Elamur Research⌕
☰
home / papers / 2014 / A Generic Anaphora Resolution Engine for Indian Languages

A Generic Anaphora Resolution Engine for Indian Languages

Summary

Most Indian languages lack parsers and other advanced preprocessing tools, and existing multilingual anaphora resolvers still depend on plug-in modules for each language. The authors build a single pronominal anaphora resolution engine meant to work across Indian languages without language-specific components. Input text receives only shallow processing. Morphological analysis, POS tagging and chunking are mandatory, while clause boundary identification and NER are optional sources of constraint features. The main obstacle is that pronouns mark gender very differently across languages: Tamil separates "he" and "she," but Hindi and Bengali use one third-person form for both. The engine addresses this with detailed morphological analysis that extracts person-number-gender (PNG) information, plus heuristic agreement rules that filter candidate antecedents. A CRF classifier then selects the antecedent. The system was evaluated on Bengali, Hindi (Indo-Aryan) and Tamil (Dravidian), and the authors describe the results as encouraging. The work matters because it shows that rich morphology can replace deep parsing as a basis for anaphora resolution in low-resource language families.

Key points

  • Evaluation languages are Bengali, Hindi and Tamil, which covers both the Indo-Aryan and Dravidian families. These are the same three languages used in the ICON 2011 anaphora resolution tool contest.
  • The learning method is CRFs, with CRF++ (Taku Kudo) cited as the toolkit. Features are drawn from shallow-parsed text rather than full syntactic parses.
  • A survey table of 11 Indian languages shows that all of them distinguish singular from plural in pronouns, but gender marking varies:
    • Telugu, Kannada, Malayalam and Tamil mark gender.
    • Hindi and several other Indo-Aryan languages do not.
    • Assamese distinguishes only animate from inanimate.
  • PNG agreement heuristics, built on in-depth morphological analysis, replace the per-language agreement modules found in earlier multilingual systems such as Vasisth, BART and GuiTAR adaptations.
  • The supplied excerpt omits the experimental section, so exact precision, recall and F-scores and corpus sizes are not included here. Consult the full paper for those numbers.

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

Topics: Coreference resolution

Cited by

6 papers of our library cite it (of the 140673 whose bibliographies we read); 1 of them have a page here, 5 not yet.

Cite

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

@inproceedings{devi2014generic,
  title = {{A Generic Anaphora Resolution Engine for Indian Languages}},
  author = {Devi, Sobha Lalitha and Ram, Vijay Sundar and Rao, Pattabhi RK},
  booktitle = {Proceedings of COLING 2014, the 25th International Conference on Computational Linguistics: Technical Papers},
  year = {2014},
  url = {https://aclanthology.org/C14-1172/}
}

