Multiobjective Simulated Annealing Based Approach for Feature Selection in Anaphora Resolution

  • Asif Ekbal
  • Sriparna Saha
  • Olga Uryupina
  • Massimo Poesio
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7099)

Abstract

In this paper we propose a multiobjective simulated annealing based technique for anaphora resolution. There is no generally accepted metric for measuring the performance of anaphora resolution systems, and the existing metrics–MUC, B3 , CEAF, Blanc, among others–tend to reward significantly different behaviors. Systems optimized according to one metric tend to perform poorly with respect to other ones, making it very difficult to compare anaphora resolution systems, as clearly shown by the results of the SEMEVAL 2010 Task 1 on the Multilingual Coreference Resolution. One solution would be to find a single completely satisfactory metric, but its not clear whether this is possible and at any rate it is not going to happen any time soon. An alternative is to optimize models according to multiple metrics simultaneously. In this paper, we propose a multiobjective simulated annealing based technique to solve the feature selection problem of anaphora resolution by optimizing multiple objective functions. Experimental results show that the proposed approach performs superior in comparison to the previously developed multiobjective genetic algorithm based feature selection technique.

Keywords

Multiobjective Optimization Simulated Annealing Anaphora Resolution Feature Selection 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Bagga, A., Baldwin, B.: Algorithms for scoring coreference chains. In: LREC Workshop on Linguistic Coreference, pp. 563–566 (1998)Google Scholar
  2. 2.
    Bandyopadhyay, S., Saha, S., Maulik, U., Deb, K.: A simulated annealing based multi-objective optimization algorithm: AMOSA. IEEE Transactions on Evolutionary Computation 12(3), 269–283 (2008)CrossRefGoogle Scholar
  3. 3.
    Bengtson, E., Roth, D.: Understanding the value of features for coreference resolution. In: EMNLP (2008)Google Scholar
  4. 4.
    Deb, K.: Multi-objective Optimization Using Evolutionary Algorithms. John Wiley and Sons, Ltd., England (2001)MATHGoogle Scholar
  5. 5.
    Deb, K., Pratap, A., Agarwal, S., Meyarivan, T.: A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation 6(2), 181–197 (2002)CrossRefGoogle Scholar
  6. 6.
    Doddington, G., Mitchell, A., Przybocki, M., Ramshaw, L., Strassell, S., Weischedel, R.: The automatic content extraction (ACE) program-tasks, data, and evaluation. In: LREC (2000)Google Scholar
  7. 7.
    Hoste, V.: Optimization Issues in Machine Learning of Coreference Resolution. Ph.D. thesis, Antwerp University (2005)Google Scholar
  8. 8.
    Kirkpatrick, S., Gelatt, C.D., Vecchi, M.P.: Optimization by simulated annealing. Science 220, 671–680 (1983)MathSciNetCrossRefMATHGoogle Scholar
  9. 9.
    Luo, X.: On coreference resolution performance metrics. In: NAACL / EMNLP, Van-couver (2005)Google Scholar
  10. 10.
    Metropolis, N., Rosenbluth, A.W., Rosenbloth, M.N., Teller, A.H., Teller, E.: Equation of state calculation by fast computing machines. J. Chemical Physics 21(6), 1087–1092 (1953)CrossRefGoogle Scholar
  11. 11.
    Munson, A., Cardie, C., Caruana, R.: Optimizing to arbitrary NLP metrics using ensembleselection. In: HLT/EMNLP, pp. 539–546 (2005)Google Scholar
  12. 12.
    Ng, V., Cardie, C.: Improving machine learning approaches to coreference resolution. In: ACL, pp. 104–111 (2002)Google Scholar
  13. 13.
    Poon, H., Domingos, P.: Joint unsupervised coreference resolution with Markov logic. In: EMNLP (2008)Google Scholar
  14. 14.
    Pradhan, S., Marcus, M., Palmer, M., Ramshaw, L., Weischedel, R., Xue, N.: Conll-2011:Shared task on modeling unrestricted coreference in ontonotes, Portland, Oregon, USA, June23-24 (2011), http://www.cnts.ua.ac.be/conll/
  15. 15.
    Recasens, M., Hovy, E.: A Deeper Look into Features for Coreference Resolution. In: Lalitha Devi, S., Branco, A., Mitkov, R. (eds.) DAARC 2009. LNCS, vol. 5847, pp. 29–42. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  16. 16.
    Recasens, M., Hovy, E.: Blanc: Implementing the rand index for coreference evaluation. Natural Language Engineering (2011)Google Scholar
  17. 17.
    Recasens, M., Márquez, L., Sapena, E., Mart, M.A., Taul, M., Hoste, V., Poesio, M., Ver-rsley, Y.: Semeval-2010 task 1: Coreference resolution in multiple languages. In: SE-MEVAL 2010, Uppsala (2010)Google Scholar
  18. 18.
    Saha, S., Ekbal, A., Uryupina, O., Poesio, M.: Single and multi-objective optimization forfeature selection in anaphora resolution. In: IJCNLP (2011)Google Scholar
  19. 19.
    Soon, W.M., Ng, H.T., Lim, D.C.Y.: A machine learning approach to coreference resolution of noun phrases. Computational Linguistics 27(4), 521–544 (2001)CrossRefGoogle Scholar
  20. 20.
    Uryupina, O.: Knowledge Acquisition for Coreference Resolution. Ph.D. thesis, University of the Saarland (2007)Google Scholar
  21. 21.
    Uryupina, O.: Corry: a system for coreference resolution. In: SemEval (2010)Google Scholar
  22. 22.
    Veldhuizen, D.V., Lamont, G.: Multiobjective evolutionary algorithms: Analyzing the state-of-the-art. Evolutionary Computations 2, 125–1473 (2000)CrossRefGoogle Scholar
  23. 23.
    Versley, Y., Ponzetto, S.P., Poesio, M., Eidelman, V., Jern, A., Smith, J., Yang, X., Moschitti, A.: BART: a modular toolkit for coreference resolution. In: ACL/HLT, pp. 9–12 (2008)Google Scholar
  24. 24.
    Vilain, M., Burger, J., Aberdeen, J., Connolly, D., Hirschman, L.: A model-theoretic coreference scoring scheme. MUC 6, 45–52 (1995)CrossRefGoogle Scholar
  25. 25.
    Zhao, S., Ng, H.T.: Maximum metric score training for coreference resolution. In: COLING 2010 (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Asif Ekbal
    • 1
  • Sriparna Saha
    • 1
  • Olga Uryupina
    • 2
  • Massimo Poesio
    • 3
  1. 1.Department of Computer Science and EngineeringIIT PatnaIndia
  2. 2.Center for Mind/Brain SciencesUniversity of TrentoItaly
  3. 3.Language and Computation GroupUniversity of EssexUK

Personalised recommendations