Measures of interestingness (e.g., support, confidence, lift) MCQs

1. What does the support metric represent in association rule mining?

A) The proportion of transactions containing only the consequent
B) The proportion of transactions containing the antecedent and consequent
C) The strength of an association rule compared to random chance
D) The likelihood of the antecedent leading to the consequent

Answer: B) The proportion of transactions containing the antecedent and consequent
Explanation: Support measures how often the items in a rule appear together in the dataset.

2. Which of the following is the formula for support?

A) $Transactions\text{Support}(A \rightarrow B) = \frac{\text{Count of (A ∪ B)}}{\text{Total Transactions}}$
B) $B\text{Support}(A \rightarrow B) = \frac{\text{Count of A}}{\text{Count of B}}$
C) $A\text{Support}(A \rightarrow B) = \frac{\text{Count of (A ∩ B)}}{\text{Count of A}}$
D) $A\text{Support}(A \rightarrow B) = \frac{\text{Total Transactions}}{\text{Count of A}}$

Answer: A) $Transactions\text{Support}(A \rightarrow B) = \frac{\text{Count of (A ∪ B)}}{\text{Total Transactions}}$
Explanation: Support measures the proportion of transactions in which both $A$ and $B$ appear.

3. What does the confidence metric measure?

A) The proportion of transactions that contain the antecedent
B) The proportion of transactions containing the antecedent in which the consequent also appears
C) The proportion of transactions containing the consequent
D) The deviation of the rule from independence

Answer: B) The proportion of transactions containing the antecedent in which the consequent also appears
Explanation: Confidence indicates the likelihood of $B$ occurring given $A$ has occurred.

4. Which of the following is the formula for confidence?

A) $Confidence(A→B)=Support(A∪B)Support(A)\text{Confidence}(A \rightarrow B) = \frac{\text{Support}(A \cup B)}{\text{Support}(A)}$
B) $Confidence(A→B)=Support(A)Support(B)\text{Confidence}(A \rightarrow B) = \frac{\text{Support}(A)}{\text{Support}(B)}$
C) $Confidence(A→B)=Support(A∩B)Support(A)\text{Confidence}(A \rightarrow B) = \frac{\text{Support}(A \cap B)}{\text{Support}(A)}$
D) $Confidence(A→B)=Support(A∩B)Support(B)\text{Confidence}(A \rightarrow B) = \frac{\text{Support}(A \cap B)}{\text{Support}(B)}$

Answer: A) $Confidence(A→B)=Support(A∪B)Support(A)\text{Confidence}(A \rightarrow B) = \frac{\text{Support}(A \cup B)}{\text{Support}(A)}$
Explanation: Confidence measures the conditional probability of $B$ given $A$ .

5. What does a confidence value of 1 indicate?

A) $A$ and $B$ are completely independent.
B) $B$ always occurs when $A$ occurs.
C) $A$ never occurs without $B$ .
D) $A$ and $B$ are negatively associated.

Answer: B) $B$ always occurs when $A$ occurs.
Explanation: A confidence value of 1 means the rule $\rightarrow B$ is perfectly reliable.

6. What is the lift metric used for?

A) To measure the frequency of an itemset in the dataset
B) To measure how much more likely $A$ and $B$ occur together than by chance
C) To measure the reliability of an association rule
D) To filter out infrequent itemsets

Answer: B) To measure how much more likely $A$ and $B$ occur together than by chance
Explanation: Lift compares the observed support of the rule to the support expected if $A$ and $B$ were independent.

7. Which of the following is the formula for lift?

A) $Lift(A→B)=Confidence(A→B)Support(B)\text{Lift}(A \rightarrow B) = \frac{\text{Confidence}(A \rightarrow B)}{\text{Support}(B)}$
B) $Lift(A→B)=Support(A∩B)Support(A)⋅Support(B)\text{Lift}(A \rightarrow B) = \frac{\text{Support}(A \cap B)}{\text{Support}(A) \cdot \text{Support}(B)}$
C) $Lift(A→B)=Support(A)Support(A∩B)\text{Lift}(A \rightarrow B) = \frac{\text{Support}(A)}{\text{Support}(A \cap B)}$
D) $Lift(A→B)=Support(A)⋅Support(B)Support(A∩B)\text{Lift}(A \rightarrow B) = \frac{\text{Support}(A) \cdot \text{Support}(B)}{\text{Support}(A \cap B)}$

Answer: B) $Lift(A→B)=Support(A∩B)Support(A)⋅Support(B)\text{Lift}(A \rightarrow B) = \frac{\text{Support}(A \cap B)}{\text{Support}(A) \cdot \text{Support}(B)}$
Explanation: Lift measures the ratio of the observed co-occurrence of $A$ and $B$ to their expected co-occurrence if they were independent.

8. What does a lift value of 1 indicate?

A) $A$ and $B$ are completely independent.
B) $A$ always leads to $B$ .
C) $A$ and $B$ are positively correlated.
D) $A$ and $B$ are negatively correlated.

Answer: A) $A$ and $B$ are completely independent.
Explanation: A lift value of 1 means there is no association between $A$ and $B$ .

9. What does a lift value greater than 1 imply?

A) $A$ and $B$ are negatively associated.
B) $A$ and $B$ are more likely to occur together than expected by chance.
C) $A$ and $B$ are independent.
D) $A$ and $B$ are rare in the dataset.

Answer: B) $A$ and $B$ are more likely to occur together than expected by chance.
Explanation: A lift value greater than 1 indicates a positive association between $A$ and $B$ .

10. Which of the following measures is primarily used to filter out infrequent itemsets?

A) Support
B) Confidence
C) Lift
D) Conviction

Answer: A) Support
Explanation: Support is used to determine whether an itemset occurs frequently enough to be considered for rule generation.