On The World Wide Web Search Engine Returns The Results Of Fuzzy Clustering Study | Posted on:2003-01-27 | Degree:Master | Type:Thesis | Country:China | Candidate:X P Chen | Full Text:PDF | GTID:2208360065460052 | Subject:Computer application technology | Abstract/Summary: | PDF Full Text Request | Today,search engines are the most commonly used tools for Web information retrieval. However,their current status is still far from user's satisfaction.lt includes:(1) The content that search engine returns is a enormous flat bill (information overloading question);(2) The items return with search engine are not the content that user requisite in deed(low precision question)This paper presents a fuzzy (soft) clustering algorithm HTSC (Hyperlink-Text based Soft Clustering) using a mixed similarity metric of document content and inter-document hyperlinks,for clustering Web search results from a search engine in order to help users find relevant Web information more easily. The main contributions of this paper include the following:(1) An effect method for computing inter-document similarities based on content and link analysis;(2) Identifying the advantages of fuzzy clustering methods in comparison with normal clustering methods,and presenting a fuzzy (soft) clustering algorithm HTSC base on a mixed similarity metric of content and link;(3) Theoretic analysis and preliminary experiments of the algorithm. | Keywords/Search Tags: | Web information retrieval, search engine, clustering, fuzzy clustering, similarity, content analysis, link analysis | PDF Full Text Request | Related items |
| |
|