Fast random walk with restart and its applications

Fast random walk with restart and its applications (2006)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Hanghang Tong

Venue:	In ICDM ’06: Proceedings of the 6th IEEE International Conference on Data Mining
Citations:	85 - 14 self

BibTeX

@INPROCEEDINGS{Tong06fastrandom,
    author = {Hanghang Tong},
    title = {Fast random walk with restart and its applications},
    booktitle = {In ICDM ’06: Proceedings of the 6th IEEE International Conference on Data Mining},
    year = {2006},
    pages = {613--622},
    publisher = {IEEE Computer Society}
}

Years of Citing Articles

Bookmark

OpenURL

Abstract

How closely related are two nodes in a graph? How to compute this score quickly, on huge, disk-resident, real graphs? Random walk with restart (RWR) provides a good relevance score between two nodes in a weighted graph, and it has been successfully used in numerous settings, like automatic captioning of images, generalizations to the “connection subgraphs”, personalized PageRank, and many more. However, the straightforward implementations of RWR do not scale for large graphs, requiring either quadratic space and cubic pre-computation time, or slow response time on queries. We propose fast solutions to this problem. The heart of our approach is to exploit two important properties shared by many real graphs: (a) linear correlations and (b) blockwise, community-like structure. We exploit the linearity by using low-rank matrix approximation, and the community structure by graph partitioning, followed by the Sherman-Morrison lemma for matrix inversion. Experimental results on the Corel image and the DBLP dabasets demonstrate that our proposed methods achieve significant savings over the straightforward implementations: they can save several orders of magnitude in pre-computation and storage cost, and they achieve up to 150x speed up with 90%+ quality preservation. 1

Citations

2546	Indexing by latent semantic analysis - Deerwester, Dumais, et al. - 1990
1969	T.: The pagerank citation ranking: Bringing order to the web - Page, Brin, et al. - 1999
1822	Principal Component Analysis - Jolliffe - 2002
968	On spectral clustering: Analysis and an algorithm - Ng, Jordan, et al. - 2001
651	Community structure in social and biological networks - Girvan, Newman - 2002
438	Semi-Supervised Learning Using Gaussian Fields and Harmonic Functions - Zhu, Ghahramani, et al. - 2003
386	Topic-sensitive pagerank - Haveliwala - 2002
382	Learning with local and global consistency - ZHOU, BOUSQUET, et al. - 2004
272	Scaling personalized web search - Jeh, Widom - 2002
240	C.: Gaussian Processes for Machine Learning - Rasmussen, Williams - 2006
219	Information-theoretic co-clustering - Dhillon, Mallela, et al. - 2003
214	Efficient identification of web communities - Flake, Lawrence, et al. - 2000
204	SimRank: A Measure of Structural-Context Similarity - Jeh, Widom - 2002
142	ObjectRank: Authority-based keyword search in databases - Balmin, Hristidis, et al. - 2004
127	Exploiting the Block Structure of the Web for Computing PageRank - Kamvar, Haveliwala, et al. - 2003
123	Fast computation of low rank matrix approximations - Achlioptas, McSherry - 2001
99	Fast discovery of connection subgraphs - Faloutsos, McCurley, et al. - 2004
90	Automatic multimedia cross-modal correlation discovery - Pan, Yang, et al. - 2004
62	Towards scaling fully personalized PageRank: Algorithms, lower bounds, and experiments - Fogaras, Rácz, et al.
56	Manifoldranking based image retrieval - He, Li, et al. - 2004
45	Center-piece subgraphs: Problem definition and fast solutions - Tong, Faloutsos - 2006
35	Neighborhood Formation and Anomaly Detection in Bipartite Graphs - Sun, Qu, et al. - 2005
29	Relational link-based ranking - Geerts, Mannila, et al. - 2004
27	Parallel multilevel k-way partitioning for irregular graphs - Karypis, Kumar - 1999
24	BANKS: Browsing and keyword searching in relational databases - Aditya, Bhalotia, et al.
15	Vagelis Hristidis, and Yannis Papakonstantinou. Objectrank: authority-based keyword search in databases - Balmin - 2004
10	Inverting a sum of matrices - Piegorsch, Casella - 1990
8	Hyung-Jeong Yang, Christos Faloutsos, and Pinar Duygulu. Automatic multimedia crossmodal correlation discovery - Pan - 2004
6	Mingjing Li, HongJiang Zhang, Hanghang Tong, and Changshui Zhang. Manifold-ranking based image retrieval - He - 2004
4	Gaurav Bhalotia, Soumen Chakrabarti, Arvind Hulgeri, Charuta Nakhe, Parag Parag - Aditya - 2002
3	Heikki Mannila, and Evimaria Terzi. Relational link-based ranking - Geerts - 2004