Paper: Using Derivation Trees for Treebank Error Detection

ACL ID P11-2122
Title Using Derivation Trees for Treebank Error Detection
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2011
Authors

This work introduces a new approach to checking treebank consistency. Derivation trees based on a variant of Tree Adjoining Grammar are used to compare the annotation of word sequences based on their structural similarity. This overcomes the problems of earlier approaches based on using strings of words rather than tree structure to identify the appropriate contexts for comparison. We re- port on the result of applying this approach to the Penn Arabic Treebank and how this ap- proach leads to high precision of error detec- tion.