Please use this identifier to cite or link to this item:
Title: New approaches in web prefetching to improve content access by end users
Researcher: Venketesh, P
Guide(s): Venkatesan, R
Keywords: cache
Content Access
End Users
Fuzzy Logic
Web Prefetching
Upload Date: 1-Sep-2014
University: Anna University
Completed Date: n.d.
Abstract: The growth of Internet at a rapid pace with enormous number of users newlineand web services constantly demands good infrastructure to deliver the web newlinecontents to users with minimal delay Tremendous increase in the global traffic newlinedue to demands from large number of users strains the servers and network newlineresulting in poor quality of service availability reliability and latency perceived newlineby the users Web caching and prefetching provides effective mechanisms to newlinemitigate the user perceived latency This thesis is focused on the study of existing newlineprefetching mechanism and suggesting new approaches for effective prefetching newlinein web environment The goal of web prefetching is to download prefetch the newlinecontents and store it in local cache before user actually requests them It newlineminimizes the latency time perceived by users when accessing the content newlineThe thesis primarily focuses on two key aspects: newlineMethods to generate predictions that improve prefetching newlineactivity newlineMechanism to effectively manage the contents of cache regular newlineand prefetch by designing cache replacement scheme newlineWeb predictions can be generated at server proxy or client using newlinevariety of information depending on the location where they are implemented newlineServer based predictions consider access history of several users stored in a log newlinefile to generate predictions Client based predictions consider contents of web newlinepages accessed by a user to generate predictions newlineThe major contributions of this research work are as follows: newlineThe first part of thesis focuses on improving the client based newlinepredictions by designing two approaches NaïveBayes and Fuzzy Logic that newlineuses hyperlinks accessed by users to generate the predictions The hypertext newlineinformation associated with each hyperlink is used to compute its priority and newlinethen sorted highest to lowest to create prediction hint list Both prediction and newlineprefetching engine are implemented in the client machine focusing on browsing newlinebehavior of single or multiple users newline newline
Pagination: xvii,164p.
Appears in Departments:Faculty of Science and Humanities

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File170.64 kBAdobe PDFView/Open
02_certificate.pdf1.72 MBAdobe PDFView/Open
03_abstract.pdf86.79 kBAdobe PDFView/Open
04_acknowledgement.pdf693.58 kBAdobe PDFView/Open
05_contents.pdf107.15 kBAdobe PDFView/Open
06_chapter 1.pdf152.72 kBAdobe PDFView/Open
07_chapter 2.pdf233.46 kBAdobe PDFView/Open
08_chapter 3.pdf283.99 kBAdobe PDFView/Open
09_chapter 4.pdf388.15 kBAdobe PDFView/Open
10_chapter 5.pdf692.32 kBAdobe PDFView/Open
11_chapter 6.pdf88.33 kBAdobe PDFView/Open
12_references.pdf158.06 kBAdobe PDFView/Open
13_publications.pdf89.79 kBAdobe PDFView/Open
14_vitae.pdf51.02 kBAdobe PDFView/Open

Items in Shodhganga are protected by copyright, with all rights reserved, unless otherwise indicated.