GEometrical Models of Natural Language Semantics


Patrick Pantel
"Lexical Semantics in the Prime Time: Applications to Web Search"

Similarity modeling is a key task in computational lexical semantics for finding word senses, concepts, paraphrases, topics, and distributional synonyms, just to name a few. In this talk, we present a flexible Map/Reduce infrastructure for very large-scale unsupervised and semi-supervised learning and show its application to several web search tasks such as query intent modeling, ad matching, and automatic set expansion using corpus statistic from a large web crawl.

Patrick Pantel is currently a Senior Scientist at Yahoo! Labs and a Research Assistant Professor and Research Scientist in the Natural Language Group at the USC Information Sciences Institute, where he conducts research in large-scale natural language processing, text mining, knowledge acquisition, and predictive systems. In 2003, he received a Ph.D. in Computing Science from the University of Alberta in Edmonton, Canada.