Pagerank Increase under Different Collusion Topologies

We study the impact of collusion –nepotistic linking– in a Web graph in terms of Pagerank. We prove a bound on the Pagerank increase that depends both on the reset probability of the random walk ε and on the original Pagerank of the colluding set. In particular, due to the power law distribution of Pagerank, we show that highly-ranked Web sites do not benefit that much from collusion.

• Type: Conference
• Year: 2005
• Where: AIRWEB
• Authors: Ricardo A. Baeza-Yates, Carlos Castillo, Vicente López
• Keywords: AIRWEB 2005, Collusion –nepotistic Linking–, Internet Technology, Original Pagerank, Pagerank Increase