Probabilistic Learning Approaches for Indexing and Retrieval with the TREC-2 Collection

  • Citation-Key:
    Fuhr/etal:94
  • Title:
    Probabilistic Learning Approaches for Indexing and Retrieval with the TREC-2 Collection
  • Author(s):
    N. Fuhr
    U. Pfeifer
    C. Bremkamp
    M. Pollmann
    C. Buckley
  • In:
    • Citation-Key:
      TREC-2
    • Title:
      The Second Text REtrieval Conference (TREC-2)
    • Editor(s):
      D. Harman
    • Publisher:
      National Institute of Standards and Technology
    • In:
      The Second Text REtrieval Conference (TREC-2)
    • Year:
      1994
    Classification(s):
    H.3.3
    General terms:
    experimentation
  • Page(s):
    67--74
  • Year:
    1994

Abstract:


In this paper, we describe the application of probabilistic models for indexing and retrieval with the TREC-2 collection. This database consists of about a million documents (2 gigabytes of data) and 100 queries (50 routing and 50 adhoc topics). For document indexing, we use a description-oriented approach which exploits relevance feedback data in order to produce a probabilistic indexing with single terms as well as with phrases. With the adhoc queries, we present a new query term weighting method based on a training sample of other queries. For the routing queries, the RPI model is applied which combines probabilistic indexing with query term weighting based on query-specific feedback data. The experimental results of our approach show very good performance for both types of queries.
Classification(s):
H.3.3, G.1.2, H.3.1
Subject descriptor(s):
least squares approximation, indexing methods, retrieval models
Keywords:
query expansion, probabilistic models

Fulltext as PS