Two-level text classification using hybrid machine learning techniques

Subspace detection and processing is receivingmore attention nowadays as a method to speed up searchand reduce processing overload. Subspace Learningalgorithms try to detect low dimensional subspaces in thedata which minimize the intra-class separation whilemaximizing the inter-class separation. In this paper wepresent a novel technique using the maximum significancevalue to detect a semantic subspace. We further modify thedocument vector using conditional significance to representthe subspace. This enhances the distinction between classeswithin the subspace. We compare our method againstTFIDF with PCA and show that it consistently outperformsthe baseline with a large margin when tested with a widevariety of learning algorithms. Our results show that thecombination of subspace detection and conditionalsignificance vectors improves subspace learning.