Efficiency, Robustness, and Accuracy in Picky Chart Parsing

Abstract

This paper describes Picky, a probabilistic agenda-based chart parsing algorithm which uses a technique called {\em probabilistic prediction} to predict which grammar rules are likely to lead to an acceptable parse of the input. Using a suboptimal search method, Picky significantly reduces the number of edges produced by CKY-like chart parsing algorithms, while maintaining the robustness of pure bottom-up parsers and the accuracy of existing probabilistic parsers. Experiments using Picky demonstrate how probabilistic modelling can impact upon the efficiency, robustness and accuracy of a parser.