---
title: "Coreference Systems Based on Kernels Methods — Versley 2008"
description: "The paper investigates how kernel methods can supply coreference resolution with knowledge that is partly structural and partly lexical."
authors: ["Yannick Versley", "Alessandro Moschitti", "Massimo Poesio", "Xiaofeng Yang"]
year: 2008
venue: "Proceedings of the 22nd International Conference on Computational Linguistics (Coling 2008)"
source: "https://aclanthology.org/C08-1121/"
licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)"
licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/papers/versley-2008-coreference-systems-based-kernels-methods/"
---
[Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/versley-2008-coreference-systems-based-kernels-methods/)

[home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2008](https://md.research.elamur.ai/free/papers/2008/) / Coreference Systems Based on Kernels Methods

# Coreference Systems Based on Kernels Methods

**Authors:** [Yannick Versley](https://md.research.elamur.ai/free/authors/yannick-versley/), [Alessandro Moschitti](https://md.research.elamur.ai/free/authors/alessandro-moschitti/), [Massimo Poesio](https://md.research.elamur.ai/free/authors/massimo-poesio/), [Xiaofeng Yang](https://md.research.elamur.ai/free/authors/xiaofeng-yang/) · **Year:** 2008 · **Venue:** Proceedings of the 22nd International Conference on Computational Linguistics (Coling 2008) · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/)

**Links:** [Publisher's page](https://aclanthology.org/C08-1121/) · [PDF](https://aclanthology.org/C08-1121.pdf) · [OpenAlex](https://openalex.org/W2042609668)

**Citations:** 30 citations (openalex-snapshot · 2026-10-14) · [13](#cited-by) in our library, of the 140673 papers whose bibliographies we read

## Summary
The paper investigates how kernel methods can supply coreference resolution with knowledge that is partly structural and partly lexical. Examples include binding constraints, recognizing non-referential ("expletive") *it*, and matching name variants. The authors build on the Soon et al. (2001) pairwise anaphor–antecedent framework, implemented with SVMs and a polynomial kernel (PK). They add kernels from three families: String Kernels, the Syntactic Tree Kernel (STK) and the Partial Tree Kernel (PTK). These are applied to task-specific data views:
- for expletives, the pronoun's clause in both a full and a pruned form;
- for binding, the shortest tree path linking a pronoun to its candidate antecedent;
- for context, word sequences.

Some outputs are fed to the resolver as features from separate classifiers. Others are merged directly into the coreference kernel. On MUC-6 and ACE02-BNews, the binding tree kernel substantially improves pronoun resolution, and full coreference gains about 1–2 F points. The work matters because it shows that structural knowledge can be exploited without hand-enumerating features. It also exposes a trade-off between accuracy and efficiency when classifiers are kept separate rather than integrated.

## Key points
- The expletive classifier combines an unmodified tree with a reduced tree that keeps only pattern-relevant structure. It is trained on the BBN pronoun corpus and used as a feature on MUC-6. Pronoun F rose from 66.0 to 66.5, against an upper bound of 68.4 with gold expletives.
- Adding the binding STK to PK sharply improved pronoun resolution:
  - On MUC-6, F rose from 63.7 to 71.9, with precision going from 63.1 to 80.1.
  - On ACE02-BNews, F rose from 63.1 to 67.6.
- On validation, the best combination of the two kernels was STK·PK + PK. Adding a Word Sequence Kernel (WSK) over context windows helped further.
- For full coreference:
  - On MUC-6, F went from 64.2 (PK) to 65.4 (+STK) and 66.0 (+WSK).
  - On ACE02-BNews, the same steps gave 59.9, 60.9 and 61.0.
- Kernels for name aliasing were also proposed. The authors note that the size of the ACE corpora prevented them from combining all the kernels at once. Using separate classifiers as binary features is faster but loses some accuracy.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/shelves/linguistics/)

Topics: [Coreference resolution](https://md.research.elamur.ai/free/topics/coreference-resolution/)

## Cited by

Cited by

13 papers of our library cite it (of the 140673 whose bibliographies we read); 5 of them have a page here, 8 not yet.

