,
, CS 7641 Machine Learning Practice Midterm
Exam | Georgia Tech | Practice Questions,
Answers & Concept Review | 2026/2027
update.
SECTION 1: DECISION TREES (Questions 1–12)
1. What is entropy in the context of decision trees?
A) A measure of impurity in a dataset
B) The total number of examples
C) The depth of the tree
D) The accuracy of predictions
Correct Answer: A
Rationale: Entropy measures the impurity or uncertainty in a dataset. In
decision trees, entropy is used to quantify how mixed the classes are within
a node. A pure node (all examples same class) has entropy of 0, while a
maximally mixed node has entropy of 1 (for binary classification) .
Source: CS 7641 SL Unit Quiz Practice Questions
2. How is information gain in decision trees measured?
A) Entropy reduction: the greatest decrease in probability of seeing multiple
different values
B) Increase in the number of branches
C) Total number of leaves in the tree
D) Depth of the tree
Correct Answer: A
Rationale: Information gain is measured as entropy reduction — the
decrease in entropy achieved by splitting on a particular attribute. The
, CS 7641 Machine Learning Practice Midterm
Exam | Georgia Tech | Practice Questions,
Answers & Concept Review | 2026/2027
update.
SECTION 1: DECISION TREES (Questions 1–12)
1. What is entropy in the context of decision trees?
A) A measure of impurity in a dataset
B) The total number of examples
C) The depth of the tree
D) The accuracy of predictions
Correct Answer: A
Rationale: Entropy measures the impurity or uncertainty in a dataset. In
decision trees, entropy is used to quantify how mixed the classes are within
a node. A pure node (all examples same class) has entropy of 0, while a
maximally mixed node has entropy of 1 (for binary classification) .
Source: CS 7641 SL Unit Quiz Practice Questions
2. How is information gain in decision trees measured?
A) Entropy reduction: the greatest decrease in probability of seeing multiple
different values
B) Increase in the number of branches
C) Total number of leaves in the tree
D) Depth of the tree
Correct Answer: A
Rationale: Information gain is measured as entropy reduction — the
decrease in entropy achieved by splitting on a particular attribute. The
