1. What is the main goal of association rule mining?
a) To find hidden patterns and relationships in a dataset
b) To cluster data into groups
c) To predict continuous values
d) To reduce the dimensionality of the data
Answer: a) To find hidden patterns and relationships in a dataset
2. In association rule mining, what does an itemset refer to?
a) A set of items that always appear together in the dataset
b) A single item in the dataset
c) A rule that describes the relationship between items
d) A subset of transactions containing at least one item
Answer: a) A set of items that always appear together in the dataset
3. Which of the following is a measure of how frequently an itemset appears in the dataset?
a) Confidence
b) Lift
c) Support
d) Frequency
Answer: c) Support
4. What is the confidence of an association rule A→BA \rightarrow B?
a) The proportion of transactions that contain both AA and BB
b) The proportion of transactions that contain AA
c) The probability of BB occurring given that AA has occurred
d) The ratio of the frequency of AA to the frequency of BB
Answer: c) The probability of BB occurring given that AA has occurred
5. Which of the following measures indicates how much more likely the rule is to occur compared to random chance?
a) Confidence
b) Support
c) Lift
d) Frequency
Answer: c) Lift
6. What does the Apriori algorithm use to reduce the search space in association rule mining?
a) It uses random sampling to reduce computation
b) It eliminates itemsets that do not meet the minimum support
c) It combines itemsets into new ones
d) It generates frequent itemsets using clustering
Answer: b) It eliminates itemsets that do not meet the minimum support
7. In association rule mining, what is the significance of support?
a) It indicates how strong the relationship is between items
b) It tells how frequently the rule is applicable to a transaction
c) It measures the correlation between items
d) It measures the number of unique items in the dataset
Answer: b) It tells how frequently the rule is applicable to a transaction
8. What is the typical use of association rule mining in market basket analysis?
a) To predict the future sales of items
b) To identify associations between products bought together
c) To segment customers into distinct groups
d) To predict the next item a customer will buy
Answer: b) To identify associations between products bought together
9. Which of the following is NOT a valid measure for evaluating the quality of an association rule?
a) Support
b) Confidence
c) Lift
d) Precision
Answer: d) Precision
10. What does Lift measure in association rule mining?
a) The likelihood of finding itemset BB when itemset AA occurs
b) The relative increase in probability of BB occurring given AA
c) The strength of the rule between two items
d) The increase in the overall frequency of items
Answer: b) The relative increase in probability of BB occurring given AA
11. Which of the following statements is true regarding association rules?
a) They are typically used for clustering similar data points together
b) The rule A→BA \rightarrow B implies that AA is bought after BB
c) Rules with high confidence always have high support
d) Association rules cannot be used for predicting numerical values
Answer: d) Association rules cannot be used for predicting numerical values
12. In the context of the Apriori algorithm, what does pruning refer to?
a) Removing irrelevant items from the dataset
b) Removing itemsets that are unlikely to be frequent
c) Reducing the size of the dataset
d) Eliminating redundant rules
Answer: b) Removing itemsets that are unlikely to be frequent
13. What is the minimum support threshold used for in association rule mining?
a) It filters out rules that do not meet the desired frequency level
b) It defines the maximum number of rules to be generated
c) It helps in identifying the most confident rules
d) It determines the minimum correlation between items
Answer: a) It filters out rules that do not meet the desired frequency level
14. Which of the following is a limitation of the Apriori algorithm?
a) It requires less memory for large datasets
b) It generates a large number of candidate itemsets
c) It is particularly efficient for categorical data
d) It uses a greedy approach to generate rules
Answer: b) It generates a large number of candidate itemsets
15. What type of data does association rule mining typically work with?
a) Structured data with continuous values
b) Time series data
c) Transactional data or data in the form of a market basket
d) Sequential data with ordered elements
Answer: c) Transactional data or data in the form of a market basket