Skip to Main content Skip to Navigation
Journal articles

Nouvelles perspectives en extraction d'information

Abstract : The aim of Information Extraction (IE) consists in recognizing and extracting a specific set of information from a corpus of texts, and structuring it in a predefined format. IE has fostered an increasing interest in the last decade, and is likely to lead to industrial applications in a close future. After a short presentation of the principles of the technology, this paper describes researches performed by our group. An operational system, designed for analysis of car crash reports is presented. Then we show how IE techniques can be used in new tasks of document processing: semantic encoding, reading assistance, composite document structuring. Finally we present some work in semantics aimed to improve the performances of nowadays systems.
Complete list of metadatas

Cited literature [11 references]  Display  Hide  Download

https://halshs.archives-ouvertes.fr/halshs-00009485
Contributor : Bernard Victorri <>
Submitted on : Wednesday, March 8, 2006 - 5:06:30 PM
Last modification on : Saturday, March 28, 2020 - 2:20:14 AM
Document(s) archivé(s) le : Saturday, April 3, 2010 - 10:43:50 PM

Identifiers

  • HAL Id : halshs-00009485, version 1

Citation

Michel Dupont, Jean-Marc Vuillaume, Bernard Victorri, Patrice Enjalbert, Yann Mathet, et al.. Nouvelles perspectives en extraction d'information. Revue des Sciences et Technologies de l'Information - Série TSI : Technique et Science Informatiques, Lavoisier, 2002, 1 (21), pp.37-63. ⟨halshs-00009485⟩

Share

Metrics

Record views

1079

Files downloads

1271