Share Email Print

Proceedings Paper

Applying fuzzy clustering optimization algorithm to extracting traffic spatial pattern
Author(s): Chunchun Hu; Wenzhong Shi; Lingkui Meng; Min Liu
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Traditional analytical methods for traffic information can't meet to need of intelligent traffic system. Mining value-add information can deal with more traffic problems. The paper exploits a new clustering optimization algorithm to extract useful spatial clustered pattern for predicting long-term traffic flow from macroscopic view. Considering the sensitivity of initial parameters and easy falling into local extreme in FCM algorithm, the new algorithm applies Particle Swarm Optimization method, which can discovery the globe optimal result, to the FCM algorithm. And the algorithm exploits the union of the clustering validity index and objective function of the FCM algorithm as the fitness function of the PSO algorithm. The experimental result indicates that it is effective and efficient. For fuzzy clustering of road traffic data, it can produce useful spatial clustered pattern. And the clustered centers represent the locations which have heavy traffic flow. Moreover, the parameters of the patterns can provide intelligent traffic system with assistant decision support.

Paper Details

Date Published: 15 October 2009
PDF: 6 pages
Proc. SPIE 7492, International Symposium on Spatial Analysis, Spatial-Temporal Data Modeling, and Data Mining, 74921T (15 October 2009); doi: 10.1117/12.838628
Show Author Affiliations
Chunchun Hu, Wuhan Univ. (China)
Wenzhong Shi, Hong Kong Polytechnic Univ. (Hong Kong, China)
Lingkui Meng, Wuhan Univ. (China)
Min Liu, Wuhan Univ. (China)

Published in SPIE Proceedings Vol. 7492:
International Symposium on Spatial Analysis, Spatial-Temporal Data Modeling, and Data Mining
Yaolin Liu; Xinming Tang, Editor(s)

© SPIE. Terms of Use
Back to Top