Class Watset<V,​E>

  • Type Parameters:
    V - the type of nodes in the graph
    E - the type of edges in the graph
    All Implemented Interfaces:
    org.jgrapht.alg.interfaces.ClusteringAlgorithm<V>

    public class Watset<V,​E>
    extends Object
    implements org.jgrapht.alg.interfaces.ClusteringAlgorithm<V>
    Watset is a local-global meta-algorithm for fuzzy graph clustering.

    Watset builds an intermediate undirected graph by inducing different senses of each node in the input graph.

    This implementation of Watset is known as Simplified Watset. It does not need a context similarity measure.

    See Also:
    Ustalov et al. (COLI 45:3)
    • Field Detail

      • graph

        protected final org.jgrapht.Graph<V,​E> graph
        The graph.
      • inducer

        protected final SenseInduction<V,​E> inducer
        The node sense induction approach.
    • Constructor Detail

      • Watset

        public Watset​(org.jgrapht.Graph<V,​E> graph,
                      ClusteringAlgorithmBuilder<V,​E,​?> local,
                      ClusteringAlgorithmBuilder<Sense<V>,​org.jgrapht.graph.DefaultWeightedEdge,​?> global)
        Create an instance of the Watset clustering algorithm.
        Parameters:
        graph - the graph
        local - the local clustering algorithm supplier
        global - the global clustering algorithm supplier
    • Method Detail

      • builder

        public static <V,​E> Watset.Builder<V,​E> builder()
        Create a builder.
        Type Parameters:
        V - the type of nodes in the graph
        E - the type of edges in the graph
        Returns:
        a builder
      • getClustering

        public WatsetClustering<V> getClustering()
        Specified by:
        getClustering in interface org.jgrapht.alg.interfaces.ClusteringAlgorithm<V>