The International Journal of Innovative Research in Computer and Communication Engineering
Feature reduction of pattern dimensionality using feature extraction and feature selection belongs to the data mining. To enhance the robustness of the k-means clustering algorithm and for visualization purpose the dimension reduction techniques may be employed. Randomized dimensionality reduction is the transformation of high-dimensional data into a significant illustration of reduced dimensionality that corresponds to the fundamental dimensionality of the data. K-means clustering algorithm often not well for high dimension datasets and error dimensionality reduction, hence, to improve the efficiency, the proposed system apply roughset theory based k-means on original data set and obtain a reduced dataset containing possibly uncorrelated variables.