Improved Parsing for Argument-Clusters Coordination

TL;DR

This paper improves the parsing accuracy of Argument-Cluster Coordination structures by modifying the training data representation, leading to significant gains on a specialized corpus with prevalent ACC structures.

Contribution

The authors propose a new PTB representation for ACC and demonstrate its effectiveness in improving parser performance on relevant structures.

Findings

Slight improvement in EVALB scores on standard sections

2.7 times better ACC recovery on science exam corpus

Effective data representation change enhances ACC parsing

Abstract

Syntactic parsers perform poorly in prediction of Argument-Cluster Coordination (ACC). We change the PTB representation of ACC to be more suitable for learning by a statistical PCFG parser, affecting 125 trees in the training set. Training on the modified trees yields a slight improvement in EVALB scores on sections 22 and 23. The main evaluation is on a corpus of 4th grade science exams, in which ACC structures are prevalent. On this corpus, we obtain an impressive x2.7 improvement in recovering ACC structures compared to a parser trained on the original PTB trees.