An artificial neural network approach for classifying e-Commerce Web pages

Anagnostopoulos, I; Kouzas, G; Anagnostopoulos, C; Psoroulas, I; Vergados, D; Loumos, V; Kayafas, E

dc.contributor.author	Anagnostopoulos, I	en
dc.contributor.author	Kouzas, G	en
dc.contributor.author	Anagnostopoulos, C	en
dc.contributor.author	Psoroulas, I	en
dc.contributor.author	Vergados, D	en
dc.contributor.author	Loumos, V	en
dc.contributor.author	Kayafas, E	en
dc.date.accessioned	2014-03-01T02:42:12Z
dc.date.available	2014-03-01T02:42:12Z
dc.date.issued	2003	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/30854
dc.relation.uri	http://www.scopus.com/inward/record.url?eid=2-s2.0-1442351317&partnerID=40&md5=5c4ed72a1d6da02620b524c750b99a16	en
dc.relation.uri	http://www.informatik.uni-trier.de/~ley/db/conf/appinf/appinf2003.html#AnagnostopoulosKAPVLK03	en
dc.subject	Artificial Neural Network	en
dc.subject	e-commerce	en
dc.subject	Information system	en
dc.subject	Web page classification	en
dc.subject.other	Algorithms	en
dc.subject.other	Classification (of information)	en
dc.subject.other	Information management	en
dc.subject.other	Neural networks	en
dc.subject.other	Portals	en
dc.subject.other	Servers	en
dc.subject.other	Information systems	en
dc.subject.other	Web page classification	en
dc.subject.other	Electronic commerce	en
dc.title	An artificial neural network approach for classifying e-Commerce Web pages	en
heal.type	conferenceItem	en
heal.publicationDate	2003	en
heal.abstract	In this paper, an information system capable of identifying and categorizing e-commerce web pages is proposed, on the basis of information filtering combined with a Artificial Neural Network (ANN). It includes term transformation techniques along with pattern recognition methods for content identification. The information representation techniques are based on a multi-dimensional descriptor vector with 432 word stems collected either manually or in an automatic way. This vector is called e-Commerce Descriptor Vector (e-CDV) and when applied in a proper way according to a proposed technique, it assigns a unique profile to every tested web page. The created profile is then considered as a pattern, and it is categorized by the system according the Business Media Framework (BMF). The system classifies twelve classes of web pages. Eleven of them follow the concepts of the BMF, while the last represents all the other web pages.	en
heal.journalName	IASTED International Multi-Conference on Applied Informatics	en
dc.identifier.volume	21	en
dc.identifier.spage	237	en
dc.identifier.epage	242	en