Automatic indexing of online health resources for a French quality controlled gateway

  • Authors:
  • Aurélie Névéol;Alexandrina Rogozan;Stéfan Darmoni

  • Affiliations:
  • PSI Laboratory, CNRS FRE 2645, INSA de Rouen, Place Emile Blondel, BP 08, 76131 Mont-Saint-Aignan Cedex, France and CISMeF, Rouen University Hospital, and L@STICS, Rouen Medical School, 1, rue de ...;PSI Laboratory, CNRS FRE 2645, INSA de Rouen, Place Emile Blondel, BP 08, 76131 Mont-Saint-Aignan Cedex, France;PSI Laboratory, CNRS FRE 2645, INSA de Rouen, Place Emile Blondel, BP 08, 76131 Mont-Saint-Aignan Cedex, France and CISMeF, Rouen University Hospital, and L@STICS, Rouen Medical School, 1, rue de ...

  • Venue:
  • Information Processing and Management: an International Journal
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

The profusion of online resources calls for tools and methods to help Internet users find precisely what they are looking for. Quality controlled gateway CISMeF provides such services for health resources. However, the human cost of maintaining and updating the catalogue are increasingly high. This paper presents the automatic indexing system currently developed in the CISMeF team to be used as such for preliminary indexing, or after human reviewing for the final indexing. The system architecture, using the INTEX platform for MeSH term extraction is detailed. The results of a first evaluation tend to indicate that the automatic indexing strategy is relevant, as it achieves a precision comparable to that of other existing operational systems. Moreover, the system presented in this paper retrieves keyword/qualifier pairs as opposed to single terms, therefore providing a significantly more precise indexing. Further development and tests will be carried out in order to improve the coverage of the dictionaries, and validate the efficiency of the system in the indexers' everyday work.