Clustering an un supervised technique of Data Mining is a process that places similar objects in same cluster based on some similarity measure ..In this paper we provide a new measure for calculation of similarity between two clusters for categorical attributes in a relational database .We have suggested a Fuzzy algorithm which considers inter similarity( which should be low) in addition to intra similarity(which should be high) for clustering categorical datasets. The experimental results have been validated using indexes like Precision, Recall, F -Measure .The accuracy has also been calculated.
展开▼
