High Accurate and a Variant of k-fold Cross Validation Technique for Predicting the Decision Tree Classifier Accuracy

In machine learning data usage is the most important criterion than logic of program. With very big and moderate sized datasets it possible to obtain robust high classification accuracies but not with small datasets. particular only large training are potential for producing decision tree results. The results obtained by using one testing dataset pair reliable. Cross validation technique uses many random folds same validation. order reliable statistically correct there a need apply algorithm on different pairs To overcome problem single existing k-fold cross plan obtaining increased accuracy this paper new called prime fold proposed experimentally tested thoroughly then verified correctly bench mark UCI It observed that based after experimentation far better techniques finding accuracies.