Title :
An Improved WaveCluster Algorithm Based on ICA
Author :
Li, Xiaoli ; Luo, Min
Author_Institution :
Modern Educ. Technol. Center, Nantong Univ., Nantong, China
Abstract :
In this paper, we introduce an improved WaveCluster algorithm, which can deal with the high-dimensional data issue (IWCA algorithm for short). The algorithm combines the advantages of ICA algorithm that can reduce the dimension of database. By this means, we can not only cluster the high-dimensional data, but also reduce the time complexities. Finally, the experimental results on KDD Cup 1999 data sets show that IWCA algorithm can efficiently find clusters in high-dimensional data space.
Keywords :
computational complexity; independent component analysis; pattern clustering; ICA algorithm; KDD Cup 1999 data; database; high-dimensional data space; improved WaveCluster algorithm; independent component analysis; time complexity; Algorithm design and analysis; Clustering algorithms; Covariance matrix; Databases; Educational technology; Feature extraction; Independent component analysis; Partitioning algorithms; Principal component analysis; Space technology;
Conference_Titel :
Wireless Communications, Networking and Mobile Computing, 2009. WiCom '09. 5th International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-3692-7
Electronic_ISBN :
978-1-4244-3693-4
DOI :
10.1109/WICOM.2009.5302493
