Small noise limits of Markov chains and the PageRank
By: Vivek S Borkar, S Sowmya, Raghavendra Tripathi
Potential Business Impact:
Makes Google search results better and faster.
We recall the classical formulation of PageRank as the stationary distribution of a singularly perturbed irreducible Markov chain that is not irreducible when the perturbation parameter goes to zero. Specifically, we use the Markov chain tree theorem to derive explicit expressions for the PageRank. This analysis leads to some surprising results. These results are then extended to a much more general class of perturbations that subsume personalized PageRank. We also give examples where even simpler formulas for PageRank are possible.
Similar Papers
On Algorithmic Robustness of Corrupted Markov Chains
Probability
PageRank resists changes, keeping results accurate.
Fully personalized PageRank and algebraic methods to distribute a random walker
Social and Information Networks
Changes how important websites appear online.
On Limiting Probability Distributions of Higher Order Markov Chains
Probability
Predicts long-term behavior of complex systems.