A comparison of text retrieval models

Howard R. Turtle, W. Bruce Croft

Research output: Contribution to journalArticle

58 Citations (Scopus)

Abstract

Many retrieval models have been proposed as the basis of text retrieval systems. The three main classes that have been investigated are the exact-match, vector space and probabilistic models. The retrieval effectiveness of strategies based on these models has been evaluated experimentally, but there has been little in the way of comparison in terms of their formal properties. In this paper we introduce a recent form of the probabilistic model based on inference networks, and show how the vector space and exact-match can be described in this framework. Differences between these models can be explained as differences in the estimation of probabilities, both in the initial search and during relevance feedback.

Original languageEnglish (US)
Pages (from-to)279-290
Number of pages12
JournalComputer Journal
Volume35
Issue number3
DOIs
StatePublished - Jun 1992

Fingerprint

Text Retrieval
Statistical Models
Vector spaces
Space Simulation
Probabilistic Model
Retrieval
Vector Space Model
Relevance Feedback
Vector space
Model
Model-based
Feedback
comparison

ASJC Scopus subject areas

  • Molecular Biology
  • Statistics and Probability
  • Computational Mathematics
  • Development
  • Management, Monitoring, Policy and Law
  • Demography
  • Computer Science(all)

Cite this

A comparison of text retrieval models. / Turtle, Howard R.; Croft, W. Bruce.

In: Computer Journal, Vol. 35, No. 3, 06.1992, p. 279-290.

Research output: Contribution to journalArticle

Turtle, HR & Croft, WB 1992, 'A comparison of text retrieval models', Computer Journal, vol. 35, no. 3, pp. 279-290. https://doi.org/10.1093/comjnl/35.3.279
Turtle, Howard R. ; Croft, W. Bruce. / A comparison of text retrieval models. In: Computer Journal. 1992 ; Vol. 35, No. 3. pp. 279-290.
@article{f99cc3a365d2497fa110a71faf509a1b,
title = "A comparison of text retrieval models",
abstract = "Many retrieval models have been proposed as the basis of text retrieval systems. The three main classes that have been investigated are the exact-match, vector space and probabilistic models. The retrieval effectiveness of strategies based on these models has been evaluated experimentally, but there has been little in the way of comparison in terms of their formal properties. In this paper we introduce a recent form of the probabilistic model based on inference networks, and show how the vector space and exact-match can be described in this framework. Differences between these models can be explained as differences in the estimation of probabilities, both in the initial search and during relevance feedback.",
author = "Turtle, {Howard R.} and Croft, {W. Bruce}",
year = "1992",
month = "6",
doi = "10.1093/comjnl/35.3.279",
language = "English (US)",
volume = "35",
pages = "279--290",
journal = "Computer Journal",
issn = "0010-4620",
publisher = "Oxford University Press",
number = "3",

}

TY - JOUR

T1 - A comparison of text retrieval models

AU - Turtle, Howard R.

AU - Croft, W. Bruce

PY - 1992/6

Y1 - 1992/6

N2 - Many retrieval models have been proposed as the basis of text retrieval systems. The three main classes that have been investigated are the exact-match, vector space and probabilistic models. The retrieval effectiveness of strategies based on these models has been evaluated experimentally, but there has been little in the way of comparison in terms of their formal properties. In this paper we introduce a recent form of the probabilistic model based on inference networks, and show how the vector space and exact-match can be described in this framework. Differences between these models can be explained as differences in the estimation of probabilities, both in the initial search and during relevance feedback.

AB - Many retrieval models have been proposed as the basis of text retrieval systems. The three main classes that have been investigated are the exact-match, vector space and probabilistic models. The retrieval effectiveness of strategies based on these models has been evaluated experimentally, but there has been little in the way of comparison in terms of their formal properties. In this paper we introduce a recent form of the probabilistic model based on inference networks, and show how the vector space and exact-match can be described in this framework. Differences between these models can be explained as differences in the estimation of probabilities, both in the initial search and during relevance feedback.

UR - http://www.scopus.com/inward/record.url?scp=0003136577&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0003136577&partnerID=8YFLogxK

U2 - 10.1093/comjnl/35.3.279

DO - 10.1093/comjnl/35.3.279

M3 - Article

AN - SCOPUS:0003136577

VL - 35

SP - 279

EP - 290

JO - Computer Journal

JF - Computer Journal

SN - 0010-4620

IS - 3

ER -