DocumentCode :
424326
Title :
The study on immune spatial clustering model based on obstacle
Author :
Yang, Hai-dong ; Deng, Fei-qi
Author_Institution :
Inst. of Autom., South China Univ. of Tech., Guangzhou, China
Volume :
2
fYear :
2004
fDate :
26-29 Aug. 2004
Firstpage :
1214
Abstract :
Spatial clustering methods are mainly to group spatial objects based on their characteristics such as distance, connectivity, or their relative density in space. In the real world, many physical obstacles exist such as rivers, lakes and highways, and their presence may affect the results of clustering substantially. The problem of clustering in the presence of obstacles was studied and defined. As a solution to this problem and based on K-medoids, a scalable new clustering algorithm, called immune spatial clustering model based on obstacle was proposed. Various forms of pre-processed information that could enhance the efficiency of immune spatial clustering model were discussed. Various test data show that immune spatial clustering model is both efficient and effective.
Keywords :
pattern clustering; spatial data structures; K-medoids; immune spatial clustering model; Automated highways; Automation; Clustering algorithms; Clustering methods; Euclidean distance; Iterative algorithms; Lakes; Partitioning algorithms; Rivers; Stochastic processes;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Machine Learning and Cybernetics, 2004. Proceedings of 2004 International Conference on
Print_ISBN :
0-7803-8403-2
Type :
conf
DOI :
10.1109/ICMLC.2004.1382376
Filename :
1382376
Link To Document :
بازگشت