Automatic Profile Reformulation Using a Local Document Analysis

  • Authors:
  • Anis Benammar;Gilles Hubert;Josiane Mothe

  • Affiliations:
  • -;-;-

  • Venue:
  • Proceedings of the 24th BCS-IRSG European Colloquium on IR Research: Advances in Information Retrieval
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

User profiles are more and more used in information retrieval system in order to assist users in finding relevant information. Profiles are continuously updated to evolve at the same time the user information need does. In this paper we present a reformulation strategy used to automatically update the profile content. In a first stage, a local document set is computed from the search results. In a second stage, the local set is analyzed to select the terms to add to the profile expression. Experiments have been performed on an extract from the OHSUMED database to evaluate the effectiveness of the adaptation process.