Абстрактный

MO & ESC - The New Approach in Impurity Centroids

G.Rajasekar, R.Vijayakumar, T.Aravind

Clustering is used to determine the intrinsic grouping in a set of unlabeled data. Subspace clustering is one type of clustering model that solves many normal clustering problems. The both novel subspace clustering algorithms known as fixed and optimal centroids are allows getting more profitable objects in database. But this also provides information with impurity data. So we proposed a new approach called Multi Objective and Evolutionary Subspace Clustering (MO&ESC) that provides statistic of the dimensions and the impurity measure within each cluster. This technique is used to provide Centroid-based Actionable 3D Subspace clusters and also returns the information based on impurity dimensional data value.