作者: Arthur Ramer , Maria do Carmo Nicoletti , Flávia O. Santos de Sá Lisboa
关键词: Hyperrectangle 、 Mathematics 、 Fuzzy logic 、 Set (abstract data type) 、 Euclidean distance 、 Artificial intelligence 、 Similarity (network science) 、 Pattern recognition 、 k-nearest neighbors algorithm 、 Euclidean space 、 Training set
摘要: The Nested Generalized Exemplar (NGE) model is an incremental form of inductive learning that generalizes a given training set into hypotheses represented as hyperrectangles in n-dimensional Euclidean space. NGE algorithm can be considered descendent either Nearest Neighbor (NN) or K-Nearest (KNN) algorithms. based systems classify new instances by calculating their similarity to the nearest generalized exemplar (i.e. hyperrectangle). Similarity implemented distance metric namely distance. This paper describes version suitable for fuzzy domains called Fuzzy (F-NGE). F-NGE learns rules classifying crisp classes. An implementation has been tested several different knowledge which results are presented and discussed. Results versions NN KNN using same also presented, comparison.