Improving Retrieval Effectiveness by Automatically Creating Multiscaled Links between Text and Pictures

Nicolas Malandain 1 Mauro Gaio 1, 2 Jacques Madelaine 3
1 GREYC - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen
2 LIUPPA - Laboratoire Informatique de l'Université de Pau et des Pays de l'Adour
3 Equipe AMACC - Laboratoire GREYC - UMR6072
GREYC - Groupe de Recherche en Informatique, Image, Automatique et Instrumentation de Caen
Abstract : This paper describes a method to improve retrieval of composite documents (text and graphic) by creating a set of internal links. We propose the concept of granularity to add structure for this given set of typed semantic links. That is obtained by a multi-scaling processing. We propose three major classes based on the capacity of a link to include or to be included by others one. Global links are obtained with a classical IR methods. A computational model allows an automatic extraction of the textual information units contained in the text source of the global links. In our geographic corpus the units denotes georefenced entities. A semantic representation of these entities is proposed that allows further cooperation with processing of the graphical part of the document.
Document type :
Conference papers
Complete list of metadatas
https://hal-univ-pau.archives-ouvertes.fr/hal-01912868
Contributor : Gaelle Chancerel <>
Submitted on : Monday, November 5, 2018 - 5:17:56 PM
Last modification on : Sunday, April 7, 2019 - 3:00:39 PM

Identifiers

Collections

UNIV-PAU | LIUPPA | GREYC | GREYC-AMACC | COMUE-NORMANDIE | ENSICAEN | UNICAEN

Citation

Nicolas Malandain, Mauro Gaio, Jacques Madelaine. Improving Retrieval Effectiveness by Automatically Creating Multiscaled Links between Text and Pictures. Document Recognition and Retrieval VIII, San Jose, CA, USA, January 20, 2001, 2001, Unknown, Unknown Region. pp.89-98, ⟨10.1117/12.410826⟩. ⟨hal-01912868⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

10

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD