PowerWalk

Abstract

Most methods for Personalized PageRank (PPR) precompute and store all accurate PPR vectors, and at query time, return the ones of interest directly. However, the storage and computation of all accurate PPR vectors can be prohibitive for large graphs, especially in caching them in memory for real-time online querying. In this paper, we propose a distributed framework that strikes a better balance between offline indexing and online querying. The offline indexing attains a fingerprint of the PPR vector of each vertex by performing billions of "short" random walks in parallel across a cluster of machines. We prove that our indexing method has an exponential convergence, achieving the same precision with previous methods using a much smaller number of random walks. At query time, the new PPR vector is composed by a linear combination of related fingerprints, in a highly efficient vertex-centric decomposition manner. Interestingly, the resulting PPR vector is much more accurate than its offline counterpart because it actually uses more random walks in its estimation. More importantly, we show that such decomposition for a batch of queries can be very efficiently processed using a shared decomposition. Our implementation, PowerWalk, takes advantage of advanced distributed graph engines and it outperforms the state-of-the-art algorithms by orders of magnitude. Particularly, it responses to tens of thousands of queries on graphs with billions of edges in just a few seconds.

Efficient and Effective Similarity Search over Bipartite Graphs

“…There exists a large body of literature [10, 12, 18, 24, 38-40, 45, 47, 52, 57, 64, 65, 69, 81, 83, 85, 88] for single-source PPR queries. Among them, [10, 38-40, 52, 57, 69, 85, 88] rely on expensive matrix operations, [12,24,47,65] estimate PPR values using a large number of random walks, and several recent studies [45,81,83] combine deterministic graph traversal with random walks for improved efficiency. Particularly, in Ref.…”

Parallel-friendly personalized pageranks with operation shuffling

“…For example, some studies [37,25] utilize the reverse push operation as the main computation method. Also, some studies [31,22] find out that analyzing the structure of the graph can accelerate the computation of PPR.…”