Representation and quality of the instance data are the foremost factors that affects classification accuracy of the statistical - based method Decision tree algorithm which gives less accuracy for binary classification problems. Experiments shows that by using clustering and hyper-parameter tuning, the decision tree accuracy can be achieved above 95%, better than the 75% recognition using decision tree alone.

Keywords : Classification, Clustering, K-means, Decision Tree, Hyper-parameter Tuning, Grid Search, Customer Churn, Logistic Regression.