Dimensionality Using Optimization Algorithm for High Dimensional Data Clustering

This paper present an efficient approach to a feature selection problem based on genetic algorithm for high dimensional data clustering. We proposed a new algorithm uses an inconsistency rate to evaluate the fitness of individuals in population which results in a significant reduction in terms of computational time needed to reach a subset of relevant features and finally the relevant features are clustered by clustering algorithm k-means algorithm. The performances of efficient approach of features selection are evaluated on several high dimensional dataset. The experiment result shows that the proposed algorithm is faster process than the existing algorithm for feature selection and increasing the predictive accuracy and the dimensionality is reduced