Publication Type

Journal Article

Version

submittedVersion

Publication Date

10-2015

Abstract

As Personalized PageRank has been widely leveraged for ranking on a graph, the efficient computation of Personalized PageRank Vector (PPV) becomes a prominent issue. In this paper, we propose FastPPV, an approximate PPV computation algorithm that is incremental and accuracy-aware. Our approach hinges on a novel paradigm of scheduled approximation: the computation is partitioned and scheduled for processing in an “organized” way, such that we can gradually improve our PPV estimation in an incremental manner and quantify the accuracy of our approximation at query time. Guided by this principle, we develop an efficient hub-based realization, where we adopt the metric of hub length to partition and schedule random walk tours so that the approximation error reduces exponentially over iterations. In addition, as tours are segmented by hubs, the shared substructures between different tours (around the same hub) can be reused to speed up query processing both within and across iterations. Given the key roles played by the hubs, we further investigate the problem of hub selection. In particular, we develop a conceptual model to select hubs based on the two desirable properties of hubs—sharing and discriminating, and present several different strategies to realize the conceptual model. Finally, we evaluate FastPPV over two real-world graphs, and show that it not only significantly outperforms two state-of-the-art baselines in both online and offline phrases, but also scales well on larger graphs. In particular, we are able to achieve near-constant time online query processing irrespective of graph size.

Keywords

Accuracy-aware, Incremental enhancement, Hub selection, Scheduled approximation, Personalized PageRank

Discipline

Databases and Information Systems | Numerical Analysis and Scientific Computing

Research Areas

Data Science and Engineering

Publication

VLDB Journal

Volume

24

Issue

5

First Page

655

Last Page

679

ISSN

1066-8888

Identifier

10.1007/s00778-014-0376-8

Publisher

Springer Verlag (Germany)

Additional URL

https://doi.org/10.1007/s00778-014-0376-8

Download

Share

COinS