Advertisement

Two-Way Parsimonious Classification Models for Evolving Hierarchies

  • Mostafa Dehghani
  • Hosein Azarbonyad
  • Jaap Kamps
  • Maarten Marx
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9822)

Abstract

There is an increasing volume of semantically annotated data available, in particular due to the emerging use of knowledge bases to annotate or classify dynamic data on the web. This is challenging as these knowledge bases have a dynamic hierarchical or graph structure demanding robustness against changes in the data structure over time. In general, this requires us to develop appropriate models for the hierarchical classes that capture all, and only, the essential solid features of the classes which remain valid even as the structure changes. We propose hierarchical significant words language models of textual objects in the intermediate levels of hierarchies as robust models for hierarchical classification by taking the hierarchical relations into consideration. We conduct extensive experiments on richly annotated parliamentary proceedings linking every speech to the respective speaker, their political party, and their role in the parliament. Our main findings are the following. First, we define hierarchical significant words language models as an iterative estimation process across the hierarchy, resulting in tiny models capturing only well grounded text features at each level. Second, we apply the resulting models to party membership and party position classification across time periods, where the structure of the parliament changes, and see the models dramatically better transfer across time periods, relative to the baselines.

Keywords

Significant words language models Evolving hierarchies 

Notes

Acknowledgments

This research is funded in part by Netherlands Organization for Scientific Research through the Exploratory Political Search project (ExPoSe, NWO CI # 314.99.108), and by the Digging into Data Challenge through the Digging Into Linked Parliamentary Data project (DiLiPaD, NWO Digging into Data # 600.006.014).

