Interspeech'2005 - Eurospeech

Lisbon, Portugal
September 4-8, 2005

Generalized Hebbian Algorithm for Incremental Latent Semantic Analysis

Genevieve Gorrell, Brandyn Webb

Linköping University, Sweden

The Generalized Hebbian Algorithm is shown to be equivalent to Latent Semantic Analysis, and applicable to a range of LSA-style tasks. GHA is a learning algorithm which converges on an approximation of the eigen decomposition of an unseen frequency matrix given observations presented in sequence. Use of GHA allows very large datasets to be processed.

Full Paper

Bibliographic reference.  Gorrell, Genevieve / Webb, Brandyn (2005): "Generalized hebbian algorithm for incremental latent semantic analysis", In INTERSPEECH-2005, 1325-1328.