Author :
Cheng, Paxton ; Forward, Kevin
Author_Institution :
Dept. of Electr. & Electron. Eng., Melbourne Univ., Vic., Australia
Abstract :
Proposes a new model of a fuzzy Petri net and an algorithm to generate such a network automatically. As an example of the application of the fuzzy Petri net, it is used to classify the Iris data set. Although there are extensive examples of neural network-based classifiers in the literature, they all share the undesirable characteristic of a long learning time. We attempt to remedy this problem by using a totally different architecture, and the resulting Petri net attains comparable performance with conventional systems with just a few seconds of training
Keywords :
Petri nets; fuzzy set theory; learning (artificial intelligence); mathematics computing; neural net architecture; pattern classification; Iris data set classification; fuzzy Petri net generation algorithm; learning time; neural net architecture; neural network-based classifier; performance; training; Data mining; Error analysis; Fuzzy logic; Fuzzy sets; Iris; Neural networks; Performance analysis; Petri nets; Solar power generation; Training data;
Conference_Titel :
Knowledge-Based Intelligent Electronic Systems, 1997. KES '97. Proceedings., 1997 First International Conference on
Conference_Location :
Adelaide, SA
Print_ISBN :
0-7803-3755-7
DOI :
10.1109/KES.1997.619416
