TODO: Implement gini index using the method of intropy for the better results and accuracy
Implementation of decision tree in python dataset used : bank_note dataset with five attributes.
link: https://www.openml.org/d/1462
links: https://machinelearningmastery.com/implement-decision-tree-algorithm-scratch-python/
Linkes for confusion matrix: http://chem-eng.utoronto.ca/~datamining/dmc/model_evaluation_c.htm
creating the binary decision tree is actually dividing the input space
A greedy approach is used to divide the space called recursivebinary spliting.
split with best cost is selected.
Regression: The cost function that is minimized to choose split point is the sum squared error across all training samples that fall within the rectangle.
Dataset used : BankNote dataset.
Zero rule classification: This predicts the majority class irrespective of the classifier.
Gini index: is the name of cost function to split the dataset.