Empirical Study of Classification Models for Web Page Categorization

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F14%3A00220119" target="_blank" >RIV/68407700:21230/14:00220119 - isvavai.cz</a>
Result on the web
<a href="http://ebooks.iospress.nl/volumearticle/37218" target="_blank" >http://ebooks.iospress.nl/volumearticle/37218</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.3233/978-1-61499-421-3-290" target="_blank" >10.3233/978-1-61499-421-3-290</a>

Result language
angličtina
Original language name
Empirical Study of Classification Models for Web Page Categorization
Original language description
We describe one part of a web page content classification system used for contextual advertising. The system classifies the content of a web page to one of many predefined ad categories, i.e. performing text classification [2,8,10]. Our goal is to identify the best performing classification model for Czech language. We present a comprehensive comparison study of selected models, text representations, and feature selection techniques on a collection of datasets with different numbers of documents, numbers of categories, and varying category sizes. We conclude the work with the recommendation for the best performing models.
Czech name
—
Czech description
—

Publication year
2014
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Frontiers in Artificial Intelligence and Applications
ISBN
978-1-61499-420-6
ISSN
0922-6389
e-ISSN
—
Number of pages
10
Pages from-to
290-299
Publisher name
IOS Press
Place of publication
Amsterdam
Event location
Praha
Event date
Aug 18, 2014
Type of event by nationality
EUR - Evropská akce
UT code for WoS article
000350218400030

Similar results(10)