Beruflich Dokumente
Kultur Dokumente
HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 71
Abstract— Lexeme is a prototype for advertising network that connects Web sites that want to host advertisements with
advertisers who want to run advertisements. Lexeme aims to implement better approaches for reaching and attracting target
customers by integrating semantic Web technology and enables computers to know what particular ads mean, to know what
particular Web sites are about, and to understand the relationships between them all. Advertising networks’ reliance on only the
keywords in the content results in displaying irrelevant and unappealing ads on the Web page. The Semantic-Based Advertising
Networks moves beyond simple keywords by understanding all the words on a page, and how they relate to one another. In
Lexeme, the description of ads and Web site content relies on the ontology that represents the conceptualization of the
knowledge domain. Advertiser defines the concept that corresponds to the product or services to sell in their ad along with
properties specifying the characteristics of the product and relationships with other concepts. The paper proposes a novel
approach for matching ads with Web site content using semantic Web technology, illustrated by Lexeme prototype.
Index Terms— Intelligent Web Services and Semantic Web, Ontology, Electronic commerce, Context Analysis and Indexing.
—————————— ——————————
1 INTRODUCTION
3. CONTENT CONCEPTUALIZATION For refinement, after choosing the parent concept the
In this section we explain in detail the content conceptua- system suggests a list of properties, expressed in a natural
lization process of both the ad and Web site. In our pro- language, which corresponds, here in our case, to the
posed model, advertiser describes the content of the ad "notebook" as “computer” product, namely: "is colored",
that relies on EAO ontology as a concept followed by a set "is priced for", "is dimensioned", "is manufactured by ", as
of attribute-value pairs, which is then interpreted and shown in figure 5. Advertiser selects a property and then
matched against different Web sites (publishers). When assigns to it, a specific value. In our example the price of
the advertiser registers, the system suggests a number of the product was set to "2000" and manufactured by
terms that may describe the ad depending on the content "Asus". That represents the value of the selected property
of the advertiser’s Web site and the ontology. In addition, "price" and the property "is manufactured by".
the advertiser may contribute in choosing a number of
terms that describes the content of the ad. For example,
in our case, “Notebook” term was supplied by the adver-
tiser.
Lexeme checks the EAO Ontology content and indenti-
fies the sub-tree in which these concepts belong to. Below
JOURNAL OF COMPUTING, VOLUME 2, ISSUE 10, OCTOBER 2010, ISSN 2151-9617
HTTPS://SITES.GOOGLE.COM/SITE/JOURNALOFCOMPUTING/
WWW.JOURNALOFCOMPUTING.ORG 74
were asked to gather some samples that they considered
to be related to the educational domain. We then made
sure that they were 100% strict XHTML pages and then
manually marked them up with RDFa. Java-RDFa 0.3
was used to extract RDF from Web sites. Jena comes with
a number of built-in functions.
Among the set of publishers returned by Lexeme
Fig. 5. Properties of the selected concept where it displayed the ad on, we selected the result
shown in the below figure.
The formulated conceptualization of the publisher’s
Web site is sent to Lexeme server in charge of analyzing it
and performing an http request to Lexeme application
server. The server analyzes the RDF-based response rep-
lied from the application server and extracts the informa-
tion corresponding to the matching ad and direct it to the
ad spaces placed on publisher's Web site.
ACKNOWLEDGMENT
The authors wish to thank Arwa Al-Tameem, Ghada Ab-
uguyan, May Abu Melah, Nora Al-Zaid, Nouf Al-Najran
for developing the prototype.
REFERENCES
[1] "isense" [Online]. Available: http//www.isense.net/. [Accessed Nov. 19,
2009].
[2] "new ways to target your customer" 2006. [Online]. Available:
http://www.emarketer.com. [Accessed Nov. 3, 2009].
[3] "online ad targeting: engaging the audience" 2006. [Online]. Available:
http//www.emarketer.com. [Accessed Nov. 14, 2009].
[4] "Over 875 million consumers have shopped online—the number of Inter-
net shoppers up to 40% in two years", Jan. 28, 2008. [Online]. Available:
http://en-
us.neilsen.com/main/news/news_releases/2008/jan/over_875_million_co
nsumers. [Accessed Nov. 16, 2009].
[5] D. Allemang, and J. Hendler, "Semantic web for the working ontologist:
effective modeling in rdfs and owl". Burlington, Massachusetts: Morgan
Kaufmann, 2008.
[6] G. Scheider and J.Evans, "new perspectives on the internet", 7th ed. Bos-
ton: Course Technology, 2008.
[7] J. Plummer, S. Rappaport, T. Hall, and R. Barocci, "The Online Advertising
Playbook". Hoboken, New Jersey: Wiley, 2007.
[8] J. Strickland, "How Web 3.0 will work," [Online]. Available:
http//computer.howstuffworks.com. [Accessed Oct. 31, 2009].
[9] L. Al-Safadi, N. Abdulateef, "Educational Advertising Ontology: A Do-
main-Dependent Ontology for Semantic Advertising Networks". Jounal of
Computer Sciences 6(9), 2010.