Improve coreference resolution with parameter tunable anaphoricity identification and global optimization

Authors:
Shuhan Qi;Xuan Wang;Xinxin Li
Affiliations:
Computer Application Research Center, Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen, China;Computer Application Research Center, Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen, China;Computer Application Research Center, Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen, China
Venue:
ICIC'11 Proceedings of the 7th international conference on Intelligent Computing: bio-inspired computing and applications
Year:
2011

Citing 14
Cited 0

A machine learning approach to coreference resolution of noun phrases

Computational Linguistics - Special issue on computational anaphora resolution
Evaluating automated and manual acquisition of anaphora resolution strategies

ACL '95 Proceedings of the 33rd annual meeting on Association for Computational Linguistics
Identifying anaphoric and non-anaphoric noun phrases to improve coreference resolution

COLING '02 Proceedings of the 19th international conference on Computational linguistics - Volume 1
Improving machine learning approaches to coreference resolution

ACL '02 Proceedings of the 40th Annual Meeting on Association for Computational Linguistics
The influence of minimum edit distance on reference resolution

EMNLP '02 Proceedings of the ACL-02 conference on Empirical methods in natural language processing - Volume 10
A mention-synchronous coreference resolution algorithm based on the Bell tree

ACL '04 Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics
Learning noun phrase anaphoricity to improve coreference resolution: issues in representation and optimization

ACL '04 Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics
Machine learning for coreference resolution: from local classification to global ranking

ACL '05 Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics
Kernel-based pronoun resolution with structured syntactic knowledge

ACL-44 Proceedings of the 21st International Conference on Computational Linguistics and the 44th annual meeting of the Association for Computational Linguistics
A large-scale exploration of effective global features for a joint entity detection and tracking model

HLT '05 Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing
Exploiting semantic role labeling, WordNet and Wikipedia for coreference resolution

HLT-NAACL '06 Proceedings of the main conference on Human Language Technology Conference of the North American Chapter of the Association of Computational Linguistics
Shallow semantics for coreference resolution

IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
Using decision trees for conference resolution

IJCAI'95 Proceedings of the 14th international joint conference on Artificial intelligence - Volume 2
Coreference resolution with loose transitivity constraints

CONLL Shared Task '11 Proceedings of the Fifteenth Conference on Computational Natural Language Learning: Shared Task

Quantified Score

Hi-index	0.00

Visualization

Abstract

We build an anaphoric classifier with tunable parameters and realize a global connection between the classifier and coreference resolution. 60 features are used to build the anaphoric classifier. A corpus ratio control method is proposed and a "probability threshold" method is introduced to tune the precision and recall of the anaphoric classifier. The anaphoricity identification joints with the coreference resolution in a way of global optimization, and the parameters of anaphoricity identification are tuned according the result of coreference resolution. Maximum entropy is used for anaphoricity identification and coreference resolution with selected features. The results that combine the coreference resolution with the anaphoric classifier with different recall and precision are analyzed, and a comparison between our system and other coreference resolution systems is taken in the experiments analyze part. Our system improves the baseline coreference resolution system from 50.57 raise up to 53.35 on CoNLL'11 share tasks development data set.