Skip to main navigation Skip to search Skip to main content

Coreference Resolution with Loose Transitivity Constraints

  • Harbin Institute of Technology Shenzhen

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Our system treats coreference resolution as an integer linear programming (ILP) problem. Extending Denis and Baldridge (2007) and Finkel and Manning (2008)'s work, we exploit loose transitivity constraints on coreference pairs. Instead of enforcing transitivity closure constraints, which brings O (n 3 ) complexity, we employ a strategy to reduce the number of constraints without large performance decrease, i.e., eliminating coreference pairs with probability below a threshold. Experimental results show that it achieves a better performance than pairwise classifiers.

Original languageEnglish
Title of host publicationCoNLL 2011 - 15th Conference on Computational Natural Language Learning
Subtitle of host publicationShared Task, Proceedings
EditorsSameer Pradhan
PublisherAssociation for Computational Linguistics (ACL)
Pages107-111
Number of pages5
ISBN (Electronic)9781937284084
StatePublished - 2011
Externally publishedYes
Event15th Conference on Computational Natural Language Learning: Shared Task, CoNLL 2011 - Portland, United States
Duration: 23 Jun 201124 Jun 2011

Publication series

NameCoNLL 2011 - 15th Conference on Computational Natural Language Learning: Shared Task, Proceedings

Conference

Conference15th Conference on Computational Natural Language Learning: Shared Task, CoNLL 2011
Country/TerritoryUnited States
CityPortland
Period23/06/1124/06/11

Fingerprint

Dive into the research topics of 'Coreference Resolution with Loose Transitivity Constraints'. Together they form a unique fingerprint.

Cite this