Continuous Attributes Partition Approach Based on Improved Statistical Independence
MAO Mingyang
Abstract：A new improved continuous attributes partition approach based on the statistical independence called APA-SI is proposed and described in this paper.APA-SI considers the effect of variance in the two merged intervals.It not only considers the effect of variance on degrees of freedom,but also takes the effect of variance on data distribution into consideration.A series of ex-periments to evaluate the utility of APA-SI method on the credit risk dataset are done.The data mining techniques,such as C4.5 de-cision tree,Naive Bayes and SVM classifiers,to classify and predict the quantified data are applied.The simulation results show that the approach significantly improves the mean accuracy of classification than continuous approval data and other known quantiza-tion methods such as EFD,MDLP,Extended Chi2.
Keywords：continuous attributespartition approachclassification methodsdata miningmachine learning
Publication Date：2025-04-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 926-929,947 )