road segments tensor in order to compute the recommended hot route. To prove the validity and efficiency of the method, we conduct a collaborative route recommendation system, and the experimental result indicated that the solution can recommend a route with considerable accuracy.

Keywords


Cite This Article

C. Yu, Q. Hong, D. Yao and H. Jin, "Tensor-based user trajectory mining," Computer Systems Science and Engineering, vol. 33, no.2, pp. 87–94, 2018.



cc This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 1260

    View

  • 961

    Download

  • 0

    Like

Share Link