We describe a generative probabilistic model of natural language, which we call HBG, that takes advantage of detailed linguistic information to resolve ambiguity. HBG incorporates...
Ezra Black, Frederick Jelinek, John D. Lafferty, D...
Corpus-based methods for natural language processing often use supervised training, requiring expensive manual annotation of training corpora. This paper investigates methods for ...
A supercompiler is a program which can performa deep transformation of programsusing a principle which is similarto partial evaluation, and can be referred to as metacomputation. ...
Andrei P. Nemytykh, Victoria A. Pinchuk, Valentin ...
Macrophone is a corpus of approximately 200,000 utterances, recorded over the telephone from a broad sample of about 5,000 American speakers. Sponsored by the Linguistic Data Cons...
While complete understanding of arbitrary input text remains in the future, it is currently possible to construct natural language processing systems that provide a partial unders...
Peggy M. Andersen, Philip J. Hayes, Steven P. Wein...