New models for improving supertag disambiguation
John Chen, Srinivas Bangalore, K. Vijay‐Shanker · 1999
In previous work, supertag disambiguation has been presented as a robust partial parsing technique. In this paper we present two approaches: contextual models, which exploit a variety of features in order to improve supertag performance, and class-based models, which assign sets of supertags to words in order to substantially improve accuracy with only a slight increase in ambiguity.