Natural Language Understanding
TINA is a natural language understanding system developed by the
SLS group. This natural language system parses sentences using a
probabilistic context-free grammar. The probabilistic model uses an
approach called layered bigrams to estimate the likelihood of a
particular parse. The underlying syntactic grammar is augmented with
semantic tags allowing TINA to extract a semantic representaion of a
sentence in the form of a semantic frame.
"TINA: A natural language system for spoken language applications,"
Computational Linguistics, vol. 18, no. 1., pp. 61-86, March 1992.
"Robust Parsing for Spoken Language Systems,"
Proc. ICASSP, San Francisco, March, 1992.
S. Seneff, H. Meng and V. Zue,
"Language Modelling for Recognition and Understanding Using Layered Bigrams,"
Proc. ICSLP, Banff, Canada, October, 1992.