edu.uci.ics.jung.algorithms.scoring

Class PageRankWithPriors<V,E>

java.lang.Object

edu.uci.ics.jung.algorithms.scoring.AbstractIterativeScorer<V,E,S>

edu.uci.ics.jung.algorithms.scoring.AbstractIterativeScorerWithPriors<V,E,java.lang.Double>

edu.uci.ics.jung.algorithms.scoring.PageRankWithPriors<V,E>

All Implemented Interfaces:

VertexScorer<V,java.lang.Double>, IterativeContext

Direct Known Subclasses:

KStepMarkov, PageRank

public class PageRankWithPriors<V,E>
extends AbstractIterativeScorerWithPriors<V,E,java.lang.Double>

A generalization of PageRank that permits non-uniformly-distributed random jumps. The 'vertex_priors' (that is, prior probabilities for each vertex) may be thought of as the fraction of the total 'potential' that is assigned to that vertex at each step out of the portion that is assigned according to random jumps (this portion is specified by 'alpha').

See Also:

"Algorithms for Estimating Relative Importance in Graphs by Scott White and Padhraic Smyth, 2003", PageRank

Constructor Summary

Constructors

Constructor and Description
PageRankWithPriors(Hypergraph<V,E> graph, com.google.common.base.Function<E,? extends java.lang.Number> edge_weights, com.google.common.base.Function<V,java.lang.Double> vertex_priors, double alpha) Creates an instance with the specified graph, edge weights, vertex priors, and 'random jump' probability (alpha).
PageRankWithPriors(Hypergraph<V,E> graph, com.google.common.base.Function<V,java.lang.Double> vertex_priors, double alpha) Creates an instance with the specified graph, vertex priors, and 'random jump' probability (alpha).

Method Summary

Modifier and Type	Method and Description
double	update(V v) Updates the value for this vertex.

Methods inherited from class edu.uci.ics.jung.algorithms.scoring.AbstractIterativeScorerWithPriors

getAlpha, getVertexPriors, initialize

Methods inherited from class edu.uci.ics.jung.algorithms.scoring.AbstractIterativeScorer

acceptDisconnectedGraph, done, evaluate, getEdgeWeights, getIterations, getMaxIterations, getTolerance, getVertexScore, isDisconnectedGraphOK, setEdgeWeights, setHyperedgesAreSelfLoops, setMaxIterations, setTolerance, step

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface edu.uci.ics.jung.algorithms.scoring.VertexScorer

getVertexScore

Constructor Detail

PageRankWithPriors

public PageRankWithPriors(Hypergraph<V,E> graph,
                          com.google.common.base.Function<E,? extends java.lang.Number> edge_weights,
                          com.google.common.base.Function<V,java.lang.Double> vertex_priors,
                          double alpha)

Creates an instance with the specified graph, edge weights, vertex priors, and 'random jump' probability (alpha).

Parameters:

graph - the input graph

edge_weights - the edge weights, denoting transition probabilities from source to destination

vertex_priors - the prior probabilities for each vertex

alpha - the probability of executing a 'random jump' at each step

PageRankWithPriors

public PageRankWithPriors(Hypergraph<V,E> graph,
                          com.google.common.base.Function<V,java.lang.Double> vertex_priors,
                          double alpha)

Creates an instance with the specified graph, vertex priors, and 'random jump' probability (alpha). The outgoing edge weights for each vertex will be equal and sum to 1.

Parameters:

graph - the input graph

vertex_priors - the prior probabilities for each vertex

alpha - the probability of executing a 'random jump' at each step

Method Detail

update

public double update(V v)

Updates the value for this vertex. Called by step().

Parameters:

v - the vertex whose value is to be updated

Returns:

the updated value