Research on Improved K-Means Clustering Algorithm
2011 ◽
Vol 403-408
◽
pp. 1977-1980
Keyword(s):
The traditional K-means clustering algorithm prematurely plunges into a local optimum because of sensitive selection of the initial cluster center. Hierarchical clustering algorithm can be used to generate the initial cluster center of K-means clustering algorithm. The geometric features of input data can achieve a good distribution by means of pretreatment and feature extraction and selection. In the learning of fuzzy neural network, Java language is used to write source code of the algorithm. The experimental results show that new algorithm has improved the clustering quality effectively.
2014 ◽
Vol 701-702
◽
pp. 88-93
◽
Keyword(s):
Keyword(s):
2010 ◽
Vol 29-32
◽
pp. 802-808
Keyword(s):
2010 ◽
Vol 108-111
◽
pp. 106-111
◽
Keyword(s):
2014 ◽
Vol 998-999
◽
pp. 873-877
2014 ◽
Vol 31
(8)
◽
pp. 1661-1667
◽
Keyword(s):