On CORI Results Merging

  • Ilya Markov
  • Avi Arampatzis
  • Fabio Crestani
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7814)

Abstract

Score normalization and results merging are important components of many IR applications. Recently MinMax—an unsupervised linear score normalization method—was shown to perform quite well across various distributed retrieval testbeds, although based on strong assumptions. The CORI results merging method relaxes these assumptions to some extent and significantly improves the performance of MinMax. We parameterize CORI and evaluate its performance across a range of parameter settings. Experimental results on three distributed retrieval testbeds show that CORI significantly outperforms state-of-the-art results merging and score normalization methods when its parameter goes to infinity.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Arguello, J., Callan, J., Diaz, F.: Classification-based resource selection. In: Proceedings of the ACM CIKM, pp. 1277–1286. ACM (2009)Google Scholar
  2. 2.
    Callan, J.: Distributed Information Retrieval. In: Advances in Information Retrieval, ch. 5, pp. 127–150. Kluwer Academic Publishers (2000)Google Scholar
  3. 3.
    Fernández, M., Vallet, D., Castells, P.: Using historical data to enhance rank aggregation. In: Proceeding of the ACM SIGIR, pp. 643–644 (2006)Google Scholar
  4. 4.
    Lee, J.H.: Analyses of multiple evidence combination. In: Proceedings of the ACM SIGIR, pp. 267–276. ACM (1997)Google Scholar
  5. 5.
    Markov, I., Arampatzis, A., Crestani, F.: Unsupervised linear score normalization revisited. In: Proceedings of the ACM SIGIR, pp. 1161–1162 (2012)Google Scholar
  6. 6.
    Shokouhi, M., Zobel, J.: Robust result merging using sample-based score estimates. ACM Trans. Inf. Syst. 27(3), 1–29 (2009)CrossRefGoogle Scholar
  7. 7.
    Si, L., Callan, J.: Relevant document distribution estimation method for resource selection. In: Proceedings of the ACM SIGIR, pp. 298–305 (2003)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Ilya Markov
    • 1
  • Avi Arampatzis
    • 2
  • Fabio Crestani
    • 1
  1. 1.University of LuganoLuganoSwitzerland
  2. 2.Democritus University of ThraceXanthiGreece

Personalised recommendations