Language understanding using hidden understanding models
Richard M. Schwartz, Scott L. Miller, David G. Stallard, J. Makhoul · 2002
Describes a sentence understanding system that is completely based on learned methods both for understanding individual sentences and for determining their meaning in the context of the preceding sentences. We describe the models used for each of three stages in the understanding: semantic parsing, semantic classification and discourse modeling. When we ran this system on the December 1994 test of the ARPA Air Travel Information System (ATIS) task, we achieved a 14.5% error rate. The error rate for those sentences that are context-independent (class A) was 9.5%.