V
- the vertex typeE
- the edge typepublic class HITS<V,E> extends HITSWithPriors<V,E>
The classic HITS algorithm essentially proceeds as follows:
assign equal initial hub and authority values to each vertex repeat for each vertex w: w.hub = sum over successors x of x.authority w.authority = sum over predecessors v of v.hub normalize hub and authority scores so that the sum of the squares of each = 1 until scores convergeHITS is somewhat different from random walk/eigenvector-based algorithms such as PageRank in that:
UniformDegreeWeight
)
so that the weights of the relevant edges incident to a vertex sum to 1.
Modifier and Type | Class and Description |
---|---|
static class |
HITS.Scores
Maintains hub and authority score information for a vertex.
|
disappearing_potential
alpha, vertex_priors
edge_weights, graph, hyperedges_are_self_loops, max_delta, max_iterations, output_reversed, tolerance, total_iterations
Constructor and Description |
---|
HITS(Graph<V,E> g)
Creates an instance for the specified graph.
|
HITS(Graph<V,E> g,
double alpha)
Creates an instance for the specified graph and alpha (random jump probability)
parameter.
|
HITS(Graph<V,E> g,
com.google.common.base.Function<E,Double> edge_weights,
double alpha)
Creates an instance for the specified graph, edge weights, and alpha
(random jump probability) parameter.
|
afterStep, collectDisappearingPotential, normalizeScores, update
getAlpha, getVertexPrior, getVertexPriors, initialize
acceptDisconnectedGraph, done, evaluate, getAdjustedIncidentCount, getCurrentValue, getEdgeWeight, getEdgeWeights, getIterations, getMaxIterations, getOutputValue, getTolerance, getVertexScore, isDisconnectedGraphOK, setCurrentValue, setEdgeWeights, setHyperedgesAreSelfLoops, setMaxIterations, setOutputValue, setTolerance, step, swapOutputForCurrent, updateMaxDelta
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
getVertexScore
public HITS(Graph<V,E> g, com.google.common.base.Function<E,Double> edge_weights, double alpha)
g
- the input graphedge_weights
- the weights to use for each edgealpha
- the probability of a hub giving some authority to all vertices,
and of an authority increasing the score of all hubs (not just those connected
via links)public HITS(Graph<V,E> g, double alpha)
g
- the input graphalpha
- the probability of a hub giving some authority to all vertices,
and of an authority increasing the score of all hubs (not just those connected
via links)Copyright © 2015. All rights reserved.