A software tool for the steady-state analysis of Google-like stochastic matrices

Authors:
Tuğrul Dayar;Gökçe Nil Noyan
Affiliations:
Bilkent University, Bilkent, Ankara, Turkey;Undersecretariat for Defence Industries, Balgat, Ankara, Turkey
Venue:
Proceedings of the Fourth International ICST Conference on Performance Evaluation Methodologies and Tools
Year:
2009

Citing 4
Cited 0

The anatomy of a large-scale hypertextual Web search engine

WWW7 Proceedings of the seventh international conference on World Wide Web 7
Comparison of Partitioning Techniques for Two-Level Iterative Solvers on Large, Sparse Markov Chains

SIAM Journal on Scientific Computing
An Implementation of Tarjan's Algorithm for the Block Triangularization of a Matrix

ACM Transactions on Mathematical Software (TOMS)
Extrapolation methods for accelerating PageRank computations

WWW '03 Proceedings of the 12th international conference on World Wide Web

Quantified Score

Hi-index	0.00

Visualization

Abstract

The problem of computing the steady-state vector of positive stochastic matrices which are convex combinations of sparse, nonnegative matrices possibly having zero rows with appropriately chosen rank-1 matrices is addressed by a software tool. The dynamically changing matrices used by the Google search engine in ranking web pages are among the largest of such matrices. Ranking pages amounts to solving for the steady-state vectors of these matrices. The tool implements the power, quadratically extrapolated power, and iterative methods based on various block partitionings, including those with block triangular form and with triangular blocks obtained using cutsets.