Refinement of K-Means and Fuzzy C-Means

Clustering is widely used technique in data mining application for discovering patterns in large data set. In this paper the K-Means and Fuzzy C-Means algorithm is analyzed and found that quality of the resultant cluster is based on the initial seeds where it is selected either sequentially or randomly. For real time large database it's difficult to predict the number of cluster and initial seeds accurately. In order overcome this drawback the authors propose two new algorithms Unique Clustering through Affinity Measure (UCAM) and Fuzzy-UCAM clustering algorithm. Both UCAM and Fuzzy-UCAM clustering algorithms works without fixing initial seeds, number of resultant cluster to be obtained. Unique clustering is obtained with the help of affinity measures.