Improving Link Analysis through Considering Hosts and Blocks

Authors:
Qiang Wang;Yan Liu;Jun Yong Luo;Jing Ning;Qing Yao
Affiliations:
Information Engineering Institute, Information Engineering University;Information Engineering Institute, Information Engineering University;Information Engineering Institute, Information Engineering University;Information Engineering Institute, Information Engineering University;Information Engineering Institute, Information Engineering University
Venue:
WI '06 Proceedings of the 2006 IEEE/WIC/ACM International Conference on Web Intelligence
Year:
2006

Citing 0
Cited 1

An algebraic multigrid solution of large hierarchical markovian models arising in web information retrieval

Network performance engineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

Link analysis has shown great potential in web structure mining. Most existing link analysis methods treat a web page as a node, a hyperlink as an edge in the web graph. However, in reality, a web page's importance is also affected by its semantic factor, the structure of WWW, and other factors. Therefore, traditional link analysis methods may not be so precise in reality. It is observed that the WWW has three layers: host layer, web page layer and host layer. Based on this concept, it is considered merging the host-page relationship and the page-block relationship to improve the result of link analysis. Experiment results demonstrate the effectiveness of our approach.