Benjamin Van Durme

PhD Candidate
Department of Computer Science
Department of Linguistics
University of Rochester
email:

(2006) MS Computer Science, U. of Rochester
(2004) MS Language Technologies, Carnegie Mellon University
(2001) BS Computer Science/BA Cognitive Science, U. of Rochester

NEWS: Spring 2010 will find me as Research Scientist at the Human Language Technology Center of Excellence (HLTCOE) at Johns Hopkins University.

I am interested in things like natural language processing (specifically computational semantics), probabilistic counting, knowledge representation, psycholinguistics, and computational theories of ethics.

While a grad student at CMU (under the direction of Eric Nyberg and Bob Frederking), I was involved with the department's efforts in the AQUAINT project on Question Answering, as well as Project HALO, aimed at building a system that could answer AP science exam questions.

I came to Rochester in order to pursue a thesis focused on commonsense knowledge extraction. On September 28th, 2007, I defended my proposal. My committee consists of Len Schubert and Dan Gildea of Computer Science, Greg Carlson of Linguistics, and William Cohen of the Machine Learning Department at CMU.

I spent the summers of '06 and '07 doing research at Google with Marius Pasca. This work was focussed on finding characteristic attributes (e.g., "mayor") for concept classes (e.g., "city"), as well as collecting such classes automatically. This work benefited from interactions with Deepak Ravichandran and Dekang Lin.

Ashwin Lall and I have been collaborating since summer '08 on topics related to space efficient counting over streaming data. My interest in this area was partially motivated by conversations with Miles Osborne and David Talbot.

I'm sometimes found lurking in the basement of the building which houses Brain and Cognitive Science, with a group led by T. Florian Jaeger, looking for ways to further mingle topics from computer science and (human) language processing. I've been primarily concerned with measuring the impact of corpora source selection when deriving n-gram statistics (these statistics are known to correlate with human performance in language production and comprehension, but which collection gives the best frequencies?). In this work I interact with fellow lab members such as Austin Frank, Alex Fine, Judith Degen, Ting Qian and Celeste Kidd of BCS, and Matthew Post of CS.


user: va nd ur me domain: cs.rochester.edu

References


Benjamin Van Durme and Ashwin Lall. Streaming Pointwise Mutual Information. NIPS. Vancouver, Canada. December 7-10, 2009.

Jonathan Gordon, Benjamin Van Durme and Lenhart K. Schubert. Weblogs as a Source for Extracting General World Knowledge. K-CAP. Redondo Beach, California. September 1-4, 2009. (POSTER)

Benjamin Van Durme and Ashwin Lall. Probabilistic Counting with Randomized Storage. Twenty-First International Joint Conference on Artificial Intelligence (IJCAI-09). Pasadena, CA. July 11 - 17, 2009. Earlier version available as TR 942, Department of Computer Science, University of Rochester, December 2008. [PDF] [SLIDES]

Benjamin Van Durme, Austin Frank and T. Florian Jaeger. Resources for cross-linguistic psycholinguistic research: the web is good enough. International Cognitive Linguistics Conference 11 (ICLC-11). Berkeley, CA. July 28 - August 3, 2009. (Event canceled)

Ting Qian, Benjamin Van Durme and Lenhart K. Schubert. Building a Semantic Lexicon of English Nouns via Bootstrapping. NAACL '09, Student Research Workshop. Boulder, CO. June, 2009.

Benjamin Van Durme, Phillip Michalak and Lenhart K. Schubert. Deriving Generalized Knowledge from Corpora using WordNet Abstraction. EACL'09. Athens, Greece. March 30 - April 3, 2009. [PDF]

Benjamin Van Durme, Austin Frank and T. Florian Jaeger. Comparing Sources of Corpus Frequency Information. The 22nd Annual Meeting of the CUNY Conference on Human Sentence Processing (CUNY-09). Davis, CA. March 26 - 29, 2009. (POSTER)

Lenhart K. Schubert and Benjamin Van Durme. Open Extraction of General Knowledge through Compositional Semantics. NSF Symposium on Semantic Knowledge Discovery, Organization and Use. New York, New York. November 14-17, 2008. [PDF]

Benjamin Van Durme and Lenhart K. Schubert. Open Knowledge Extraction through Compositional Language Processing. Symposium on Semantics in Systems for Text Processing (STEP'08). Venice, Italy. September 22-24, 2008. [PDF]

Benjamin Van Durme, Ting Qian and Lenhart K. Schubert. Class-Driven Attribute Extraction. COLING'08. Manchester, UK. August 18-22, 2008. [PDF]

Austin F. Frank, Celeste Kidd, Matthew Post, Benjamin Van Durme and T. Florian Jaeger. The Web as a Psycholinguistic Resource. Presented at the 5th International Workshop on Language Production. Annapolis, MD. July 28-30, 2008. (POSTER)

Benjamin Van Durme and Marius Pasca. Finding Cars, Goddesses and Enzymes: Parametrizable Acquisition of Labeled Instances for Open-Domain Information Extraction. Twenty-Third AAAI Conference on Artificial Intelligence. Chicago, Illinois, USA. July 13-17, 2008. [PDF]

Benjamin Van Durme. Notes on the Acquisition of Conditional Knowledge. Technical Report TR-937, Department of Computer Science, University of Rochester, Rochester, NY 14627. June 2008.

Dekang Lin, Shaojun Zhao, Benjamin Van Durme and Marius Pasca. Mining Parenthetical Translations from the Web by Word Alignment. The 46th Annual Meeting of the Association of Computational Linguistics: Human Language Technologies (ACL-08: HLT). Columbus, Ohio, USA. June 15-20, 2008. [PDF]

Marius Pasca and Benjamin Van Durme. Weakly-Supervised Acquisition of Open-Domain Classes and Class Attributes from Web Documents and Query Logs. The 46th Annual Meeting of the Association of Computational Linguistics: Human Language Technologies (ACL-08: HLT). Columbus, Ohio, USA. June 15-20, 2008. [PDF]

Marius Pasca, Benjamin Van Durme, Nikesh Garera. The Role of Documents vs. Queries in Extracting Class Attributes from Text . ACM Sixteenth Conference on Information and Knowledge Management (CIKM 2007). Lisboa, Portugal. November 6-9, 2007. [PDF]

Marius Pasca and Benjamin Van Durme. What You Seek is What You Get: Extraction of Class Attributes from Query Logs. Hyderabad, India, February 2007. International Joint Conference on Artificial Intelligence (IJCAI-07). [PDF]

Anna Kupsc, Teruko Mitamura, Benjamin Van Durme, Eric Nyberg. Pronominal Anaphora Resolution for Unrestricted Text. Lisbon, Portugal, May 24-30 2004. LREC.

Benjamin Van Durme, Yifen Huang, Anna Kupsc, Eric Nyberg. Towards Light Semantic Processing for Question Answering. Edmonton, Canada, May 31 2003. HLT/NAACL Workshop on Text Meaning. [PDF]

E. Nyberg, T. Mitamura, J. Callan, J. Carbonell, R. Frederking, K. Collins-Thompson, L. Hiyakumoto, Y. Huang, C. Huttenhower, S. Judy, J. Ko, A. Kupsc, L. Lita, V. Pedro, D. Svoboda and B. Van Durme. The JAVELIN Question-Answering System at TREC 2003: A Multi-Strategy Approach with Dynamic Planning. 12th Text REtrieval Conference, November 2003.