Collaboration with Wayne Ward and Jim Martin at Boulder, and Kathy McKeown and Vasilis Hatzivasiloglou at Columbia University, funded by ARDA/AQUAINT. We are working on building automatic question-answering systems. Our focus is on automatic detection of opinions for answering opinion question, on the use of semantic role-parsing in question-answering, and on Chinese question-answering. An early paper on opinion-detection is:
Steven Bethard, Hong Yu, Ashley Thornton, Vasieleios Hativassiloglou, and Dan Jurafsky. 2004. Automatic Extraction of Opinion Propositions and their Holders. In Proceedings of AAAI Spring Symposium on Exploring Attitude and Affect in Text.
We work with a number of collaborators, beginning with Dan Gildea in his dissertation work, on automatic semantic parsing: assigning domain-independent semantic role labels (Agent, Patient, Instrument, etc) to input sentences. Much of Dan Gildeas's dissertation work was written up here:
Daniel Gildea and Daniel Jurafsky. 2002. Automatic Labeling of Semantic Roles. Computational Linguistics 28:3, 245-288.A recent paper on semantic role parsing is:
Pradhan, Sameer, Wayne Ward, Kadri Hacioglu, James H. Martin, and Daniel Jurafsky. 2004. Shallow Semantic Parsing Using Support Vector Machines. In Proceedings of NAACL-HLT 2004.
This work also involves close collaboration with the FrameNet and PropBank projects.
Rion Snow, Daniel Jurafsky, and Andrew Y. Ng. 2005. "Learning syntactic patterns for automatic hypernym discovery". In press, Proceedings of NIPS 2004.
Past projects include work with Dan Gildea on Induction of Phonological Rules:, a project focusing on the way that prior biases can simplify the task of inductive learning of phonological rules. See:
Gildea, Daniel and Daniel Jurafsky. (1996). Learning Bias and Phonological Rule Induction. Computational Linguistics 22, 497-530.Other past projects in the lab include Patrick Schone's work on automatic building of dictionaries based only on unlabeled input corpora. Some papers on automatic induction of morphology, using Latent Semantic Analysis and other tools, include:
Schone, Patrick and Daniel Jurafsky. 2000. Knowlege-Free Induction of Morphology using Latent Semantic Analysis. Proceedings of the Conference on Computational Natural Language Learning (CoNLL-2000).Other work of Pat's focused on using knowledge about language universals to induce part of speech labels:
and
Schone, Patrick and Daniel Jurafsky. 2001. Knowlege-Free Induction of Inflectional Morphologies. Proceedings of the North American chapter of the Association for Computational Linguistics (NAACL-2001). .
Schone, Patrick and Daniel Jurafsky. 2001. Language-Independent Induction of Part of Speech Class Labels Using Only Language Universals. In IJCAI-2001 Workshop "Text Learning: Beyond Supervision".