Chinese Zero Pronoun Resolution: Some Recent Advances

34680

Authors

nan

Abstract

We extend Zhao and Ng's (2007) Chinese anaphoric zero pronoun resolver by (1) using a richer set of features and (2) exploiting the coreference links between zero pronouns dur- ing resolution. Results on OntoNotes show that our approach significantly outperforms two state-of-the-art anaphoric zero pronoun re- solvers. To our knowledge, this is the first work to report results obtained by an end-to- end Chinese zero pronoun resolver. 1

Keywords

nan

Full Content

Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 1360–1365, Seattle, Washington, USA, 18-21 October 2013. cāƒ2013 Association for Computational Linguistics Chinese Zero Pronoun Resolution: Some Recent Advances Chen Chen and Vincent Ng Human Language Technology Research Institute University of Texas at Dallas Richardson, TX 75083-0688 {yzcchen,vince}@hlt.utdallas.edu Abstract We extend Zhao and Ng's (2007) Chinese anaphoric zero pronoun resolver by (1) using a richer set of features and (2) exploiting the coreference links between zero pronouns dur- ing resolution. Results on OntoNotes show that our approach significantly outperforms two state-of-the-art anaphoric zero pronoun re- solvers. To our knowledge, this is the first work to report results obtained by an end-to- end Chinese zero pronoun resolver. 1 Introduction A zero pronoun (ZP) is a gap in a sentence that is found when a phonetically null form is used to refer to a real-world entity. An anaphoric zero pronoun (AZP) is a ZP that corefers with one or more preced- ing noun phrases (NPs) in the associated text. Un- like overt pronouns, ZPs lack grammatical attributes that are useful for overt pronoun resolution such as number and gender. This makes ZP resolution more challenging than overt pronoun resolution. We aim to improve the state of the art in Chinese AZP resolution by proposing two extensions. First, while previous approaches to this task have primarily focused on employing positional and syntactic fea- tures (e.g., Zhao and Ng (2007) [Z&N], Kong and Zhou (2010) [K&Z]), we exploit a richer set of fea- tures for capturing the context of an AZP and its candidate antecedents. Second, to alleviate the diffi- culty of resolving an AZP to an antecedent far away from it, we break down the process into smaller, in- termediate steps, where we allow coreference links between AZPs to be established. We apply our two extensions to a state-of-the-art Chinese AZP resolver proposed by Z&N and eval- uate the resulting resolver on the OntoNotes cor- pus. Experimental results show that this resolver sig- nificantly outperforms both Z&N's resolver and an- other state-of-the-art resolver proposed by K&Z. It is worth noting that while previous work on Chinese ZP resolution has reported results obtained via gold information (e.g., using gold AZPs and extracting candidate antecedents and other features from gold syntactic parse trees), this is the first work to report the results of an end-to-end Chinese ZP resolver. The rest of this paper is organized as follows. Sec- tion 2 describes the two baseline AZP resolvers. Sec- tions 3 and 4 discuss our two extensions. We present our evaluation results in Section 5 and our conclu- sions in Section 6. 2 Baseline AZP Resolution Systems An AZP resolution algorithm takes as input a set of AZPs produced by an AZP identification system. Below we first describe the AZP identifier we em- ploy, followed by our two baseline AZP resolvers. 2.1 Anaphoric Zero Pronoun Identification We employ two steps to identify AZPs. In the extrac- tion step, we heuristically extract candidate ZPs. In …

Content truncated to 500 words

Labeled Text

We extend Zhao and Ng's (2007) Chinese anaphoric zero pronoun resolver by (1) using a richer set of features and (2) exploiting the coreference links between zero pronouns during resolution. Results on OntoNotes show that our approach significantly outperforms two state-of-the-art anaphoric zero pronoun resolvers. To our knowledge, this is the first work to report results obtained by an end-toend Chinese zero pronoun resolver. 1

Publication Details

Paper ID:
34680
Year:
2013
Publication:
TIPSTER
Publication Type:
conf
DOI:
nan