Changes: Open-source entity coreference resolution systems

Revision as of 12:54, 13 April 2018

System	Syntax feature?	Semantic role feature?	Semantic type feature? ^{[note 1]}	Word-window feature?	Mention-pair?	Entity-mention?	Mention-ranking?	Cluster-ranking?	Cluster-pair?	Rule-based?	Base ML model	Integer linear programming?	Reference	Notes
cort	Yes ^{[note 2]}	No	Yes ^{[note 3]}		Yes	Yes	Yes	No	No	No	perceptron	No	Martschat and Strube (2015)^[1]
nn_coref	Yes ^{[note 4]}	No	Yes ^{[note 5]}		No	Yes	No	No	No	No	neural net (RNN for encoding clusters)	No	Wiseman et al. (2016)^[2]
huggingface's neural coref	Yes	No	Yes ^{[note 6]}	Yes ^{[note 7]}	No	No	Yes	No	No	No	neural net	No	medium post	impl. of Clark and Manning (2016)^[3]
deep-coref	Yes	No	Yes ^{[note 6]}	Yes ^{[note 7]}	Yes	No	Yes	Yes	No	No	neural net	No	Clark and Manning (2016)^[4]
hcoref (Hybrid Coref)	Yes ^{[note 8]}	No	Yes ^{[note 9]}	No	No	Yes?	No	No	Yes?	Yes ^{[note 10]}	random forest ^{[note 10]}	No	Lee et al. (2017)^[5]
dcoref(Stanford sieve)	Yes	No	Yes ^{[note 11]}	No	No	No	No	No	Yes	Yes	None	No	Lee et al. (2013)^[6]	part of Stanford CoreNLP
Berkeley CR	No	No	Yes	No	No	Yes ^{[note 12]}	Yes ^{[note 13]}	No	No	No	log-linear	No	Durrett and Klein (2013)^[7]
Illinois CR
xrenner														eXternally configurable REference and Non Named Entity Recognizer
e2e-coref														end-to-end coreference resolution system from AllenAI

Notes

↑ Different from semantic role features, this includes features about mentions alone: semantic type (person/object/number), NER type (person/location/organization), or other taxonomies.
↑ deprel: dependency relation of a mention to its governor
↑ sem_class: one of 'PERSON', 'OBJECT', 'NUMERIC' and 'UNKNOWN' and head_ner: named entity tag of the mention's head word
↑ From syntactic ancestry features in BASIC+ (Wiseman et al. 2015)
↑ From entity type features in BASIC+ (Wiseman et al. 2015)
↑ ^6.0 ^6.1 In Clark and Manning (2016): "The type of the mention (pronoun, nominal, proper, or list)"
↑ ^7.0 ^7.1 From Clark and Manning (2016): "first word, last word, two preceding words, and two following words of the mention. Averaged word embed- dings of the five preceding words, five following words, all words in the mention, all words in the mention’s sentence, and all words in the mention’s document."
↑ Feature: "The path in the parse tree from the root to the (antecedent/anaphor)"
↑ Feature: "named entity type attributes of (antecedent/anaphor)"
↑ ^10.0 ^10.1 They combine rule-based and statistical classifiers.
↑ "NER label – fromthe Stanford NER"
↑ TRANSITIVE model: "each mention to maintain its own distributions over values for a number of proper- ties; these properties could include gender, named- entity type, or semantic class. Then, we will require each anaphoric mention to agree with its antecedent on the value of each of these properties"
↑ BASIC model: "This approach is similar to the mention- ranking model of Rahman and Ng (2009)."

References

↑ Sebastian Martschat and Michael Strube. 2015. Latent structures for coreference resolution. TACL, 3:405– 418.
↑ Wiseman, Sam, Alexander M. Rush, and Stuart M. Shieber. "Learning Global Features for Coreference Resolution." arXiv preprint arXiv:1604.03035(2016).
↑ Clark, K., & Manning, C. D. (2016). Deep Reinforcement Learning for Mention-Ranking Coreference Models. Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing (EMNLP-16), 2256–2262.
↑ Clark, K., & Manning, C. D. (2016a). Improving Coreference Resolution by Learning Entity-Level Distributed Representations. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 643–653. http://doi.org/10.18653/v1/P16-1061
↑ LEE, HEEYOUNG, MIHAI SURDEANU, and DAN JURAFSKY. "A scaffolding approach to coreference resolution integrating statistical and rule-based models." Natural Language Engineering (2017): 1-30.
↑ Heeyoung Lee, Angel Chang, Yves Peirsman, Nathanael Chambers, Mihai Surdeanu and Dan Jurafsky. Deterministic coreference resolution based on entity-centric, precision-ranked rules. Computational Linguistics 39(4), 2013.
↑ Durrett, G., & Klein, D. (2013). Easy victories and uphill battles in coreference resolution. EMNLP ’13, (October), 1971–1982.

