Prof. LEUNG Kwong SakKing IrwinTse Ming-Fun2023-03-292023-03-291999Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, 1999, vol. 5, pp. V-178 - V-18408843627http://hdl.handle.net/20.500.11861/7667This paper describes a novel learning system, named FF99 that learns fuzzy first-order logic concepts from various kinds of data. FF99 builds on the ideas from both fuzzy set theory and first-order logic. Object relationships are described using fuzzy relations based on which FF99 generates classification rules expressed in a restricted from fuzzy first-order logic. This new system has been applied successfully to several tasks taken from the machine learning literature. We demonstrate its usefulness in the applications of data mining through several experiments.enLearning SystemFirst-Order LogicData MiningFuzzy LogicFuzzy SetFuzzy Set TheoryFuzzy ConceptData Mining ApplicationsFuzzy RelationLearning AlgorithmsUniverseDistribution AnalysisRelated InformationReal ApplicationsInformation TheoryError DistributionError FunctionMembership FunctionHeuristic SearchRelational DatabaseKnowledge RepresentationTarget ConceptDifferential EntropyKernel EstimationLinearly SeparablePruning TechniquesWindow WidthReal-life ApplicationsNegative ErrorDistribution Of NodesConference Paper10.1109/ICSMC.1999.815544