Enhancing Multilingual Latent Semantic Analysis with Term Alignment Information

Latent Semantic Analysis (LSA) is based on the Singular Value Decomposition (SVD) of a term-by-document matrix for identifying relationships among terms and documents from cooccurrence patterns. Among the multiple ways of computing the SVD of a rectangular matrix X, one approach is to compute the eigenvalue decomposition (EVD) of a square 2

• Type: Conference
• Year: 2008
• Where: COLING
• Authors: Brett W. Bader, Peter A. Chew
• Keywords: COLING 2008, Computational Linguistics, Diagonal Blocks, Matrix, Singular Value Decomposition