Parse Correction with Specialized Models for Difficult Attachment Types

Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, pages 1222–1233, Edinburgh, Scotland, UK, July 27–31, 2011. c⃝2011 Association for Computational Linguistics Parse Correction with Specialized Models for Difficult Attachment Types Enrique Henestroza Anguiano and Marie Candito Alpage (Universit´e Paris Diderot / INRIA) Paris, France henestro@inria.fr, marie.candito@linguist.jussieu.fr Abstract This paper develops a framework for syntac- tic dependency parse correction. Dependen- cies in an input parse tree are revised by se- lecting, for a given dependent, the best gov- ernor from within a small set of candidates. We use a discriminative linear ranking model to select the best governor from a group of candidates for a dependent, and our model in- cludes a rich feature set that encodes syntac- tic structure in the input parse tree. The parse correction framework is parser-agnostic, and can correct attachments using either a generic model or specialized models tailored to dif- ficult attachment types like coordination and pp-attachment. Our experiments show that parse correction, combining a generic model with specialized models for difficult attach- ment types, can successfully improve the qual- ity of predicted parse trees output by sev- eral representative state-of-the-art dependency parsers for French. 1 Introduction In syntactic dependency parse correction, attach- ments in an input parse tree are revised by selecting, for a given dependent, the best governor from within a small set of candidates. The motivation behind parse correction is that attachment decisions, espe- cially traditionally difficult ones like pp-attachment and coordination, may require substantial contextual information in order to be made accurately. Because syntactic dependency parsers predict the parse tree for an entire sentence, they may not be able to take into account sufficient context when making attach- ment decisions, due to computational complexity. Assuming nonetheless that a predicted parse tree is mostly accurate, parse correction can revise difficult attachments by using the predicted tree’s syntactic structure to restrict the set of candidate governors and extract a rich set of features to help select among them. Parse correction is also appealing because it is parser-agnostic: it can be trained to correct the output of any dependency parser. In Section 2 we discuss work related to parse correction, pp-attachment and coordination resolu- tion. In Section 3 we discuss dependency struc- ture and various statistical dependency parsing ap- proaches. In Section 4 we introduce the parse cor- rection framework, and Section 5 describes the fea- tures and learning model used in our implementa- tion. In Section 6 we present experiments in which parse correction revises the predicted parse trees of four state-of-the-art dependency parsers for French. We provide concluding remarks in Section 7. 2 Related Work Previous research directly concerning parse correc- tion includes that of Attardi and Ciaramita (2007), working on English and Swedish, who use an ap- proach that considers a fixed set of revision rules: each rule describes movements in the parse tree leading from a dependent’s original governor to a new governor, and a classifier is trained to select the correct revision rule for a …

Content truncated to 500 words