Class ProbabilisticRankingReasoner

All Implemented Interfaces:
ModelProvider<Argument,DungTheory,NumericalPartialOrder<Argument,DungTheory>>, PostulateEvaluatable<Argument>

public class ProbabilisticRankingReasoner extends AbstractRankingReasoner<NumericalPartialOrder<Argument,DungTheory>>
Implements a graded semantics reasoner based on the ideas from [Thimm, Cerutti, Rienstra. Probabilistic Graded Semantics. COMMA 2018].
Author:
Matthias Thimm
  • Field Details

    • NUMBER_OF_TRIALS

      public static int NUMBER_OF_TRIALS
      Number of trials for the used monte carlo search (this is a factor multiplied with the number of arguments of the actual framework)
  • Constructor Details

    • ProbabilisticRankingReasoner

      public ProbabilisticRankingReasoner(Semantics sem, Probability p, boolean exactInference)
      Creates a new ProbabilisticRankingReasoner.
      Parameters:
      sem - The classical semantics used for evaluating subgraphs
      p - The probability used for all arguments to instantiate a probabilistic argumentation framework
      exactInference - Whether to use exact inference.
  • Method Details