Fast and Improved Feature subset selection Algorithm Based Clustering for High Dimensional Data

The Clustering is a method of grouping the information into modules or clusters. Their dimensionality increases usually with a tiny number of dimensions that are significant to definite clusters, but data in the unrelated dimensions may produce much noise and wrap the actual clusters to be exposed. Attribute subset selection method is frequently used for data reduction through removing unrelated and redundant dimensions (or attribute). The Ant colony optimization technique is used for solving computational problems which can be reduced to find good path during graphs in the minimum spanning tree problem and traveling salesman problem. In my paper the feature subset selection algorithm and Ant colony optimization algorithm are employed to improve the feature subset selection. The proposed method helps us in improved feature subset selection algorithm based on hierarchical cluster and it’s minimizes redundant data set and improves the attribute subset accuracy. Keywords— Hierarchical clustering based algorithm-filter technique,Graph-based cluster.