Criteria for Cluster-Based Personalized Search

We study personalized web ranking algorithms based on the existence of document clusterings. Motivated by the topic sensitive page ranking of Haveliwala [20], we develop and implement an efficient “local-cluster” algorithm by extending the web search algorithm of Achlioptas, Fiat, Karlin and McSherry [10]. We propose some formal criteria for evaluating such personalized ranking algorithms and provide some preliminary experiments in support of our analysis. Both theoretically and experimentally, our algorithm differs significantly from Topic Sensitive PageRank.