- [Resolving Object and Attribute Coreference in Opinion Mining](https://md.research.elamur.ai/free/papers/ding-2010-resolving-object-attribute-coreference-opinion/)
- [Improving pairwise coreference models through feature space hierarchy learning](https://md.research.elamur.ai/free/papers/lassalle-2013-improving-pairwise-coreference-models-through/)
- [Graph-Cut-Based Anaphoricity Determination for Coreference Resolution](https://md.research.elamur.ai/free/papers/ng-2009-graph-cut-based-anaphoricity-determination/)
- [Supervised Noun Phrase Coreference Research: The First Fifteen Years](https://md.research.elamur.ai/free/papers/ng-2010-supervised-noun-phrase-coreference-research/)
- [A Cluster Ranking Model for Full Anaphora Resolution](https://md.research.elamur.ai/free/papers/yu-2020-cluster-ranking-model-full-anaphora/)

## Cite

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{versley2008coreference,
  title = {{Coreference Systems Based on Kernels Methods}},
  author = {Versley, Yannick and Moschitti, Alessandro and Poesio, Massimo and Yang, Xiaofeng},
  booktitle = {Proceedings of the 22nd International Conference on Computational Linguistics (Coling 2008)},
  year = {2008},
  doi = {10.3115/1599081.1599202},
  url = {https://doi.org/10.3115/1599081.1599202}
}
```

- [← previous Context-Sensitive Convolution Tree Kernel for Pronoun Resolution](https://md.research.elamur.ai/free/papers/zhou-2008-context-sensitive-convolution-tree-kernel/)
- [next → Coreference-inspired Coherence Modeling](https://md.research.elamur.ai/free/papers/elsner-2008-coreference-inspired-coherence-modeling/)
- [↑ back to papers of 2008](https://md.research.elamur.ai/free/papers/2008/)

## References

35 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Aone, C. and Bennett, S. W. (1995). Evaluating automated and manual acquisition of anaphora resolution strategies. In Proc. [find](https://scholar.google.com/scholar?q=Aone%2C+C%2E+and+Bennett%2C+S%2E+W%2E+%281995%29%2E+Evaluating+automated+and+manual+acquisition+of+anaphora+resolution+strategies%2E+In+Proc%2E) [our card](https://md.research.elamur.ai/free/papers/aone-1995-evaluating-automated-manual-acquisition-anaphora/)
2. Barzilay, R. and Lapata, M. (2005). Modelling local coherence: An entity-based approach. In Proc. of ACL, Ann Arbor, MI. [find](https://scholar.google.com/scholar?q=Barzilay%2C+R%2E+and+Lapata%2C+M%2E+%282005%29%2E+Modelling+local+coherence%3A+An+entity%2Dbased+approach%2E+In+Proc%2E+of+ACL%2C+Ann+Arbor%2C+MI%2E)
3. Boyd, A., Gegg-Harrison, W., and Byron, D. (2005). Identifying non-referential it: a machine learning approach incorporating linguistically motivated features. In Proc. ACL WS on Feature Engineering for Machine Learning in Natural Language Processing. [find](https://scholar.google.com/scholar?q=Boyd%2C+A%2E%2C+Gegg%2DHarrison%2C+W%2E%2C+and+Byron%2C+D%2E+%282005%29%2E+Identifying+non%2Dreferential+it%3A+a+machine+learning+approach+incorporating+linguistically+motivated+features%2E+In+Proc%2E+ACL+WS+on+Feature+Engineering+f)
4. Cancedda, N., Gaussier, E., Goutte, C., and Renders, J. M. (2003). Word sequence kernels. JMLR, 3:1059–1082. [find](https://scholar.google.com/scholar?q=Cancedda%2C+N%2E%2C+Gaussier%2C+E%2E%2C+Goutte%2C+C%2E%2C+and+Renders%2C+J%2E+M%2E+%282003%29%2E+Word+sequence+kernels%2E+JMLR%2C+3%3A1059%E2%80%931082%2E)
5. Chinchor, N. and Sundheim, B. (2003). Muc 6 corpus. Message Understanding Conference (MUC) 6. [find](https://scholar.google.com/scholar?q=Chinchor%2C+N%2E+and+Sundheim%2C+B%2E+%282003%29%2E+Muc+6+corpus%2E+Message+Understanding+Conference+%28MUC%29+6%2E)
