Suppose you are training a decision tree classifier. As the size of tree grows (the x-axis...

Transcribed Image Text:

Suppose you are training a decision tree classifier. As the size of tree grows (the x-axis in the plot shows the number of tree nodes), the classification accuracy curves of training and test data set are as plotted. classification accuracy 0.9 0.85 0.8 0.75 0.7 0.65 0.6 0.55 0.5 training data set test data set LO 20 30 40 50 60 70 80 90 100 (1) Judge the correctness of each statement below and give the reason. (a) Every point on the training accuracy curve provides an unbiased estimate of true error. (b) Every point on the test accuracy curve provides an unbiased estimate of true error. (c) Training accuracy minus test accuracy provides an unbiased estimate of the degree of overfitting. (2) Given the training and test accuracy in the plot, which tree size is optimal for the classifier? Explain your answer. Acti Suppose you are training a decision tree classifier. As the size of tree grows (the x-axis in the plot shows the number of tree nodes), the classification accuracy curves of training and test data set are as plotted. classification accuracy 0.9 0.85 0.8 0.75 0.7 0.65 0.6 0.55 0.5 training data set test data set LO 20 30 40 50 60 70 80 90 100 (1) Judge the correctness of each statement below and give the reason. (a) Every point on the training accuracy curve provides an unbiased estimate of true error. (b) Every point on the test accuracy curve provides an unbiased estimate of true error. (c) Training accuracy minus test accuracy provides an unbiased estimate of the degree of overfitting. (2) Given the training and test accuracy in the plot, which tree size is optimal for the classifier? Explain your answer. Acti