Paper: Finding Non-Local Dependencies: Beyond Pattern Matching

ACL ID P03-2006
Title Finding Non-Local Dependencies: Beyond Pattern Matching
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2003
Authors

We describe an algorithm for recover- ing non-local dependencies in syntac- tic dependency structures. The pattern- matching approach proposed by John- son (2002) for a similar task for phrase structure trees is extended with machine learning techniques. The algorithm is es- sentially a classifier that predicts a non- local dependency given a connected frag- ment of a dependency structure and a set of structural features for this frag- ment. Evaluating the algorithm on the Penn Treebank shows an improvement of both precision and recall, compared to the results presented in (Johnson, 2002).