Downloads: 116 | Views: 211
Comparative Studies | Computer Science & Engineering | India | Volume 3 Issue 6, June 2014
Comparative Study of Web Structure Mining Techniques for Links and Image Search
Rashmi Sharma [9] | Kamaljit Kaur [5]
Abstract: Web mining is a vast area and growing very rapidly. It employs text; audio; video; contents and images from World Wide Web. The World Wide Web contains huge number of web pages and a lot of information available within web pages. When a user put a query to the search engine; it generally returns a large amount of information in response to users query. To retrieve relevant information from Web pages; web mining performs various web structure mining techniques. This paper proposed the hybrid technique of Weighted PageRank based on Visit of Links and Fuzzy K-Means algorithms which are applied on the search result. Fuzzy K-Means algorithm is used to group the given data into clusters and Weighted PageRank is used to re-rank the data according to the visit of links to taken in the account. We have extracted the relevant information such as image links; images and total hyperlinks from a web links using the hybrid approach. In an existing work; previous approach PageRank with K-Means has done on only text or URL not on links and has several limitations as compared to Weighted PageRank and Fuzzy K-Means algorithm. In this paper; we have search on links and images and provide quality search to users and applied valid parameters like execution time; recall; precision and f-measure on our proposed method and compared with previous approach and conclude the result. The principle idea of overall our method is to provide the fast and more relevant results in response to user requirements and can be seen as future of web mining.
Keywords: PageRank, Weighted PageRank based on Visit of Links, K-Means and Fuzzy K-Means
Edition: Volume 3 Issue 6, June 2014,
Pages: 2076 - 2082