A Distribution-Index-Based Discretizer for Decision-Making with Symbolic AI Approaches

Authors:
QingXiang Wu;David A. Bell;Girijesh Prasad;Thomas Martin McGinnity
Affiliations:
IEEE;IEEE;IEEE;IEEE
Venue:
IEEE Transactions on Knowledge and Data Engineering
Year:
2007

Citing 8
Cited 0

FUSINTER: a method for discretization of continuous attributes

International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems
Machine Learning

Machine Learning
A Modified Chi2 Algorithm for Discretization

IEEE Transactions on Knowledge and Data Engineering
Induction of Decision Trees

Machine Learning
CAIM Discretization Algorithm

IEEE Transactions on Knowledge and Data Engineering
Knowledge Discovery in Multiple Databases

Knowledge Discovery in Multiple Databases
Multiknowledge for decision making

Knowledge and Information Systems
Multi-knowledge extraction and application

RSFDGrC'03 Proceedings of the 9th international conference on Rough sets, fuzzy sets, data mining, and granular computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

When symbolic AI approaches are applied to handle continuous valued attributes, there is a requirement to transform the continuous attribute values to symbolic data. In this paper, a novel distribution-index-based discretizer is proposed for such a transformation. Based on definitions of dichotomic entropy and a compound distributional index, a simple criterion is applied to discretize continuous attributes adaptively. The dichotomic entropy indicates the homogeneity degree of the decision value distribution, and is applied to determine the best splitting point. The compound distributional index combines both the homogeneity degrees of attribute value distributions and the decision value distribution, and is applied to determine which interval should be split further; thus, a potentially improved solution of the discretization problem can be found efficiently. Based on multiple reducts in rough set theory, a multiknowledge approach can attain high decision accuracy for information systems with a large number of attributes and missing values. In this paper, our discretizer is combined with the multiknowledge approach to further improve decision accuracy for information systems with continuous attributes. Experimental results on benchmark data sets show that the new discretizer can improve not only the multiknowledge approach, but also the naïve Bayes classifier and the C5.0 tree.