Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/24580
Title: Improving Precision In Ranking Semantic Associations
Researcher: Viswanathan V
Guide(s): Ilango Krishnamurthi
Keywords: genetics
intelligence analysis
pharmaceutical research
Precision
Ranking
Semantic Associations
Upload Date: 3-Sep-2014
University: Anna University
Completed Date: n.d.
Abstract: Accessing relevant information from the Web has become difficult newlinedue to the explosive growth of information on the Web Many users try to newlineacquire this information by using search engine but search engine based newlinesystems locate only documents based on the keywords or key phrases While newlineconsidering data on the Web different entities can be related in multiple ways newlinethat cannot be predefined But in the Semantic Web the RDF data model newlinecaptures the meaning of an entity by specifying its relationship with other newlineentities newlineAt present many applications such as intelligence analysis newlinegenetics pharmaceutical research and flight security require more complex newlinerelationships than simple direct relationships between entities Semantic newlineAssociation is a sequence of complex relationships between entities in a newlineknowledge base represented as a graph Searching semantic relationships newlineamong the entities like people places and events from the semantic web is an newlineessential component in the future While searching semantic associations in newlineRDF graph the result containing multiple paths connecting two entities is newlineperceived Each path has different meanings depending on the type of newlinerelationship in which some of them may be relevant while others may be newlineirrelevant to the users according to their perspective In the proposed newlinemethods irrelevant paths can be filtered using a suitable methodology while newlinediscovering the paths connecting entities or these irrelevant paths can be newlineranked lower during the ranking process In finding semantic associations newlinethe study consists of three proposed methods which help the users in newlineimproving the precision in a most appropriate manner newlineThe first one is ranking semantic association paths based on the newlineusers domain of interest using personalization in context specification In this newlineapproach the users interest level in various domains called semantic web newlineusage context is captured from the web browsing history of the users by newlineusing the personalization mechanism and it is incorporated in the ranking newlineformula newline newline
Pagination: xvi,102p
URI: http://hdl.handle.net/10603/24580
Appears in Departments:Faculty of Science and Humanities

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File251.77 kBAdobe PDFView/Open
02_certificate.pdf2.95 MBAdobe PDFView/Open
03_abstract.pdf63.34 kBAdobe PDFView/Open
04_acknowledgement.pdf59.87 kBAdobe PDFView/Open
05_contents.pdf87 kBAdobe PDFView/Open
06_chapter 1.pdf563.35 kBAdobe PDFView/Open
07_chapter 2.pdf300.52 kBAdobe PDFView/Open
08_chapter 3.pdf134.91 kBAdobe PDFView/Open
09_chapter 4.pdf1.71 MBAdobe PDFView/Open
10_chapter 5.pdf1.08 MBAdobe PDFView/Open
11_chapter 6.pdf1.88 MBAdobe PDFView/Open
12_chapter 7.pdf75.06 kBAdobe PDFView/Open
13_appendix.pdf775.07 kBAdobe PDFView/Open
14_references.pdf123.79 kBAdobe PDFView/Open
15_publications.pdf57.79 kBAdobe PDFView/Open
16_vitae.pdf56.45 kBAdobe PDFView/Open


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