1 /*
2 * Created on Jul 12, 2007
3 *
4 * Copyright (c) 2007, The JUNG Authors
5 *
6 * All rights reserved.
7 *
8 * This software is open-source under the BSD license; see either
9 * "license.txt" or
10 * https://github.com/jrtom/jung/blob/master/LICENSE for a description.
11 */
12 package edu.uci.ics.jung.algorithms.scoring;
13
14 import com.google.common.base.Function;
15
16 import edu.uci.ics.jung.algorithms.scoring.util.ScoringUtils;
17 import edu.uci.ics.jung.graph.Hypergraph;
18
19 /**
20 * Assigns scores to each vertex according to the PageRank algorithm.
21 *
22 * <p>PageRank is an eigenvector-based algorithm. The score for a given vertex may be thought of
23 * as the fraction of time spent 'visiting' that vertex (measured over all time)
24 * in a random walk over the vertices (following outgoing edges from each vertex).
25 * PageRank modifies this random walk by adding to the model a probability (specified as 'alpha'
26 * in the constructor) of jumping to any vertex. If alpha is 0, this is equivalent to the
27 * eigenvector centrality algorithm; if alpha is 1, all vertices will receive the same score
28 * (1/|V|). Thus, alpha acts as a sort of score smoothing parameter.
29 *
30 * <p>The original algorithm assumed that, for a given vertex, the probability of following any
31 * outgoing edge was the same; this is the default if edge weights are not specified.
32 * This implementation generalizes the original by permitting
33 * the user to specify edge weights; in order to maintain the original semantics, however,
34 * the weights on the outgoing edges for a given vertex must represent transition probabilities;
35 * that is, they must sum to 1.
36 *
37 * <p>If a vertex has no outgoing edges, then the probability of taking a random jump from that
38 * vertex is (by default) effectively 1. If the user wishes to instead throw an exception when this happens,
39 * call <code>acceptDisconnectedGraph(false)</code> on this instance.
40 *
41 * <p>Typical values for alpha (according to the original paper) are in the range [0.1, 0.2]
42 * but may be any value between 0 and 1 inclusive.
43 *
44 * @see "The Anatomy of a Large-Scale Hypertextual Web Search Engine by L. Page and S. Brin, 1999"
45 */
46 public class PageRank<V,E> extends PageRankWithPriors<V,E>
47 {
48
49 /**
50 * Creates an instance for the specified graph, edge weights, and random jump probability.
51 * @param graph the input graph
52 * @param edge_weight the edge weights (transition probabilities)
53 * @param alpha the probability of taking a random jump to an arbitrary vertex
54 */
55 public PageRank(Hypergraph<V,E> graph, Function<E, ? extends Number> edge_weight, double alpha)
56 {
57 super(graph, edge_weight, ScoringUtils.getUniformRootPrior(graph.getVertices()), alpha);
58 }
59
60 /**
61 * Creates an instance for the specified graph and random jump probability; the probability
62 * of following any outgoing edge from a given vertex is the same.
63 * @param graph the input graph
64 * @param alpha the probability of taking a random jump to an arbitrary vertex
65 */
66 public PageRank(Hypergraph<V,E> graph, double alpha)
67 {
68 super(graph, ScoringUtils.getUniformRootPrior(graph.getVertices()), alpha);
69 }
70 }