6. Chomsky, N. (1981). Lectures on government and binding. Foris, Dordrecht, The Netherlands. [find](https://scholar.google.com/scholar?q=Chomsky%2C+N%2E+%281981%29%2E+Lectures+on+government+and+binding%2E+Foris%2C+Dordrecht%2C+The+Netherlands%2E)
7. Collins, M. and Duffy, N. (2002). New ranking algorithms for parsing and tagging: kernels over discrete structures and the voted perceptron. In Proc. [find](https://scholar.google.com/scholar?q=Collins%2C+M%2E+and+Duffy%2C+N%2E+%282002%29%2E+New+ranking+algorithms+for+parsing+and+tagging%3A+kernels+over+discrete+structures+and+the+voted+perceptron%2E+In+Proc%2E)
8. Evans, R. (2001). Applying machine learning toward an automatic classification of it. Literary and Linguistic Computing, 16(1):45–57. [find](https://scholar.google.com/scholar?q=Evans%2C+R%2E+%282001%29%2E+Applying+machine+learning+toward+an+automatic+classification+of+it%2E+Literary+and+Linguistic+Computing%2C+16%281%29%3A45%E2%80%9357%2E)
9. Giuglea, A.-M. and Moschitti, A. (2006). Semantic role labeling via framenet, verbnet and propbank. In Proceedings of Coling-ACL, Sydney, Australia. [find](https://scholar.google.com/scholar?q=Giuglea%2C+A%2E%2DM%2E+and+Moschitti%2C+A%2E+%282006%29%2E+Semantic+role+labeling+via+framenet%2C+verbnet+and+propbank%2E+In+Proceedings+of+Coling%2DACL%2C+Sydney%2C+Australia%2E)
10. Grosz, B., Joshi, A., and Weinstein, S. (1995). Centering: a framework for modeling the local coherence of discourse. CL, 21(2):203–225. [find](https://scholar.google.com/scholar?q=Grosz%2C+B%2E%2C+Joshi%2C+A%2E%2C+and+Weinstein%2C+S%2E+%281995%29%2E+Centering%3A+a+framework+for+modeling+the+local+coherence+of+discourse%2E+CL%2C+21%282%29%3A203%E2%80%93225%2E) [our card](https://md.research.elamur.ai/free/papers/cl-1995-grosz-centering-framework-modeling-local-coherence/)
11. Hobbs, J. (1978). Resolving pronoun references. Lingua, 44:339–352. [find](https://scholar.google.com/scholar?q=Hobbs%2C+J%2E+%281978%29%2E+Resolving+pronoun+references%2E+Lingua%2C+44%3A339%E2%80%93352%2E) [our card](https://md.research.elamur.ai/free/papers/hobbs-1978-resolving-pronoun-references/)
12. Hobbs, J. (1979). Resolving pronoun references. Coherence and Coreference, 3(1):67–90. [find](https://scholar.google.com/scholar?q=Hobbs%2C+J%2E+%281979%29%2E+Resolving+pronoun+references%2E+Coherence+and+Coreference%2C+3%281%29%3A67%E2%80%9390%2E) [our card](https://md.research.elamur.ai/free/papers/hobbs-1978-resolving-pronoun-references/)
13. Iida, R., Inui, K., and Matsumoto, Y. (2006). Exploiting syntactic patterns as clues in zero-anaphora resolution. In Proc. Coling/ACL 2006, pages 625–632. [find](https://scholar.google.com/scholar?q=Iida%2C+R%2E%2C+Inui%2C+K%2E%2C+and+Matsumoto%2C+Y%2E+%282006%29%2E+Exploiting+syntactic+patterns+as+clues+in+zero%2Danaphora+resolution%2E+In+Proc%2E+Coling%2FACL+2006%2C+pages+625%E2%80%93632%2E) [our card](https://md.research.elamur.ai/free/papers/iida-2006-exploiting-syntactic-patterns-as-clues/)
14. Kennedy, C. and Boguraev, B. (1996). Anaphora for everyone: pronominal anaphora resolution without a parser. In Proc. [find](https://scholar.google.com/scholar?q=Kennedy%2C+C%2E+and+Boguraev%2C+B%2E+%281996%29%2E+Anaphora+for+everyone%3A+pronominal+anaphora+resolution+without+a+parser%2E+In+Proc%2E) [our card](https://md.research.elamur.ai/free/papers/kennedy-1996-anaphora-everyone-pronominal-anaphora-resolution/)
15. Lappin, S. and Leass, H. (1994). An algorithm for pronominal anaphora resolution. CL, 20(4):525–561. [find](https://scholar.google.com/scholar?q=Lappin%2C+S%2E+and+Leass%2C+H%2E+%281994%29%2E+An+algorithm+for+pronominal+anaphora+resolution%2E+CL%2C+20%284%29%3A525%E2%80%93561%2E) [our card](https://md.research.elamur.ai/free/papers/lappin-1994-algorithm-pronominal-anaphora-resolution/)
