Inducing Deterministic Prolog Parsers from Treebanks: A Machine Learning Approach

Abstract

This paper presents a method for constructing deterministic Prolog parsers from corpora of parsed sentences. Our approach uses recent machine learning methods for inducing Prolog rules from examples (inductive logic programming). We discuss several advantages of this method compared to recent statistical methods and present results on learning complete parsers from portions of the ATIS corpus. Introduction Recent approaches to constructing robust parsers from corpora primarily use statistical and probabilistic methods such as stochastic context-free grammars (Black et al., 1992; Pereira and Schabes, 1992). Although several current methods learn some symbolic structures such as decision trees (Black et al., 1993) and transformations (Brill, 1993), statistical methods still dominate. In this paper, we present a method that uses recent techniques in machine learning to construct symbolic, deterministic parsers from parsed corpora (treebanks). Specifically, our approach is implemented in...