References

41 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Akilandeswari A., and Sobha Lalitha Devi. (2013). Conditional Rando m Fields Based Pronominal Resolution in Tamil. International Journal on Computer Science and Engineering, Vol. 5 Issue 6 pp 601–610. find
  2. Akilandeswari A, Bakiyavathi T and Sobha Lalitha Dev i, (2012), "atu Difficult Prono minal in Tamil", In Proceedings of Lrec 2012, Istanbul Aone C., and McKee D. (1993).A Language-Independent Anaphora Resolution System for Understanding Multilingual Texts. In proceeding of ACL 1993, pp 156-163. find
  3. Aone C., and Bennett S. (1995). Evaluating automated and manual acquisit ion of anaphora resolution strategies. In: 33 rd Annual Meeting of the Association for Computational Linguistics, pp. 122-129. find
  4. Balaji J., Geetha T.V., Ranjan i Parthasarathi R., Karky M . (2012).Two -Stage Bootstrapping for Anaphora Res olution In: Proceedings of COLIN G 2012, pp 507– 516. find
  5. Carbonell J. G., and Brown R. D. (1988). Anaphora resolution: A mu lti-strategy approach. In: 12 th International Conference on Computational Linguistics, 1988, pp. 96-101. find
  6. Charniak, E., and Elsner, M. (2009). EM Works for Pronoun Anaphora Resolution. In Proceedings of the Conference of the European Chapter of the Association for Computational Linguistics (EACL 2009), Athens, Greece. find our card
  7. Daelemans, W. and van den Bosch, A. (2005). Memory -based language processing. Cambridge University Press, Camb ridge Dagan I., and Itai. find
  8. A. (1990). Automatic processing of large corpora for the resolution of anaphora references. In: 13th conference on Computational linguistics, Vol. 3, Helsinki, Fin land, pp.330-332. find our card
  9. Dakwale. P., Mujadia. V., Sharma. D.M. (2013). A Hybrid Approach for Anaphora Resolution in Hindi. In: Proc of International Joint Conference on Natural Language Processing , Nagoya, Japan, pp.977–981. find our card
  10. Dutta. K., Prakash. N. and Kaushik. S. (2008). Resolving Prono minal Anaphora in Hindi using Hobbs ‟ algorith m,” Web Journal of Formal Computation and Cognitive Linguistics, Issue 10, 2008. find
  11. Li., F., Shi., S., Chen., Y., and Lv, X. (2008). Chinese Pronominal Anaphora Resolution Based on Conditional Random Fields. In: International Conference on Computer Science and Software Engineering , Washington, DC, USA, pp. 731-734. find
  12. Hendrickx I., Hoste V., and Daelemans W. (2008). Semantic and syntactic features for Dutch coreference resolution. In Gelbukh A. (Ed.), CIC Ling-2008 conference, Vo l. 4919 LNCS, Berlin, Sp ringer Verlag, pp. 731-734. find
  13. Hobbs J. (1978). Resolving pronoun references. Lingua 44, pp. 339-352. find our card
  14. Grosz, B. J. (1977). The representation and use of focus in dialogue understanding. Technical Report 151, SRI International, 333 Ravenswood Ave, Menlo Park, Ca. 94025. find
  15. Joshi A. K., and Kuhn S. (1979). Centered logic: The role of entity centered sentence representation in natural language inferencing. In: International Joint Conference on Artificial Intelligence. find
  16. Joshi A. K., and Weinstein S. (1981). Control o f inference: Role of some aspects of discourse structure – centering”, In: International Joint Conference on Artificial Intelligence, pp. 385-387. find
  17. Kennedy, C., Boguraev, B. (1996) Anaphora for Everyone: Prono minal Anaphora Resolution without a Parser. In: 16th International Conference on Computational Linguistics COLIN G’96 , Copenhagen, Den mark, pp. 113–118. find
  18. Lappin S., and Leass H. J. (1994). An algorith m for pronominal anaphora resolution. Computational Linguistics 20 (4), pp. 535-561. find
  19. McCallu m A., and Wellner. B. (2003). Toward conditional models of identity uncertainty wit h application to proper noun coreference. In Proceedings of the IJCAI Workshop on Information Integration on the Web , pp. 79– 84. find
  20. McCarthy, J. F. and Lehnert, W. G. (1995). Using decision trees for coreference resolution. In C. Mellish (Ed.), Fourteenth International Conference on Artificial Intelligence, pp. 1050-1055 find
  21. Mitkov R. (1998). Robust pronoun resolution with limited knowledge. In: 17th International Conference on Computational Linguistics (COLING‟ 98/ACL‟98), Montreal, Canada, pp. 869-875. find our card