16. Luo, X. and Zitouni, I. (2005). Multi-lingual coreference resolution with syntactic features. In Proc. HLT/EMNLP 05. [find](https://scholar.google.com/scholar?q=Luo%2C+X%2E+and+Zitouni%2C+I%2E+%282005%29%2E+Multi%2Dlingual+coreference+resolution+with+syntactic+features%2E+In+Proc%2E+HLT%2FEMNLP+05%2E) [our card](https://md.research.elamur.ai/free/papers/luo-2005-multi-lingual-coreference-resolution-syntactic/)
17. McCarthy, J. and Lehnert, W. (1995). Using decision trees for coreference resolution. In Proc. [find](https://scholar.google.com/scholar?q=McCarthy%2C+J%2E+and+Lehnert%2C+W%2E+%281995%29%2E+Using+decision+trees+for+coreference+resolution%2E+In+Proc%2E)
18. Mitkov, R. (2002). Anaphora resolution. Longman. [find](https://scholar.google.com/scholar?q=Mitkov%2C+R%2E+%282002%29%2E+Anaphora+resolution%2E+Longman%2E)
19. Moschitti, A. (2006). Efficient convolution kernels for dependency and constituent syntactic trees. [find](https://scholar.google.com/scholar?q=Moschitti%2C+A%2E+%282006%29%2E+Efficient+convolution+kernels+for+dependency+and+constituent+syntactic+trees%2E)
20. Moschitti, A. and Bejan, C. A. (2004). A semantic kernel for predicate argument classification. In CoNLL-2004, USA. [find](https://scholar.google.com/scholar?q=Moschitti%2C+A%2E+and+Bejan%2C+C%2E+A%2E+%282004%29%2E+A+semantic+kernel+for+predicate+argument+classification%2E+In+CoNLL%2D2004%2C+USA%2E)
21. Moschitti, A., Pighin, D., and Basili, R. (2006). Semantic Role Labeling via Tree Kernel Joint Inference. In Proceedings of CoNLL-X. [find](https://scholar.google.com/scholar?q=Moschitti%2C+A%2E%2C+Pighin%2C+D%2E%2C+and+Basili%2C+R%2E+%282006%29%2E+Semantic+Role+Labeling+via+Tree+Kernel+Joint+Inference%2E+In+Proceedings+of+CoNLL%2DX%2E)
22. Moschitti, A., Quarteroni, S., Basili, R., and Manandhar, S. (2007). Exploiting syntactic and shallow semantic kernels for question answer classification. In Proceedings ACL, Prague, Czech Republic. [find](https://scholar.google.com/scholar?q=Moschitti%2C+A%2E%2C+Quarteroni%2C+S%2E%2C+Basili%2C+R%2E%2C+and+Manandhar%2C+S%2E+%282007%29%2E+Exploiting+syntactic+and+shallow+semantic+kernels+for+question+answer+classification%2E+In+Proceedings+ACL%2C+Prague%2C+Czech+Republic%2E)
23. Ng, V. and Cardie, C. (2002). Improving machine learning approaches to coreference resolution. In Proc. [find](https://scholar.google.com/scholar?q=Ng%2C+V%2E+and+Cardie%2C+C%2E+%282002%29%2E+Improving+machine+learning+approaches+to+coreference+resolution%2E+In+Proc%2E) [our card](https://md.research.elamur.ai/free/papers/ng-2002-improving-machine-learning-approaches-coreference/)
24. Paice, C. D. and Husk, G. D. (1987). Towards an automatic recognition of anaphoric features in english text: The impersonal pronoun ‘it’. Computer Speech and Language, 2:109–132. [find](https://scholar.google.com/scholar?q=Paice%2C+C%2E+D%2E+and+Husk%2C+G%2E+D%2E+%281987%29%2E+Towards+an+automatic+recognition+of+anaphoric+features+in+english+text%3A+The+impersonal+pronoun+%E2%80%98it%E2%80%99%2E+Computer+Speech+and+Language%2C+2%3A109%E2%80%93132%2E)
25. Shawe-Taylor, J. and Cristianini, N. (2004). Kernel Methods for Pattern Analysis. Cambridge University Press. [find](https://scholar.google.com/scholar?q=Shawe%2DTaylor%2C+J%2E+and+Cristianini%2C+N%2E+%282004%29%2E+Kernel+Methods+for+Pattern+Analysis%2E+Cambridge+University+Press%2E)
26. Sidner, C. (1979). Toward a computational theory of definite anaphora comprehension in english. Technical report AI-TR-537, MIT, Cambridge, MA. [find](https://scholar.google.com/scholar?q=Sidner%2C+C%2E+%281979%29%2E+Toward+a+computational+theory+of+definite+anaphora+comprehension+in+english%2E+Technical+report+AI%2DTR%2D537%2C+MIT%2C+Cambridge%2C+MA%2E)
