Abstract
In MESIA project, we tackled the problem of reach an acceptable degree of accuracy in finding links that contains relevant documents to the user consult. The current prototype interacts with the web-pages of the Madrid Local Government (www.comadrid.es) through AltaVista. The filtering process consists in the consult modification, the structural and linguistic analysis of the links texts and the use of the user model and past experience when available to obtain the final answer. The system evaluation has started not only measured by time and space, but also by the retrieved answers set precision.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sanchez Torralba, P., García-Serrano, A. (2001). MESIA: A Web Site Assistant for Document Filtering. In: Bouzeghoub, M., Kedad, Z., Métais, E. (eds) Natural Language Processing and Information Systems. NLDB 2000. Lecture Notes in Computer Science, vol 1959. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-45399-7_34
Download citation
DOI: https://doi.org/10.1007/3-540-45399-7_34
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41943-3
Online ISBN: 978-3-540-45399-4
eBook Packages: Springer Book Archive