In this paper, we consider a vector space model based information retrieval system using a principal component analysis.
Latent Semantic Indexing (LSI) is based on the singular value decomposition to the document by term matrix and is used as the method of dimension
reduction. We introduce an information retrieval method using the principal component analysis to the keyword covariance matrix and compare
this method with LSI.
By: Hironori TAKEUCHI, Mei KOBAYASHI, Masaki AONO, Hikaru SAMUKAWA
Published in: IPSJ Technical Report(FI), volume , (no ), pages in 2002
Please obtain a copy of this paper from your local library. IBM cannot distribute this paper externally.
Questions about this service can be mailed to reports@us.ibm.com .