Personalized PageRank expresses link-based page quality around userselected pages in a similar way as PageRank expresses quality over the entire web. Existing personalized PageRank algorithms can, however, serve online queries only for a restricted choice of pages. In this paper we achieve full personalization by a novel algorithm that precomputes a compact database; using this database, it can serve online responses to arbitrary user-selected personalization. The algorithm uses simulated random walks; we prove that for a fixed error probability the size of our database is linear in the number of web pages. We justify our estimation approach by asymptotic worst-case lower bounds: we show that on some sets of graphs, exact personalized PageRank values can only be obtained from a database of size quadratic in the number of vertices. Furthermore, we evaluate the precision of approximation experimentally on the Stanford WebBase graph.

To exploit the similarity information hidden in the hyperlink structure of the web, this paper introduces algorithms scalable to graphs with billions of vertices on a distributed architecture. The similarity of multi-step neighborhoods of vertices are numerically evaluated by similarity functions including SimRank [18], a recursive refinement of cocitation; PSimRank, a novel variant with better theoretical characteristics; and the Jaccard coefficient, extended to multi-step neighborhoods. Our methods are presented in a general framework of Monte Carlo similarity search algorithms that precompute an index database of random fingerprints, and at query time, similarities are estimated from the fingerprints. The performance and quality of the methods were tested on the Stanford Webbase [17] graph of 80M pages by comparing our scores to similarities extracted from the ODP directory [24]. Our experimental results suggest that the hyperlink structure of vertices within four to five steps provide more adequate information for similarity search than singlestep neighborhoods.

No abstract