An Architecture of Neural Network Classifier with Fuzzy Teaching

Autor: Wade Wang, 王文堂
Rok vydání: 1993
Druh dokumentu: 學位論文 ; thesis
Popis: 81
A neural network for classification problems with fuzzy inputs is proposed. A fuzzy input is represented as a LR-type fuzzy set. A generalized pocket algorithm, called fuzzy pocket algorithm, that utilizes LR-type fuzzy sets operations and defuzzification method is proposed to train a linear threshold unit(LTU). This LTU node will classify as many fuzzy input instances as possible. Afterwards, FV nodes that represent fuzzy interval vectors will then be generated and expanded, by proposed FVGE learning algorithm, to classify those fuzzy input instances that cannot be classified by the LTU node. The similarity degree between FV nodes and fuzzy inputs is measured by the fuzzy subsethood degree. The FVGE learning algorithm can be applied to hyperbox-based classifier, e.g., Fuzzy ART series, Fuzzy Min-Max Classifier. The Network structure is automatically generated. Besides, on-line learning is supplied and learning speed is fast. Two sample problems, called heart disease and knowledge-based evaluator, are considerd to illustrate the workings of the proposed model. The experimental results are very encouraging.
Databáze: Networked Digital Library of Theses & Dissertations