Class AbstractIterativeScorerWithPriors<V,​E,​S>

  • Type Parameters:
    V - the vertex type
    E - the edge type
    S - the score type
    All Implemented Interfaces:
    VertexScorer<V,​S>, IterativeContext
    Direct Known Subclasses:
    HITSWithPriors, PageRankWithPriors

    public abstract class AbstractIterativeScorerWithPriors<V,​E,​S>
    extends AbstractIterativeScorer<V,​E,​S>
    implements VertexScorer<V,​S>
    An abstract class for iterative random-walk-based vertex scoring algorithms that have a fixed probability, for each vertex, of 'jumping' to that vertex at each step in the algorithm (rather than following a link out of that vertex).
    • Field Detail

      • vertex_priors

        protected com.google.common.base.Function<? super V,​? extends S> vertex_priors
        The prior probability of each vertex being visited on a given 'jump' (non-link-following) step.
      • alpha

        protected double alpha
        The probability of making a 'jump' at each step.
    • Constructor Detail

      • AbstractIterativeScorerWithPriors

        public AbstractIterativeScorerWithPriors​(Hypergraph<V,​E> g,
                                                 com.google.common.base.Function<? super E,​? extends java.lang.Number> edge_weights,
                                                 com.google.common.base.Function<? super V,​? extends S> vertex_priors,
                                                 double alpha)
        Creates an instance for the specified graph, edge weights, vertex priors, and jump probability.
        Parameters:
        g - the graph whose vertices are to be assigned scores
        edge_weights - the edge weights to use in the score assignment
        vertex_priors - the prior probabilities of each vertex being 'jumped' to
        alpha - the probability of making a 'jump' at each step
      • AbstractIterativeScorerWithPriors

        public AbstractIterativeScorerWithPriors​(Hypergraph<V,​E> g,
                                                 com.google.common.base.Function<V,​? extends S> vertex_priors,
                                                 double alpha)
        Creates an instance for the specified graph, vertex priors, and jump probability, with edge weights specified by the subclass.
        Parameters:
        g - the graph whose vertices are to be assigned scores
        vertex_priors - the prior probabilities of each vertex being 'jumped' to
        alpha - the probability of making a 'jump' at each step
    • Method Detail

      • getVertexPrior

        protected S getVertexPrior​(V v)
        Returns the prior probability for v.
        Parameters:
        v - the vertex whose prior probability is being queried
        Returns:
        the prior probability for v
      • getVertexPriors

        public com.google.common.base.Function<? super V,​? extends S> getVertexPriors()
        Returns a Function which maps each vertex to its prior probability.
        Returns:
        a Function which maps each vertex to its prior probability
      • getAlpha

        public double getAlpha()
        Returns the probability of making a 'jump' (non-link-following step).
        Returns:
        the probability of making a 'jump' (non-link-following step)