Augmentation of a term/document matrix with part-of-speech tags to improve accuracy of latent semantic analysis
Tom Rishel, A. Louise Perkins, Sumanth Yenduri, Farnaz Zand, S. S. Iyengar · 2006
Abstract:- We consider the improvement in accuracy of latent semantic analysis when a part of speech tagger is used to augment a term/document matrix. We first construct an augmented term/document matrix as input into singular value decomposition (SVD). The singular values then serve as principal components for a cosine projection. The results show that the addition of POS tags can decrease ambiguities significantly.