A Decision Tree Using Correlations to Evaluate Condition Attribut es
Han Jia · Microcomputer Development · 2003
The decision tree is an important classifier in data mi ning.In the pape r a decision tree classifier based on correlation is proposed after dicussing tr aditional algorithms.The main idea is to evaluate condition attributes with corr elations,and made pruning and optimization process simplify in order to get high accuracy and fast classifying speed,which avoid incorrect classification by inf ormation entropy.The accuracy of the algorithm was proven and the complexity of time was analyzed.