Using boolean differences for discovering ill-defined attributes in propositional machine learning

Authors:
Sylvain Hallé
Affiliations:
Université du Québec à Montréal, Montréal, Canada
Venue:
MICAI'05 Proceedings of the 4th Mexican international conference on Advances in Artificial Intelligence
Year:
2005

Citing 9
Cited 0

Learning in the presence of malicious errors

STOC '88 Proceedings of the twentieth annual ACM symposium on Theory of computing
An Experimental Comparison of Three Methods for Constructing Ensembles of Decision Trees: Bagging, Boosting, and Randomization

Machine Learning
Discovering the set of fundamental rule changes

Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining
An Empirical Comparison of Pruning Methods for Decision Tree Induction

Machine Learning
Induction of Decision Trees

Machine Learning
Reliable Classifications with Machine Learning

ECML '02 Proceedings of the 13th European Conference on Machine Learning
Characterizing Model Erros and Differences

ICML '00 Proceedings of the Seventeenth International Conference on Machine Learning
Theory Refinement with Noisy Data

Theory Refinement with Noisy Data
Computer Organization and Design

Computer Organization and Design

Quantified Score

Hi-index	0.00

Visualization

Abstract

The accuracy of the rules produced by a concept learning system can be hindered by the presence of errors in the data. Although these errors are most commonly attributed to random noise, there also exist “ill-defined” attributes that are too general or too specific that can produce systematic classification errors. We present a computer program called Newton which uses the fact that ill-defined attributes create an ordered error pattern among the instances to compute hypotheses explaining the classification errors of a concept in terms of too general or too specific attributes. Extensive empirical testing shows that Newton identifies such attributes with a prediction rate over 95%.