Noise-Tolerant Instance-Based Learning Algorithms
Proceedings of the Eleventh International Joint Conference on Artificial Intelligence. Machine Learning. Pages 794-798.
Abstract
Several published reports show that instancebased learning algorithms yield high classification accuracies and have low storage requirements during supervised learning applications. However, these learning algorithms are highly sensitive to noisy training instances. This paper describes a simple extension of instancebased learning algorithms for detecting and removing noisy instances from concept descriptions. This extension requires evidence that saved instances be significantly good classifiers before it allows them to be used for subsequent classification tasks. We show that this extension's performance degrades more slowly in the presence of noise , improves classification accuracies, and further reduces storage requirements in several artificial and real-world database applications.
@inproceedings{ijcai1989p127,
title = {Noise-Tolerant Instance-Based Learning Algorithms},
author = {Aha, David W. and Kibler, Dennis F.},
booktitle = {Proceedings of the Eleventh International Joint Conference on
Artificial Intelligence, {IJCAI-89}},
publisher = {International Joint Conferences on Artificial Intelligence},
editor = {N. S. Sridharan},
pages = {794--798},
year = {1989},
month = {8},
note = {Machine Learning},
url = {https://ijcai.org/proceedings/1989/127},
}
TY - CPAPER TI - Noise-Tolerant Instance-Based Learning Algorithms AU - Aha, David W. AU - Kibler, Dennis F. T2 - Proceedings of the Eleventh International Joint Conference on Artificial Intelligence, IJCAI-89 PB - International Joint Conferences on Artificial Intelligence PY - 1989 SP - 794 EP - 798 UR - https://ijcai.org/proceedings/1989/127 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
