A New Feature Selection Technique Combined with ELM Feature Space for Text Classification

The aim of text classification is to classify the text documents into a set of pre-defined categories. But the complexity of natural languages, high dimensional feature space and low quality of feature selection become the main problem for text classification process. Hence, in order strengthen the classification technique, selection of important features, and consequently removing the unimportant ones is the need of the day. The Paper proposes an approach called Commonality-Rarity Score Computation (CRSC) for selecting top features of a corpus and highlights the importance of ML-ELM feature space in the domain of text classification. Experimental results on two benchmark datasets signify the prominence of the proposed approach compared to other established approaches.