Paper Info
Title
A Ranking Algorithm via Changing Markov Probability Matrix Based on Distribution Factor
Abstract
With the rapid growth of the web, it will become more and more difficult to provide relevant information to the users to cater to their needs. The web structure mining plays an important role in this approach. There are two classic ranking algorithms HITS and PageRank commonly used in web structure mining. These two algorithms treat all links equally while assigning rank scores. This paper provides a new ranking algorithm via changing the Markov probability matrix based on distributed factor. This algorithm assigns rank scores based on the similarity of web pages instead of equal assignment. Our experiment results show that our algorithm performs better than the standard PageRank.
Year
DOI
Venue
2008
10.1109/FSKD.2008.312
FSKD (5)
Keywords
Field
DocType
equal assignment,classic ranking algorithms hits,standard pagerank,distribution factor,experiment result,markov probability matrix,assigning rank score,ranking algorithm,web structure mining,web page,new ranking algorithm,rank score,information retrieval,markov processes,hits,probability distribution,web pages,data mining,algorithm design and analysis,mathematical model
Learning to rank,Data mining,PageRank,Structure mining,Markov process,Ranking,Ranking SVM,Web page,Computer science,Markov chain,Algorithm
Conference
Citations 
PageRank 
References 
1
0.36
6
Authors
4
Name
Order
Citations
PageRank
Xianchao Zhang131339.57
Xinxin Fan2165.10
Xinyue Liu39215.42
Hong-yu Li483.58