A Hypergraph Model for Computing Page Reputation on Web Collections

In this work we propose a representation of the web as a directed hypergraph, instead of a graph, where links can connect not only pairs of pages, but also pairs of disjoint sets of pages. In our model, the web hypergraph is derived from the web graph by dividing the set of pages into non-overlapping blocks and using the links between pages of distinct blocks to create hyperarcs. Each hyperarc connects a block of pages to a single page and is created with the goal of providing more reliable information for link analysis methods. We used the hypergraph structure to compute the reputation of web pages by experimenting hypergraph versions of two previously proposed link analysis methods, Pagerank and Indegree. We present experiments which indicate the hypergraph versions of Pagerank and Indegree produce better results when compared to their original graph versions.