Weighted PageRank Algorithm

  • Authors:
  • Affiliations:
  • Venue:
  • CNSR '04 Proceedings of the Second Annual Conference on Communication Networks and Services Research
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

With the rapid growth of the Web, users get easily lost inthe rich hyper structure. Providing relevant information tothe users to cater to their needs is the primary goal of websiteowners. Therefore, .nding the content of the Web andretrieving the usersý interests and needs from their behaviorhave become increasingly important.Web mining is usedto categorize users and pages by analyzing the usersý behavior,the content of the pages, and the order of the URLsthat tend to be accessed in order. Web structure miningplays an important role in this approach. Two page rankingalgorithms, HITS and PageRank, are commonly usedin web structure mining. Both algorithms treat all linksequally when distributing rank scores. Several algorithmshave been developed to improve the performance of thesemethods. The Weighted PageRank algorithm (WPR), an extensionto the standard PageRank algorithm, is introducedin this paper. WPR takes into account the importance ofboth the inlinks and the outlinks of the pages and distributesrank scores based on the popularity of the pages. The resultsof our simulation studies show that WPR performs betterthan the conventional PageRank algorithm in terms of returninglarger number of relevant pages to a given query.