DocumentCode :
564852
Title :
Analysing anchor links to enhance the web snippet clustering technique
Author :
Omara, F.A. ; Amoon, M. ; El-Fishawy, N.A. ; El-kazaz, S.
Author_Institution :
Fac. of Comput. & Inf., Cairo Univ., Cairo, Egypt
fYear :
2012
fDate :
14-16 May 2012
Abstract :
New generations of Search Engines aim to focus on user´s needs rather than user´s queries. This means personalization of the returned results to suit the user needs and expectations. In this paper, we believe that in order to enhance the personalization process, we have to enhance the data that used as an input to the personalization algorithm. According to this work, we will analyze the web structure (anchor text) to group all anchors refers to each search results and use it to enhance the search results returned search engines and then enhance the total personalization technique. In this paper, we introduce a comparative study between clustering search results using snippets returned from search engines and clustering search results using enriched snippet. The enriched snippet is a collection of the returned snippet and a very precise description of the original page of this snippet called the anchor text ofthat page. The summation of the enriched snippet and the anchor text will be called enriched snippet. Anchor text is used in a web page to point to a related document/picture/media application. Many existing approaches are based on the use of anchor-text contained in the anchor tag and analyze them to get out the information about an associated web page. In the information retrieval field, many search engines analyze the anchor text and use it as a main factor in its ranking algorithm. It is known that anchor text is the most important factor in Google ranking algorithm. In this paper, we show that these enriched snippets are powerful than the normal snippet and theses enriched snippets give higher precision to the resulted clusters.
Keywords :
Internet; pattern clustering; search engines; text analysis; Google ranking algorithm; Web page; Web snippet clustering technique; Web structure; anchor links; anchor tag; anchor text; clustering search results; personalization algorithm; search engines; total personalization technique; Clustering algorithms; Databases; Educational institutions; Java; Search engines; Sun; Web pages;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Informatics and Systems (INFOS), 2012 8th International Conference on
Conference_Location :
Cairo
Print_ISBN :
978-1-4673-0828-1
Type :
conf
Filename :
6236574
Link To Document :
بازگشت