Suppose you are given a dataset with 5 classes. You have trained a random forest using the training set examples and obtained four trees. Then, you have tested each tree on the test examples and obtained the following results on test samples:

It is known that the success of random forest algorithm depends on two factors: Diversity and accuracy of the individual trees.

a) Compute accuracy of each individual tree.

b) Compute the accuracy obtained if the predictions of the individual trees are combined with majority voting.

c) Propose a metric to compute diversity between two trees. Note that the proposed metric should take a value in the range of [0, 1].

d) Compute the diversity scores between each pair of trees using the proposed metric.

Class Predictions of Algorithms Tree 1 Tree 2 Tree 3 Tree 4 Sample id 1 2 3 4 5 6 7 8 9 10 2 3 2 5 4 2 4 2 3 4 3 2 4 3 1 3 3 1 2 4 5 3 2 1 5 2 3 Actual Class Label 2 4 1 5 4 3 1 2 3 5 1 4 1 5 2 1 3 2 4 5 3 2