Message Passing between Data Point on Clustering Algorithm for Gene Leukemia Dataset

D. Napoleon, G. Baskar

Abstract


Clustering (or cluster analysis ) aim to organize a collection of data item in to clusters, such that items within a cluster are more
“similar” to each other than they are to item in the other clusters. Affinity propagation (AP) is a clustering algorithm which has much better
performance than traditional clustering approach such as k-means algorithm. AP clustering handles large datasets by merging the exemplars
learned from subsets. The algorithm is tested on leukemia data set. The experimental results show that affinity propagation outperforms
clustering execution time and convergence rate.

 

Keywords: Data Mining, Clustering, k-means, x-means, Affinity propagation


Full Text:

PDF


DOI: https://doi.org/10.26483/ijarcs.v1i4.236

Refbacks

  • There are currently no refbacks.




Copyright (c) 2016 International Journal of Advanced Research in Computer Science