Using Relative Distance and Hausdorff Distance to Mine Trajectory Clusters
Along with development of location service and GPS technology, mining information from trajectory datasets becomes one of hottest research topic in data mining. How to efficiently mine the clusters from trajectories attract more and more researchers. In this paper, a new framework of trajectory clus...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | EJournal Article |
| Published: |
Institute of Advanced Engineering and Science,
2013-01-10.
|
| Subjects: | |
| Online Access: | Get fulltext |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Along with development of location service and GPS technology, mining information from trajectory datasets becomes one of hottest research topic in data mining. How to efficiently mine the clusters from trajectories attract more and more researchers. In this paper, a new framework of trajectory clustering, called Trajectory Clustering based Improved Minimum Hausdorff Distance under Translation (TraClustMHD) is proposed. In this framework, the distance between two trajectory segments based on local and relative distance is defined. And then, traditional clusters algorithm is employed to mine the clusters of trajectory segment. In additional, R-Tree is employed to improve the efficiency. The experimental results showed that our algorithm better than existing others which are based on Hausdorff distance and based on line Hausdorff distance. DOI: http://dx.doi.org/10.11591/telkomnika.v11i1.1877 |
|---|