Clustering based - A FAST algorithm on high dimensional data

Abstract

The rapid advance of computer technologies in data processing, collection and storage has provided unparalleled opportunities to expand capabilities in production, services communication and research. However, a feature selection algorithm may be evaluated from both the efficiency and effectiveness points of view. It finds the subset of features. There are two steps of FAST algorithm. First, using graph theoretic method features are divided into clusters. Second the features which are highly related to target class are selected. We are comparing FAST algorithm with the some representative feature subset selection algorithm name as Fast correlation based filter, Relief-F, Correlation based feature selection, Consist and FOCUS-SF. The results are available on high-dimensional data, microarray, text data and image data. Experimental results show that our FAST algorithm implementation can run faster and obtain better-extracted features than other methods.