Using J-Pruning to Reduce Overfitting of Classification Rules in Noisy Domains

Authors:
Max Bramer
Affiliations:
-
Venue:
DEXA '02 Proceedings of the 13th International Conference on Database and Expert Systems Applications
Year:
2002

Citing 4
Cited 3

C4.5: programs for machine learning

C4.5: programs for machine learning
An Empirical Comparison of Pruning Methods for Decision Tree Induction

Machine Learning
An Information-Theoretic Approach to the Pre-pruning of Classification Rules

Proceedings of the IFIP 17th World Computer Congress - TC12 Stream on Intelligent Information Processing
Using J-Pruning to Reduce Overfitting of Classification Rules in Noisy Domains

DEXA '02 Proceedings of the 13th International Conference on Database and Expert Systems Applications

Using J-Pruning to Reduce Overfitting of Classification Rules in Noisy Domains

DEXA '02 Proceedings of the 13th International Conference on Database and Expert Systems Applications
An evaluation of heuristics for rule ranking

Artificial Intelligence in Medicine
Computationally efficient induction of classification rules with the PMCRI and J-PMCRI frameworks

Knowledge-Based Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

The automatic induction of classification rules from examples is an important technique used in data mining. One of the problems encountered is the overfitting of rules to training data. This paper describes a means of reducing overfitting known as J-pruning, based on the J-measure, an information theoretic means of quantifying the information content of a rule, and examines its effectiveness in the presence of noisy data for two rule induction algorithms: one where the rules are generated via the intermediate representation of a decision tree and one where rules are generated directly from examples.