Question

Select all correct answers about decision trees. Decision trees are unsupervised learning algorithms The greedy learning of decision trees we have seen in class (ID3) are totally random, so each time it is run, it will be producing a different tree. Decision trees can only be used for classification. The entropy of a random variable ranges between 0 and 1. Explainable decisions and fast training are strong positive aspects of decision trees. The entropy of a random variable is largest when all the possible values it can take are equally likely. The greedy learning of decision trees we have seen in class (ID3) produces the tree that finds the global maximum. A decision tree with an arbitrary number of nodes can fit a noise-free (no intrinsic noise) training set perfectly. Decision trees can be used for both regression and classification.

          Select all correct answers about decision trees.
Decision trees are unsupervised learning algorithms
The greedy learning of decision trees we have seen in class (ID3) are totally random, so each time it is run, it will be producing a different tree.
Decision trees can only be used for classification.
The entropy of a random variable ranges between 0 and 1.
Explainable decisions and fast training are strong positive aspects of decision trees.
The entropy of a random variable is largest when all the possible values it can take are equally likely.
The greedy learning of decision trees we have seen in class (ID3) produces the tree that finds the global maximum.
A decision tree with an arbitrary number of nodes can fit a noise-free (no intrinsic noise) training set perfectly.
Decision trees can be used for both regression and classification.

Select all correct answers about decision trees.
Decision trees are unsupervised learning algorithms
The greedy learning of decision trees we have seen in class (ID3) are totally random, so each time it is run, it will be producing a different tree.
Decision trees can only be used for classification.
The entropy of a random variable ranges between 0 and 1.
Explainable decisions and fast training are strong positive aspects of decision trees.
The entropy of a random variable is largest when all the possible values it can take are equally likely.
The greedy learning of decision trees we have seen in class (ID3) produces the tree that finds the global maximum.
A decision tree with an arbitrary number of nodes can fit a noise-free (no intrinsic noise) training set perfectly.
Decision trees can be used for both regression and classification.

Added by Yolanda R.

Question

Please give Ace some feedback