References

  1. 1.
    Blei, D.M., Lafferty, J.D.: Dynamic topic models. In: ICML, pp. 113–120 (2006)Google Scholar
  2. 2.
    Chen, M., Weinberger, K.Q., Blitzer, J.: Co-training for domain adaptation. In: NIPS ’24, pp. 2456–2464 (2011)Google Scholar
  3. 3.
    Dehghani, M.: Significant words representations of entities. In: SIGIR 2016 (2016)Google Scholar
  4. 4.
    Dehghani, M., Azarbonyad, H., Kamps, J., Marx, M.: Generalized group profiling for content customization. In: CHIIR 2016, pp. 245–248 (2016)Google Scholar
  5. 5.
    Dehghani, M., Azarbonyad, H., Kamps, J., Hiemstra, D., Marx, M.: Luhn revisited: significant words language models. In: The Proceedings of The ACM International Conference on Information and Knowledge Management (CIKM’16) (2016)Google Scholar
  6. 6.
    Dehghani, M., Azarbonyad, H., Kamps, J., Marx, M.: On horizontal and vertical separation in hierarchical text classification. In: The Proceedings of ACM SIGIR International Conference on the Theory of Information Retrieval (ICTIR’16) (2016)Google Scholar
  7. 7.
    Diermeier, D., Godbout, J.-F., Yu, B., Kaufmann, S.: Language and ideology in congress. Br. J. Polit. Sci. 42(1), 31–55 (2012)CrossRefGoogle Scholar
  8. 8.
    Dumais, S., Chen, H.: Hierarchical classification of web content. In: SIGIR, pp. 256–263 (2000)Google Scholar
  9. 9.
    Frank, J.R., Kleiman-Weiner, M., Roberts, D.A., Voorhees, E.M., Soboroff, I.: Evaluating stream filtering for entity profile updates in trec 2012, 2013 and 2014. In: TREC 2014 (2012)Google Scholar
  10. 10.
    Hiemstra, D., Robertson, S., Zaragoza, H.: Parsimonious language models for information retrieval. In: SIGIR 2004, pp. 178–185 (2004)Google Scholar
  11. 11.
    Hirst, G., Riabinin, Y., Graham, J., Boizot-Roche, M.: Text to ideology or text to party status? From Text Polit. Positions: Text Anal. Across Disciplines 55, 93–116 (2014)CrossRefGoogle Scholar
  12. 12.
    Kim, D.-K., Voelker, G., Saul, L.K.: A variational approximation for topic modeling of hierarchical corpora. In: ICML, pp. 55–63 (2013)Google Scholar
  13. 13.
    Luhn, H.P.: The automatic creation of literature abstracts. IBM J. Res. Dev. 2(2), 159–165 (1958)MathSciNetCrossRefGoogle Scholar
  14. 14.
    Marx, M., Schuth, A.: Dutchparl: a corpus of parliamentary documents in dutch. In: DIR Workshop, pp. 82–83 (2010)Google Scholar
  15. 15.
    McCallum, A., Rosenfeld, R., Mitchell, T.M., Ng, A.Y.: Improving text classification by shrinkage in a hierarchy of classes. In: ICML 1998, pp. 359–367 (1998)Google Scholar
  16. 16.
    Ogilvie, P., Callan, J.: Hierarchical language models for XML component retrieval. In: Fuhr, N., Lalmas, M., Malik, S., Szlávik, Z. (eds.) INEX 2004. LNCS, vol. 3493, pp. 224–237. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  17. 17.
    Oh, H.-S., Choi, Y., Myaeng, S.-H.: Text classification for a large-scale taxonomy using dynamically mixed local and global models for a node. In: Clough, P., Foley, C., Gurrin, C., Jones, G.J.F., Kraaij, W., Lee, H., Mudoch, V. (eds.) ECIR 2011. LNCS, vol. 6611, pp. 7–18. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  18. 18.
    PoliticalMashup. Political mashup project (2015). http://search.politicalmashup.nl/. Netherlands Organization for Scientific Research
  19. 19.
    Sebastiani, F.: Machine learning in automated text categorization. ACM Comput. Surv. 34(1), 1–47 (2002)CrossRefGoogle Scholar
  20. 20.
    Silla Jr., C.N., Freitas, A.A.: A survey of hierarchical classification across different application domains. Data Min. Knowl. Discov. 22(1–2), 31–72 (2011)MathSciNetCrossRefzbMATHGoogle Scholar
  21. 21.
    Song, Y., Roth, D.: On dataless hierarchical text classification. In: AAAI, pp. 1579–1585 (2014)Google Scholar
  22. 22.
    Teh, Y.W., Jordan, M.I., Beal, M.J., Blei, D.M.: Hierarchical dirichlet processes. J. Am. Stat. Assoc. 101(476), 1566–1581 (2006)MathSciNetCrossRefzbMATHGoogle Scholar
  23. 23.
    Xue, G.-R., Dai, W., Yang, Q., Yu, Y.: Topic-bridged plsa for cross-domain text classification. In: SIGIR 2008, pp. 627–634 (2008)Google Scholar
  24. 24.
    Yao, L., Mimno, D., McCallum, A.: Efficient methods for topic model inference on streaming document collections. In: SIGKDD, pp. 937–946 (2009)Google Scholar
  25. 25.
    Yu, B., Kaufmann, S., Diermeier, D.: Classifying party affiliation from political speech. J. Inf. Technol. Politics 5(1), 33–48 (2008)CrossRefGoogle Scholar
  26. 26.
    Zavitsanos, E., Paliouras, G., Vouros, G.A.: Non-parametric estimation of topic hierarchies from texts with hierarchical dirichlet processes. J. Mach. Learn. Res. 12, 2749–2775 (2011)MathSciNetzbMATHGoogle Scholar
  27. 27.
    Zhai, C., Lafferty, J.: A study of smoothing methods for language models applied to ad hoc information retrieval. In: SIGIR 2001, pp. 334–342 (2001)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Mostafa Dehghani
    • 1
  • Hosein Azarbonyad
    • 2
  • Jaap Kamps
    • 1
  • Maarten Marx
    • 2
  1. 1.Institute for Logic, Language and ComputationUniversity of AmsterdamAmsterdamThe Netherlands
  2. 2.Informatics InstituteUniversity of AmsterdamAmsterdamThe Netherlands

Personalised recommendations