Elective Ensembling Methods (Eem) On Classification Using Data Anonymization For Health Care Data

Enhancing the classification performance mostly used ensemble classification techniques. Research studies shows that classification through ensembling techniques shows the good classification concert in dynamic model representation in data anonymization approach. This paper we propose a elective ensembling methods based on the dynamic model data anonymization (EEM-DM-DA). This proposed technique enable to understanding the numerous trials met in privacy preserving data mining and also support us to discover best appropriate technique for numerous data modification techniques. The proposed anonymization technique can simultaneously disturb attributes presenting in the elected dataset. This can increase the diversity among different classifiers. Tentative stage of EEM-DM-DA is compared with the existing ensemble methods on maximum UCI data sets, where the SVM classification algorithm is used to train the ensemble classifiers. Proposed EEM-DM-DA technique results provides competitive solution for elective ensemble Method.