Clustering and Hiding Sensitive Data for Social Network Dataset?

Data Mining is one of the main processing step in KDD (Knowledge Discovery in Database), which provides the potentially useful but unknown information to the user. The mining of the databases includes the preprocessing, removing duplication and irrelevant data from the database. For the mining process several techniques were evolves such as association, classification, clustering, feature selection etc. Each of the technique is specifically designed and in use. This proposed work concentrates on clustering technique. The clustering is made in sequential manner, in particular to provide privacy to the data. Hence, the proposed work with the sequential clustering technique, hide very sensitive information from the third party authorization. The application considered here is the Social Networks, in which the sequential clustering methodology is applied. The sensitive data that needs the privacy is computed here with the symbols. The results show that the proposed method remains better at its performance in authentication.