Decision Tree is one of the successful data mining
techniques used in the diagnosis of heart disease. Yet its
accuracy is not perfect. Most research applies the J4.8
Decision Tree that is based on Gain Ratio and binary
discretization. This research systematically tested
combinations of discretization, decision tree type and
voting to identify a more robust, more accurate method.
The supervised discretization methods do not show any
enhancement in the Decision Tree accuracy either with or
without voting. Applying voting shows increase in the
accuracy of different types of Decision Tree. Systematic
testing against a widely-used benchmark data set shows
that nine voting with equal frequency discretization and
Gain Ratio Decision Tree can enhance the accuracy of the
diagnosis of heart disease.