Paper: Using CCG categories to improve Hindi dependency parsing

ACL ID P13-2107
Title Using CCG categories to improve Hindi dependency parsing
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2013
Authors

We show that informative lexical cate- gories from a strongly lexicalised for- malism such as Combinatory Categorial Grammar (CCG) can improve dependency parsing of Hindi, a free word order lan- guage. We first describe a novel way to obtain a CCG lexicon and treebank from an existing dependency treebank, using a CCG parser. We use the output of a su- pertagger trained on the CCGbank as a feature for a state-of-the-art Hindi depen- dency parser (Malt). Our results show that using CCG categories improves the accu- racy of Malt on long distance dependen- cies, for which it is known to have weak rates of recovery.