Updating pagerank with iterative aggregation
Section 5 contains conclusion and points out directions for future research.That is to say, the current iterate can be expressed as a linear combination of some of the first eigenvectors, combined with the Power method, up to the converge of the principal eigenvector.GQE is derived in this light and can be given as follows .In our study, we consider the neighborhood aggregation method as described in [12–14], since it is able to result in well-balanced aggregates of approximately equal size and provide a more regular coarsening throughout the automatic coarsening process [12, 29, 33].
Our aggregation strategies are based on the problem matrix by the current iterate can be interpreted as approximations to the stationary probability vector.
Isensee and Horton considered a kind of multilevel methods for the steady state solution of continuous-time and discrete-time Markov chains in [13, 14], respectively. proposed a multilevel adaptive aggregation method for calculating the stationary probability vector of Markov matrices in , as shown in their context, which is a special case of the adaptive smoothed aggregation  and adaptive algebraic multigrid methods  for sparse linear systems.
Updating pagerank with iterative aggregation comments