Q.1 What is the main goal of association rule mining?
To classify data into groups
To find hidden patterns of co-occurrence
To reduce dimensionality
To cluster similar items
Explanation - Association rule mining discovers interesting relationships between items in large datasets, such as items often purchased together.
Correct answer is: To find hidden patterns of co-occurrence
Q.2 Which metric measures how often items appear together in a dataset?
Support
Confidence
Lift
Entropy
Explanation - Support measures the frequency of itemsets in the dataset.
Correct answer is: Support
Q.3 Confidence in association rule mining refers to:
The probability that an item appears in a dataset
The accuracy of classification
The probability of Y given X
The correlation of attributes
Explanation - Confidence is defined as P(Y|X), indicating how often Y appears in transactions that contain X.
Correct answer is: The probability of Y given X
Q.4 Lift is used to measure:
Frequency of occurrence
Independence of rules
Strength of an association rule compared to random chance
The size of the dataset
Explanation - Lift measures how much more often X and Y occur together than expected if they were independent.
Correct answer is: Strength of an association rule compared to random chance
Q.5 In association rule mining, which of the following is a common algorithm?
K-means
Decision Tree
Apriori
Naive Bayes
Explanation - Apriori is a classic algorithm for mining frequent itemsets and generating association rules.
Correct answer is: Apriori
Q.6 Apriori algorithm is based on which property?
Monotonicity property
Anti-monotonicity property
Transitivity property
Symmetry property
Explanation - Apriori uses the anti-monotonicity property which states that all subsets of a frequent itemset must also be frequent.
Correct answer is: Anti-monotonicity property
Q.7 Which of the following is NOT an application of association rule mining?
Market basket analysis
Fraud detection
DNA sequence analysis
Sorting algorithms
Explanation - Association rule mining is not used for sorting but for finding hidden relationships in data.
Correct answer is: Sorting algorithms
Q.8 What does FP in FP-Growth stand for?
Frequent Pattern
Fast Processing
Full Precision
Function Prediction
Explanation - FP-Growth stands for Frequent Pattern Growth, an algorithm for mining frequent itemsets without candidate generation.
Correct answer is: Frequent Pattern
Q.9 Compared to Apriori, FP-Growth is generally:
Slower
Faster
Less memory-efficient
Less accurate
Explanation - FP-Growth avoids generating candidate sets, making it faster than Apriori.
Correct answer is: Faster
Q.10 Which of the following best describes support?
Probability of Y given X
Fraction of transactions containing an itemset
Correlation coefficient
Error rate
Explanation - Support is the proportion of transactions that include a particular itemset.
Correct answer is: Fraction of transactions containing an itemset
Q.11 If an association rule has high lift, it means:
The rule is always true
The rule is stronger than random co-occurrence
The items are negatively correlated
The dataset is very large
Explanation - Lift > 1 indicates a positive correlation between items beyond random chance.
Correct answer is: The rule is stronger than random co-occurrence
Q.12 Which type of rules are generated in association rule mining?
IF-THEN rules
Boolean algebra rules
Arithmetic rules
Sorting rules
Explanation - Association rules are expressed as IF-THEN relationships between itemsets.
Correct answer is: IF-THEN rules
Q.13 In market basket analysis, association rules help to:
Find the cheapest items
Find items often bought together
Predict customer gender
Sort items by price
Explanation - Association rules in market basket analysis identify co-occurrence patterns of products.
Correct answer is: Find items often bought together
Q.14 Which of the following is true about Apriori?
It generates candidate sets explicitly
It avoids candidate generation
It uses a tree-based approach
It cannot be used for large datasets
Explanation - Apriori generates candidate itemsets and prunes them using support thresholds.
Correct answer is: It generates candidate sets explicitly
Q.15 What is the minimum threshold parameter for frequent itemsets called?
Minimum confidence
Minimum support
Minimum lift
Minimum entropy
Explanation - Minimum support is used to filter out infrequent itemsets.
Correct answer is: Minimum support
Q.16 Confidence can be calculated as:
Support(X ∪ Y) / Support(X)
Support(X) / Support(Y)
Support(X) * Support(Y)
1 - Support(Y)
Explanation - Confidence = Support(X ∪ Y) ÷ Support(X).
Correct answer is: Support(X ∪ Y) / Support(X)
Q.17 Lift can be calculated as:
Confidence / Support(Y)
Support(X) / Support(Y)
Support(X) * Support(Y)
Confidence - Support(X)
Explanation - Lift(X→Y) = Confidence(X→Y) ÷ Support(Y).
Correct answer is: Confidence / Support(Y)
Q.18 Which algorithm uses a tree structure to represent itemsets?
Apriori
FP-Growth
Naive Bayes
KNN
Explanation - FP-Growth uses an FP-Tree (Frequent Pattern Tree) for efficient mining.
Correct answer is: FP-Growth
Q.19 Which of these is NOT a metric in association rule mining?
Support
Confidence
Recall
Lift
Explanation - Recall is a metric in classification, not association rule mining.
Correct answer is: Recall
Q.20 What is the output of association rule mining?
Clusters
Decision trees
Rules showing relationships between items
Sorted lists
Explanation - The output is IF-THEN rules describing item co-occurrence.
Correct answer is: Rules showing relationships between items
Q.21 Which step comes first in Apriori?
Generate association rules
Generate frequent itemsets
Prune infrequent items
Calculate lift
Explanation - Frequent itemsets are generated first, and then rules are derived from them.
Correct answer is: Generate frequent itemsets
Q.22 Which data type is most suitable for association rule mining?
Transaction data
Continuous numeric data
Time series data
Streaming sensor data
Explanation - Association rule mining is commonly applied to transaction databases such as market baskets.
Correct answer is: Transaction data
Q.23 If support is too low, the result may include:
Too many irrelevant rules
Too few rules
High confidence rules
High lift rules
Explanation - Low support threshold may generate an overwhelming number of uninteresting rules.
Correct answer is: Too many irrelevant rules
Q.24 If confidence is too high, rules generated may:
Always be useful
Miss important relationships
Be redundant
Not exist
Explanation - High confidence may ignore interesting but less frequent rules.
Correct answer is: Miss important relationships
Q.25 Which of these combinations defines a strong association rule?
High support & low confidence
High support & high confidence
Low support & high lift
High entropy & low support
Explanation - Strong association rules typically have both high support and high confidence.
Correct answer is: High support & high confidence