Incremental Learning in a Fuzzy Intelligent System
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.
Cite
Text
Murphey and Chen. "Incremental Learning in a Fuzzy Intelligent System." International Joint Conference on Artificial Intelligence, 1999.Markdown
[Murphey and Chen. "Incremental Learning in a Fuzzy Intelligent System." International Joint Conference on Artificial Intelligence, 1999.](https://mlanthology.org/ijcai/1999/murphey1999ijcai-incremental/)BibTeX
@inproceedings{murphey1999ijcai-incremental,
title = {{Incremental Learning in a Fuzzy Intelligent System}},
author = {Murphey, Yi Lu and Chen, Tie Qi},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {1999},
pages = {1376-1384},
url = {https://mlanthology.org/ijcai/1999/murphey1999ijcai-incremental/}
}