Unsupervised feature selection technique based on harmony search algorithm for improving the text clustering

Abstract

The increasing amount of text information on the Internet web pages affects the clustering analysis. The text clustering is a favorable analysis technique used for partitioning a massive amount of information into clusters. Hence, the major problem that affects the text clustering technique is the presence uninformative and sparse features in text documents. The feature selection (FS) is an important unsupervised technique used to eliminate uninformative features to encourage the text clustering technique. Recently, the meta-heuristic algorithms are successfully applied to solve several optimization problems. In this paper, we proposed the harmony search (HS) algorithm to solve the feature selection problem (FSHSTC). The proposed method is used to enhance the text clustering (TC) technique by obtaining a new subset of informative or useful features. Experiments were applied using four benchmark text datasets. The results show that the proposed FSHSTC is improved the performance of the k-mean clustering algorithm measured by F-measure and Accuracy.