Abstract—The Web is a lifestyle of this era. User searches
information on Web data by daily usage. The problem is that
when user browsing a Web page and interested in similar pages,
then an application is needed to find out related information
locations (web pages) called similar Web page advisor. It is
obvious that this task requires more than a Web search engine.
In this study, a simple text processing technique for English is
devised in order to rearrange the output of the Web search
engine. In other words, the HTML content of the Web pages on
the links suggested by Web search engine are further processed
and evaluated so that enhanced ranking of the top ten links is
presented to the user.
The output of the System is compared with the well-known
similar tool Chrome “similar Web pages” add-on application.
The average Cosine similarity of the original Web page and
suggested ten Web pages is considered. Our System overwhelms
Chrome “similar Web pages” add-on. Moreover, it is more
stable if different types of Web pages are considered.
Index Terms—Social recommendation, content analysis and
feature selection, text processing.
Metin Turan is with the İstanbul Commerce University, Küçükyalı,
İstanbul, Turkey (e-mail: mturan@ ticaret.edu.tr).
[PDF]
Cite: Metin Turan, "Enhancing Online Similar Web Pages Advisor with Support of Text Processing," International Journal of Future Computer and Communication vol. 6, no. 1, pp. 15-20, 2017.
