Incremental learning in a fuzzy intelligent system

Authors:
Yi Lu Murphey;Tie Qi Chen
Affiliations:
ECE Department, The University of Michigan-Dearborn, Dearborn, MI;ECE Department, The University of Michigan-Dearborn, Dearborn, MI
Venue:
IJCAI'99 Proceedings of the 16th international joint conference on Artificial intelligence - Volume 2
Year:
1999

Citing 5
Cited 0

Algorithmic learning

Algorithmic learning
Machine Learning

Machine Learning
Incremental Induction of Decision Trees

Machine Learning
Fast rule generation and membership function optimization for a fuzzy diagnosis system

IEA/AIE'1997 Proceedings of the 10th international conference on Industrial and Engineering Applications of Artificial Intelligence and Expert Systems
Incremental backpropagation learning networks

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents an incremental learning algorithm within the framework of a fuzzy intelligent system. The incremental learning algorithm is based on priority values attached to fuzzy rules. The priority value of a fuzzy rule is generated based on the fuzzy belief values of the fuzzy rule derived from the training data. The fuzzy incremental algorithm has three important properties. It can detect and recover from incorrect knowledge once new knowledge is available; it will not lose the useful knowledge generated from the old data while it attempts to learn from new data; and it provides a mechanism allowing to emphasize on knowledge learnt from the new data. The incremental fuzzy learning algorithm has been implemented in a fuzzy intelligent system for automotive engineering diagnosis. Its performance is presented in the paper.