Experiments on Average Distance Measure

  • Vincenzo Della Mea
  • Gianluca Demartini
  • Luca Di Gaspero
  • Stefano Mizzaro
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3936)

Abstract

ADM (Average Distance Measure) is an IR effectiveness metric based on the assumptions of continuous relevance and retrieval. This paper presents some novel experimental results on two different test collections: TREC 8, re-assessed on 4-levels relevance judgments, and TREC 13 TeraByte collection. The results confirm that ADM correlation with standard measures is high, even when using less data, i.e., few documents.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Della Mea, V., Di Gaspero, L., Mizzaro, S.: Evaluating ADM on a four-level relevance scale document set from NTCIR. In: Proceedings of NTCIR Workshop 4 Meeting - Supplement Vol. 2, pp. 30–38 (2004)Google Scholar
  2. 2.
    Della Mea, V., Mizzaro, S.: Measuring retrieval effectiveness: A new proposal and a first experimental validation. JASIS&T 55(6), 530–543 (2004)CrossRefGoogle Scholar
  3. 3.
    Mizzaro, S.: A new measure of retrieval effectiveness (Or: What’s wrong with precision and recall). In: Ojala, T. (ed.) International Workshop on Information Retrieval (IR 2001), pp. 43–52 (2001)Google Scholar
  4. 4.
    Sormunen, E.: Liberal relevance criteria of TREC - Counting on negligible documents? In: Jarvelin, K., Beaulieu, M., Baeza-Yates, R., Myaeng, S. (eds.) Proceedings of the 25th ACM SIGIR Conference, August 11-15, pp. 324–330 (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Vincenzo Della Mea
    • 1
  • Gianluca Demartini
    • 1
  • Luca Di Gaspero
    • 2
  • Stefano Mizzaro
    • 1
  1. 1.Dept. of Mathematics and Computer ScienceItaly
  2. 2.Dept. of Electrical, Management and Mechanical EngineeringUniversity of UdineUdineItaly

Personalised recommendations