Feature Subset Selection Algorithm for Clustering Using Modified Prim?s Algorithm

In the proposed framework the feature selection subset is mainly used for machine classification and data mining. It is used for the high dimensional data reduction. This data reduction and best feature selection are handled through removal of irrelevant data?s in the dataset. Best feature subset selections are obtained through modified prim?s algorithm. In prim?s algorithm although there is continuous developing forest of connected nodes but initially root node is chosen randomly which may or may not be the part of minimum edge of the graph. But proposed modified prim?s algorithm, along with continuous developing connected spanning tree, root node is also forming minimum edge of the graph. The implementation analyzes the results through 10 different bench mark datasets and two classification algorithms.