Provably Reliable Question-answering Interfaces

CSE Home

About Us

Search

Contact Info

General  Reliable NLI Home Faculty  Oren Etzioni  Henry Kautz  Dan Weld Students  Ana-Maria Popescu  Alexander Yates UW Natural Language Community  Mari Ostendorf (EE Dept.)  Speech Group (EE Dept.)  Wanda Pratt (Information School)  John Goldsmith (Linguistics Dept.)  Eric Brill (Microsoft Research)

The need for natural language interfaces to databases (NLIDBs) has become more acute as more nontechnical people access information through their web browsers, PDAs, and cell phones. Yet NLIDBs are only useful if they map natural language questions to SQL queries correctly, and natural language interfaces tend to be error-prone. The PRECISE Natural Language Interface to Databases is designed on the principle that it should guarantee the correctness of its output, or else indicate that it does not understand the input question. For a certain class of semantically tractable questions, PRECISE is able to recognize a member of that class and guarantee that its SQL output is the correct interpretation of the question. Publications Modern Natural Language Interfaces to Databases: Composing Statistical Parsing with Semantic Tractability . Ana-Maria Popescu, Alex Armanasu, Oren Etzioni, David Ko, Alexander Yates. COLING 2004 Towards a Theory of Natural Language Interfaces to Databases. Ana-Maria Popescu, Oren Etzioni, and Henry Kautz. Intelligent User Interfaces, 2003. A Reliable Natural Language Interface to Household Appliances. Alexander Yates, Oren Etzioni, and Daniel Weld. Intelligent User Interfaces, 2003. Scaling Question Answering to the Web. Cody Kwok, Oren Etzioni, and Daniel S. Weld. ACM Transactions on Information Systems, Vol. 19, No. 3, July 2001, pp. 242-262.

	Department of Computer Science & Engineering University of Washington Box 352350 Seattle, WA  98195-2350 (206) 543-1695 voice, (206) 543-2969 FAX [comments to amp@cs.washington.edu]