Paper: A Pipeline Framework For Dependency Parsing

ACL ID P06-2009
Title A Pipeline Framework For Dependency Parsing
Venue Annual Meeting of the Association of Computational Linguistics
Session Poster Session
Year 2006

Pipeline computation, in which a task is decomposed into several stages that are solved sequentially, is a common compu- tational strategy in natural language pro- cessing. The key problem of this model is that it results in error accumulation and suffers from its inability to correct mis- takes in previous stages. We develop a framework for decisions made via in pipeline models, which addresses these dif culties, and presents and evaluates it in the context of bottom up dependency parsing for English. We show improve- ments in the accuracy of the inferred trees relative to existing models. Interestingly, the proposed algorithm shines especially when evaluated globally, at a sentence level, where our results are signi cantly better than those of existing approaches.