Advertisement

Is Document Frequency Important for PRF?

  • Stéphane Clinchant
  • Eric Gaussier
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6931)

Abstract

We introduce in this paper a new heuristic constraint for PRF models, referred to as the Document Frequency (DF) constraint, which is validated through a series of experiments with an oracle. We then analyze, from a theoretical point of view, state-of-the-art PRF models according to their relation with this constraint. This analysis reveals that the standard mixture model for PRF in the language modeling family does not satisfy the DF constraint on the contrary to several recently proposed models. Lastly, we perform tests, which further validate the constraint, with a simple family of tf-idf functions based on a parameter controlling the satisfaction of the DF constraint.

Keywords

Mixture Model Information Retrieval Relevance Feedback Query Expansion Document Frequency 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Amati, G., Carpineto, C., Romano, G., Bordoni, F.U.: Fondazione Ugo Bordoni at TREC 2003: robust and web track (2003)Google Scholar
  2. 2.
    Clinchant, S., Gaussier, E.: Information-based models for ad hoc IR. In: SIGIR 2010, Conference on Research and Development in Information Retrieval (2010)Google Scholar
  3. 3.
    Collins-Thompson, K.: Reducing the risk of query expansion via robust constrained optimization. In: CIKM 2009 Conference on Information and Knowledge Management (2009)Google Scholar
  4. 4.
    Collins-Thompson, K., Callan, J.: Estimation and use of uncertainty in pseudo-relevance feedback. In: SIGIR 2007 (2007)Google Scholar
  5. 5.
    Crabtree, D.W., Andreae, P., Gao, X.: Exploiting underrepresented query aspects for automatic query expansion. In: SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD 2007 (2007)Google Scholar
  6. 6.
    Dillon, J.V., Collins-Thompson, K.: A unified optimization framework for robust pseudo-relevance feedback algorithms. In: CIKM, pp. 1069–1078 (2010)Google Scholar
  7. 7.
    Fang, H., Tao, T., Zhai, C.: A formal study of information retrieval heuristics. In: SIGIR 2004: Conference on Research and Development in Information Retrieval (2004)Google Scholar
  8. 8.
    Lavrenko, V., Croft, W.B.: Relevance based language models. In: SIGIR 2001:Conference on Research and Development in Information Retrieval (2001)Google Scholar
  9. 9.
    Lee, K.S., Croft, W.B., Allan, J.: A cluster-based resampling method for pseudo-relevance feedback. In: SIGIR 2008, Conference on Research and Development in Information Retrieval (2008)Google Scholar
  10. 10.
    Lv, Y., Zhai, C.: Adaptive relevance feedback in information retrieval. In: Conference on Information and Knowledge Management, CIKM 2009 (2009)Google Scholar
  11. 11.
    Lv, Y., Zhai, C.: A comparative study of methods for estimating query language models with pseudo feedback. In: CIKM 2009: Conference on Information and Knowledge Management, pp. 1895–1898 (2009)Google Scholar
  12. 12.
    Lv, Y., Zhai, C.: Positional relevance model for pseudo-relevance feedback. In: SIGIR 2010, Conference on Research and Development in Information Retrieval (2010)Google Scholar
  13. 13.
    Seo, J., Croft, W.B.: Geometric representations for multiple documents. In: SIGIR 2010: Conference on Research and Development in Information Retrieval (2010)Google Scholar
  14. 14.
    Xu, Z., Akella, R.: A new probabilistic retrieval model based on the dirichlet compound multinomial distribution. In: SIGIR 2008: Conference on Research and Development in Information Retrieval (2008)Google Scholar
  15. 15.
    Zhai, C., Lafferty, J.: Model-based feedback in the language modeling approach to information retrieval. In: CIKM 2001 (2001)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Stéphane Clinchant
    • 1
    • 2
  • Eric Gaussier
    • 2
  1. 1.Xerox Research Center EuropeMeylanFrance
  2. 2.UMR 5217/AMA teamLIG Université de GrenobleFrance

Personalised recommendations