1.Decision trees can be obtained by

1.Decision trees can be obtained by data mining methods but may suffer from overfitting. Suggest how to overcome this problem. (3marks)We can deal with the overfitting problem by pruning the decision tree.Prepruning: Stop growth in advance before the decision tree is fully grown.(1) We can limit the height of the tree.(2) We can use the classification index to limit the growth of the decision tree. If the next split_point segmentation does not reduce the error, we will stop the split_point segmentation.(3) we can Limit the number of point in the tree. For example, if a point has less than 100 samples, we will stop cutting the point.Postpruning: we can prune the subtree on a given split_point , and replace with leaves, to pruning the branch of the split_point.