The particular focus is on techniques that create and use Web-based corpora of "comparable" sentences and text chunks for estimating word and phrase translation probabilities, and on techniques that derive relationships from "context vectors" that represent word and phrase meanings.Part of the project will also upgrade Trevor's work on TupleFlow to work with Hadoop.
Thursday, April 23
NSF Clue Award for Mining Semantic Word Relationships
Google congratulated the projects that were awarded 2009 CLuE grants that includes access to the Google/IBM cluster. Our lab received a grant to work on mining word relationships from large corpora.