[1] Different from semantic role features, this includes features about mentions alone: semantic type (person/object/number), NER type (person/location/organization), or other taxonomies.

[2] rel: dependency relation of a mention to its governor

[3] sem_class: one of 'PERSON', 'OBJECT', 'NUMERIC' and 'UNKNOWN' and head_ner: named entity tag of the mention's head word

[5] From syntactic ancestry features in BASIC+ (Wiseman et al. 2015)

[6] From entity type features in BASIC+ (Wiseman et al. 2015)

[:0-8] 6.0 ^6.1 In Clark and Manning (2016): "The type of the mention (pronoun, nominal, proper, or list)"

[:1-9] 7.0 ^7.1 From Clark and Manning (2016): "first word, last word, two preceding words, and two following words of the mention. Averaged word embed- dings of the five preceding words, five following words, all words in the mention, all words in the mention’s sentence, and all words in the mention’s document."

[12] Feature: "The path in the parse tree from the root to the (antecedent/anaphor)"

[13] Feature: "named entity type attributes of (antecedent/anaphor)"

[:2-14] 10.0 ^10.1 They combine rule-based and statistical classifiers.

[16] "NER label – fromthe Stanford NER"

[18] TRANSITIVE model: "each mention to maintain its own distributions over values for a number of proper- ties; these properties could include gender, named- entity type, or semantic class. Then, we will require each anaphoric mention to agree with its antecedent on the value of each of these properties"

[19] BASIC model: "This approach is similar to the mention- ranking model of Rahman and Ng (2009)."

[:4-4] Sebastian Martschat and Michael Strube. 2015. Latent structures for coreference resolution. TACL, 3:405– 418.

[:2-7] Wiseman, Sam, Alexander M. Rush, and Stuart M. Shieber. "Learning Global Features for Coreference Resolution." arXiv preprint arXiv:1604.03035(2016).

[10] Clark, K., & Manning, C. D. (2016). Deep Reinforcement Learning for Mention-Ranking Coreference Models. Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing (EMNLP-16), 2256–2262.

[:3-11] Clark, K., & Manning, C. D. (2016a). Improving Coreference Resolution by Learning Entity-Level Distributed Representations. Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 643–653. http://doi.org/10.18653/v1/P16-1061

[15] LEE, HEEYOUNG, MIHAI SURDEANU, and DAN JURAFSKY. "A scaffolding approach to coreference resolution integrating statistical and rule-based models." Natural Language Engineering (2017): 1-30.

[17] Heeyoung Lee, Angel Chang, Yves Peirsman, Nathanael Chambers, Mihai Surdeanu and Dan Jurafsky. Deterministic coreference resolution based on entity-centric, precision-ranked rules. Computational Linguistics 39(4), 2013.

[20] Durrett, G., & Klein, D. (2013). Easy victories and uphill battles in coreference resolution. EMNLP ’13, (October), 1971–1982.

[note 1]

[note 2]

[note 3]

[1]

[note 4]

[note 5]

[2]

[note 6]

[note 7]

[3]

[4]

[note 8]

[note 9]

[note 10]

[5]

[note 11]

[6]

[note 12]

[note 13]

[7]

@@ Line 128: / Line 128: @@
 |None
 |No
+|Lee et al. (2013)<ref>Heeyoung Lee, Angel Chang, Yves Peirsman, Nathanael Chambers, Mihai Surdeanu and Dan Jurafsky. Deterministic coreference resolution based on entity-centric, precision-ranked rules. Computational Linguistics 39(4), 2013.</ref>
-|Lee et al. (2013)<ref><blockquote>
-Heeyoung Lee, Angel Chang, Yves Peirsman, Nathanael Chambers, Mihai Surdeanu and Dan Jurafsky.
-<br>
-Deterministic coreference resolution based on entity-centric, precision-ranked rules.
-<br>
-Computational Linguistics 39(4), 2013.
-</blockquote></ref>
 |part of [http://stanfordnlp.github.io/CoreNLP/ Stanford CoreNLP]
 |-