Gregory Grefenstette

Gregory Grefenstette is a French American computer scientist, researcher, professor, and author in the fields of artificial intelligence and natural language processing,[1][2] who currently serves as chief scientific officer at Biggerpan, a company developing a predictive contextual engine for the mobile web. Grefenstette is also a senior associate researcher at the Florida Institute for Human and Machine Cognition (IHMC).

Gregory Grefenstette
Born(1956-04-25)April 25, 1956
NationalityAmerican, French
Alma materUniversity of Pittsburgh (Ph.D., 1993)
Known forNatural language processing
Semantic maps
AwardsACM Multimedia Grand Challenge
Scientific career
FieldsComputer Science
Artificial Intelligence
Natural Language Processing & Understanding
Computational Linguistics
InstitutionsBiggerpan
IHMC
INRIA
Exalead
CEA
ThesisExploring Automatic Thesaurus Generation (1993)
Websiteweb.archive.org/web/20170705080047/https://www.lri.fr/~ggrefens/

Biography

Grefenstette was born in Pittsburgh, PA in 1956. He attended the Massachusetts Institute of Technology as an undergraduate student in mathematical science and completed his bachelor's degree at Stanford University in 1978. Grefenstette graduated with a master's degree from Paris-Sud University in 1983 and received his PhD in computer science from University of Pittsburgh in 1993. From 1984 to 1989, he has also worked as an assistant professor at University of Tours, in France.[3]

Grefenstette's research primarily focuses on natural language processing. Following his PhD work on “Exploring Automatic Thesaurus Generation”, he mostly addressed large-scale natural language processing problems and co-edited with Adam Kilgarriff a special issue of "Computational Linguistics" about how to use the Internet as a corpus for machine learning. Previous to his position at Biggerpan, Grefenstette was an Advanced Researcher at INRIA, the French national research institute in Computer Science, working on personal semantics. Prior to that, he was chief science officer of Exalead, a search engine company, managing the OSEO QUAERO CMSE program on innovative multimedia indexing technologies. Grefenstette is also a former chief scientist at the Xerox Research Centre Europe (1993-2001), at Clairvoyance Corporation (2001-2004), and with the French CEA (2004-2008).

Grefenstette was named as an inventor in 20 US patents, a majority of which were filed during his years at Xerox.[4] With his research team, he has received awards for his work on semantic maps and won a three-year grant by the Lagardere Foundation in 2007. He has also authored four books and has been part of several journal publications. Referenced in many natural language processing research papers, Grefenstette is especially renowned for his work on cross-language information retrieval and distributional semantics.

Selected works

Books

  • Grefenstette, Gregory; Wilber, Laura (2011). Search-based Applications: At the Confluence of Search and Database Technologies. Morgan & Claypool Publishers. ISBN 9781608455072.
  • Renals, Steve; Grefenstette, Gregory (2003-09-09). Text- and Speech-Triggered Information Access: 8th ELSNET Summer School, Chios Island, Greece, July 15-30, 2000, Revised Lectures. Springer. ISBN 9783540451150.
  • Grefenstette, Gregory (1998). Cross-language information retrieval. Kluwer Academic Publishers. ISBN 9781461527107.
  • Grefenstette, Gregory (2012-12-06). Explorations in Automatic Thesaurus Discovery. Springer Science & Business Media. ISBN 9781461527107.

Major publications[5]

Awards

gollark: I just know 14ish digits of it for no apparent reason. Accidental spaced repetition or whatever.
gollark: They're meant to be *omnipotent*, though, so they could just not.
gollark: It's something like 200 kilowords.
gollark: Vaguely semihyperrelatedly, I read some *great* Bible fanfiction a while ago. https://unsongbook.com/
gollark: Well, the Christian god is obviously probably very evil.

References

This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.