27. Soon, W., Ng, H., and Lim, D. (2001). A machine learning approach to coreference resolution of noun phrases. CL, 27(4):521–544. [find](https://scholar.google.com/scholar?q=Soon%2C+W%2E%2C+Ng%2C+H%2E%2C+and+Lim%2C+D%2E+%282001%29%2E+A+machine+learning+approach+to+coreference+resolution+of+noun+phrases%2E+CL%2C+27%284%29%3A521%E2%80%93544%2E) [our card](https://md.research.elamur.ai/free/papers/soon-2001-machine-learning-approach-coreference-resolution/)
28. Steinberger, J., Poesio, M., Kabadjov, M., and Jezek, K. (2007). Two uses of anaphora resolution in summarization. Information Processing and Management, 43:1663–1680. Special issue on Summarization. [find](https://scholar.google.com/scholar?q=Steinberger%2C+J%2E%2C+Poesio%2C+M%2E%2C+Kabadjov%2C+M%2E%2C+and+Jezek%2C+K%2E+%282007%29%2E+Two+uses+of+anaphora+resolution+in+summarization%2E+Information+Processing+and+Management%2C+43%3A1663%E2%80%931680%2E+Special+issue+on+Summarization%2E)
29. Sturt, P. (2003). The time-course of the application of binding constraints in reference resolution. Journal of Memory and Language. [find](https://scholar.google.com/scholar?q=Sturt%2C+P%2E+%282003%29%2E+The+time%2Dcourse+of+the+application+of+binding+constraints+in+reference+resolution%2E+Journal+of+Memory+and+Language%2E)
30. Vapnik, V. (1995). The Nature of Statistical Learning Theory. Springer. [find](https://scholar.google.com/scholar?q=Vapnik%2C+V%2E+%281995%29%2E+The+Nature+of+Statistical+Learning+Theory%2E+Springer%2E)
31. Vieira, R. and Poesio, M. (2000). An empirically based system for processing definite descriptions. CL, 27(4):539– 592. [find](https://scholar.google.com/scholar?q=Vieira%2C+R%2E+and+Poesio%2C+M%2E+%282000%29%2E+An+empirically+based+system+for+processing+definite+descriptions%2E+CL%2C+27%284%29%3A539%E2%80%93+592%2E)
32. Yang, X., Su, J., and Tan, C. (2006). Kernel-based pronoun resolution with structured syntactic knowledge. In Proc. COLING-ACL 06. [find](https://scholar.google.com/scholar?q=Yang%2C+X%2E%2C+Su%2C+J%2E%2C+and+Tan%2C+C%2E+%282006%29%2E+Kernel%2Dbased+pronoun+resolution+with+structured+syntactic+knowledge%2E+In+Proc%2E+COLING%2DACL+06%2E) [our card](https://md.research.elamur.ai/free/papers/yang-2006-kernel-based-pronoun-resolution-structured/)
33. Yang, X., Su, J., Zhou, G., and Tan, C. (2004). Improving pronoun resolution by incorporating coreferential information of candidates. In Proc. [find](https://scholar.google.com/scholar?q=Yang%2C+X%2E%2C+Su%2C+J%2E%2C+Zhou%2C+G%2E%2C+and+Tan%2C+C%2E+%282004%29%2E+Improving+pronoun+resolution+by+incorporating+coreferential+information+of+candidates%2E+In+Proc%2E) [our card](https://md.research.elamur.ai/free/papers/yang-2004-improving-pronoun-resolution-incorporating-coreferential/)
34. Zanzotto, F. M. and Moschitti, A. (2006). Automatic learning of textual entailments with cross-pair similarities. In Proceedings of Coling-ACL, Sydney, Australia. [find](https://scholar.google.com/scholar?q=Zanzotto%2C+F%2E+M%2E+and+Moschitti%2C+A%2E+%282006%29%2E+Automatic+learning+of+textual+entailments+with+cross%2Dpair+similarities%2E+In+Proceedings+of+Coling%2DACL%2C+Sydney%2C+Australia%2E)
35. Zelenko, D., Aone, C., and Richardella, A. (2003). Kernel methods for relation extraction. JMLR, 3(6):1083 – 1106. [find](https://scholar.google.com/scholar?q=Zelenko%2C+D%2E%2C+Aone%2C+C%2E%2C+and+Richardella%2C+A%2E+%282003%29%2E+Kernel+methods+for+relation+extraction%2E+JMLR%2C+3%286%29%3A1083+%E2%80%93+1106%2E)