Skip to Main content Skip to Navigation
Theses

Une approche générique pour l'analyse croisant contenu et usage des sites Web par des méthodes de bipartitionnement

Abstract : In this thesis, we propose a new approach WCUM (Web Content and Usage Mining based approach) for linking content analysis to usage analysis of a website to better understand the general behavior of the web site visitors. This work is based on the use of the block clustering algorithm CROKI2 implemented by two different strategies of optimization that we compared through experiments on artificially generated data. To mitigate the problem of determination of the number of clusters on rows and columns, we suggest to generalize the use of some indices originally proposed to evaluate the partitions obtained by clustering algorithms to evaluate bipartitions obtained by simultaneous clustering algorithms. To evaluate the performance of these indices on data with biclusters structure, we proposed an algorithm for generating artificial data to perform simulations and validate the results. Experiments on artificial data as well as on real data were realized to estimate the efficiency of the proposed approach.
Document type :
Theses
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-00516367
Contributor : Abes Star :  Contact
Submitted on : Thursday, September 9, 2010 - 1:51:09 PM
Last modification on : Saturday, December 21, 2019 - 3:52:57 AM
Long-term archiving on: : Friday, December 10, 2010 - 2:31:18 AM

File

2010CNAM0694_0_0.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-00516367, version 1

Collections

Citation

Malika Charrad. Une approche générique pour l'analyse croisant contenu et usage des sites Web par des méthodes de bipartitionnement. Linguistique. Conservatoire national des arts et metiers - CNAM; École Nationale des Sciences de l'Informatique (La Manouba, Tunisie), 2010. Français. ⟨NNT : 2010CNAM0694⟩. ⟨tel-00516367⟩

Share

Metrics

Record views

961

Files downloads

2156