  22. Mitkov, R. (1997). "Factors in anaphora resolution: they are not the only things that matter. A case study based on two different approaches". In Proceedings of the ACL'97/EACL'97 workshop on Operational factors in practical, robust anaphora resolution, Madrid, Spain. find our card
  23. Murthy K.N., Sobha L, Muthuku mari B. (2007). Pronominal Resolution in Tamil Using Machine Learning A pproach. The First Workshop on Anaphora Resolution (WAR I) , Ed Christer Johansson, Cambridge Scholars Publishing, 15 Angerton Gardens, Newcastle, NE5 2JA, UK, pp.39-50. find
  24. Ng V., and Card ie C. (2002). Improv ing machine learning approaches to coreference resolution. In. 40th Annual Meeting of the Association for Computational Linguistics, pp. 104-111. find
  25. Prasad R., and Strube,M.,(2000). Discourse Salience and Pronoun Resolution in Hindi, Penn Working Papers in Linguistics, Vol 6.3, pp. 189-208. find
  26. Ram, R.V.S. and Sobha Lalitha Dev i. (2013)."Pronominal Resolution in Tamil Using Tree CRFs", In Proceedings of 6th Language and Technology Conference, Human Language Technologies as a challenge for Computer Science and Linguistics - 2013, Po znan, Po land Recasens M., M\`arquez L., Sapena E., Mart´I… find
  27. Recasens M., Hovy E. (2009). A Deeper Look into Features for Coreference Resolution. Lalitha Dev i, S., Bra nco, A. and Mitkov, R. (eds.), Anaphora Processing and Applications (DAARC 2009) , LNAI 5847, Springer-Verlag Berlin Heidelberg, pp 535-561. find
  28. Rich, E. and LuperFoy S., (1988) An architecture for anaphora resolution. In: Proceedings of the Second Conference on Applied Natural Language Processing, Austin, Texas. find our card
  29. Senapati A., Garain U. (2013). Gu iTA R-based Pronominal Anaphora Resolution in Bengal. In: Proceedings of 51st Annual Meeting of the Association for Computational Linguistics, Sofia, Bulgaria pp 126–130. find
  30. Sikdar U.K, Ekbal A., Saha S., Uryupina O., Poesio M. (2013). Adapting a State-of-the-art Anaphora Resolution System for Resource-poor Language. In proceedings of International Joint Conference on Natural Language Processing, Nagoya, Japan pp 815– 821. find
  31. Sobha L. and Patnaik B. N. (2000). Vasisth: An Anaphora Resolution System for Indian Languages. In Proceedings of International Conference on Artificial and Computational Intelligence for Decision, Control and Automation in Engineering and Industrial Applications, Monastir, Tunisia. find
  32. Sobha L. and Patnaik,B.N. (2002). Vasisth: An anaphora resolution system for Malayalam and Hindi. In Proceedings of Symposium on Translation Support Systems. find
  33. Sobha L. (2007). Resolution of Prono minals in Tamil. Co mputing Theory and Applicat ion, The IEEE Computer Society Press, Los Alamitos, CA, pp. 475-79. find
  34. Sobha L., Pralayankar P. (2008). Algorithm fo r Anaphor Resolution in Sanskrit. In Proceedings of 2nd Sanskrit Computational Linguistics Symposium, Brown University, USA, 2008. find
  35. Sobha, Lalitha Devi., Vijay Sundar Ram and Pattabhi RK Rao. (2011). Resolution of Prono minal Anaphors u sing Linear and Tree CRFs. In. 8th DAARC, Faro, Portugal, 2011. find
  36. Sobha L., Sivaji Bandyopadhyay, Vijay Sundar Ram R., and Akilandeswari A. (2011). NLP T ool Contest @ICON2011 on Anaphora Resolution in Indian Languages. In: Proceedings of ICON 2011. find
  37. Soon W. H., Ng, and Lim D. (2001). A machine learn ing approach to coreference resolution of noun phrases. Computational Linguistics 27 (4), pp.521-544. find
  38. Strube, M. and Hahn U., (1999). Functional centering: Grounding referential coherence in information structure. Computational Linguistics, 25(3) pp 309–344 find our card
  39. Taku Kudo. 2005. CRF++, an open source toolkit for CRF, http://crfpp.sourceforge.net . ↗
  40. Uppalapu. B., and Sharma, D.M. (2009). Pronoun Resolution For Hindi. In: Proceedings of 7th Discourse Anaphora and Anaphor Resolution Colloquium (DAARC 09) , pp. 123-134. find
  41. Wick M., Singh S., and McCallu m A. (2012). A Discriminative Hierarchical Model fo r Fast Coreferenc e At Large Scale. In : Proceedings of ACL 2012. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/devi-2014-generic-anaphora-resolution-engine-indian/