association rules Flashcards
What is another common name for Association Rules?
A) Clustering Analysis
B) Market Basket Analysis
C) Regression Analysis
D) Decision Tree Analysis
B) Market Basket Analysis
Explanation: Association Rules are also called market basket analysis or affinity analysis, focusing on discovering relationships like “Customers who bought X also bought Y”
In an association rule, what is the antecedent?
A) The THEN part of the rule
B) The IF part of the rule
C) The combination of both IF and THEN
D) The item with the highest frequency
B) The IF part of the rule
Explanation: The antecedent refers to the “IF” portion of the rule, while the consequent is the “THEN” part
What does support measure in association rules?
A) How strong the relationship is between items
B) The proportion of transactions containing both antecedent and consequent
C) The percentage of transactions that include a specific itemset
D) The probability of the consequent given the antecedent
Answer: C) The percentage of transactions that include a specific itemset
Explanation: Support refers to how frequently an itemset appears in the dataset
Which metric indicates how much better a rule performs compared to random selection?
A) Confidence
B) Support
C) Lift
D) Coverage
C) Lift
Explanation: Lift > 1 suggests the rule is useful and better than random chance at predicting the consequent
Why does the Apriori algorithm focus only on frequent itemsets?
A) To improve accuracy
B) To reduce computational complexity
C) To increase confidence values
D) To ensure higher lift ratios
Answer: B) To reduce computational complexity
Explanation: Considering only itemsets that meet a minimum support threshold limits the number of combinations, making computation feasible
In the rule selection process, after identifying frequent itemsets, what is the next step?
A) Calculate lift values
B) Generate all possible item combinations
C) Generate rules that meet minimum confidence
D) Eliminate redundant transactions
Answer: C) Generate rules that meet minimum confidence
Explanation: After finding frequent itemsets, rules are created from them and filtered based on confidence thresholds
What is a potential risk when generating too many association rules?
A) Missing important patterns
B) Increasing the chance of random, meaningless rules
C) Reducing lift values
D) Decreasing support automatically
Answer: B) Increasing the chance of random, meaningless rules
Explanation: Generating too many rules can lead to patterns that occur by chance, especially in smaller datasets
What is the main purpose of collaborative filtering in recommender systems?
A) To filter out infrequent items
B) To predict future sales trends
C) To recommend items based on similar users’ preferences
D) To group items into clusters
Answer: C) To recommend items based on similar users’ preferences
Explanation: Collaborative filtering suggests items by analyzing preferences of users with similar behavior
