Feature Selection Technique for Text Document Classification: An Alternative Approach

Abstract

Text classification and feature selection plays an important role for correctly identifying the documents into particular category, due to the explosive growth of the textual information from the electronic digital documents as well as world wide web. In the text mining present challenge is to select important or relevant feature from large and vast amount of features in the data set. The aim of this paper is to improve the feature selection method for text document classification in machine learning. In machine learning the training set is generated for testing the documents. This can be achieved by selecting important new term i.e. weights of term in text document to improve both classification with relevance to accuracy and performance.