ICASSP
2009
IEEE
15 years 4 months ago
2009
IEEE
The major limitation in bilingual latent semantic analysis (bLSA) is the requirement of parallel training corpora. Motivated by semi-supervised learning, we propose a clusterbased...