Quiz 2 Flashcards
What is association rule mining?
Looking at how the data collected can be associated to each other. The frequency of purchasing an item and purchasing another item.
For Example: Diapers and Beer
- Give two examples of how association rule mining can be used in other than the market-basket use case. [Hint: Think through the various examples we discussed in class Tuesday (Jan 29 2024)]
Identifying common topics where students struggle
Health care and medical diagnosis
- Why is the confidence of association rules always less than 1?
Partial Coverage, Diverse Outcomes, Incomplete data/Noise, Conditional Nature of Confidence.
Reaching complete confidence is uncommon/unrealistic.
- Consider a transactional database where a pen is bought 15 times and a bag is bought 25 times. What is the maximum number of times that a pen and a bag are bought together?
A. 10
B. 15
C. 20
D. 25
B. 15
- What is the main goal of the market-basket model for frequent itemset mining in supermarket shelf management?
A. Maximize the number of items on the shelf
B. Identify items that are bought together by sufficiently many customers
C. Minimize the sales data collected with barcode scanners
D. Reduce variety of products in the supermarket
B. Identify items that are bought together by sufficiently many customers
- In association rules, what does the “interest” of a rule measure?
A. The likelihood of a rule being significant
B. The difference between confidence and support
C. The probability of the items being picked up
D. The overall popularity of the items in the rule
A. The likelihood of a rule being significant
- What information is typically stored in a supermarket’s market-basket data?
A. Individual item prices
B. Customer demographics
C. Lists of items purchased by each customer on a specific day
D. Sales revenue generated by each item
C. Lists of items purchased by each customer on a specific day
- True | False In the market basket analysis, association rules with high confidence are always considered interesting, regardless of the support of the itemsets involved.
False
- True | False The support threshold in market basket analysis determines the minimum number of baskets in which an itemset must appear to be considered frequent.
True
- True | False The Naïve algorithm to find frequent pairs takes n2 bytes of memory, where n is the number of universal items available.
True
- True | False The triangular matrix approach to count the number of pairs is considered to be more efficient than the triples approach in all cases.
False
- True| False All rules with high confidence are considered as interesting rules.
False
- True | False An item can be part of a frequent itemset even if it is individually not a frequent item.
False
- True | False If the confidence of association rule A, B, C -> D is low, then the confidence of
A, B -> C, D is also low.
False
