학술논문
Comparing information-theoretic attribute selection measures: a statistical approach.
Document Type
Article
Source
Subject
*DECISION trees
*INFORMATION theory
*LEARNING
*
*
Language
ISSN
0921-7126
Abstract
Introduces an information-theoretic attribute selection method for decision tree induction. Measure of the distance-based attribute selection measure; Overview of the learning data sets; Discussion on the nonparametric statistical tests for analyzing paired data; Results obtained from the different attribute selection measures.