Increasing cluster uniqueness in Fuzzy C-Means through affinity measure

摘要

Clustering is a widely used technique in data mining application for discovering patterns in large dataset. In this paper the Fuzzy C-Means algorithm is analyzed and found that quality of the resultant cluster is based on the initial seed where it is selected either sequentially or randomly. Fuzzy C-Means uses K-Means clustering approach for the initial operation of clustering and then degree of membership is calculated. Fuzzy C-Means is very similar to the K-Means algorithm and hence in this paper K-Means is outlined and proved how the drawback of K-Means algorithm is rectified through UCAM (Unique Clustering with Affinity Measure) clustering algorithm and then UCAM is refined to give a new view namely Fuzzy-UCAM. Fuzzy C-Means algorithm should be initiated with the number of cluster C and initial seeds. For real time large database it's difficult to predict the number of cluster and initial seeds accurately. In order to overcome this drawback the current paper focused on developing the Fuzzy-UCAM algorithm for clustering without giving initial seed and number of clusters for Fuzzy C-Means. Unique clustering is obtained with the help of affinity measures.