Association Analysis

Question 1

Association Rule Mining

Answer 1

Given a set of transactions, find rules that will predict the occurrence of an item based on the occurrences of other items in the transaction.

Question 2

2 Rule Evaluation Metrics

Answer 2

Support - A measure of absolute frequency (Fraction that contain an itemset)
Confidence (c): A measure of correlative frequency.

Question 3

3 Step of Brute-force approach

Answer 3

List all possible association rules
Compute the support and confidence for each rule
Prune rules that fail the minsup and minconf

Question 4

Formula of

Total number of itemsets
Total number of possible association rules
Combination Formula

Answer 4

Total number of itemsets = 2k-1
Total number of possible association rules: 3k - 2k+1 + 1
Combination Formula: C(n,r) = nCr = n! / ((n - r )! * r!)

Question 5

2 Method of Candidate Generation

Answer 5

Fk-1 * F1 Method
Fk-1 * Fk-1 Method

Question 6

Maximal frequent itemset

Answer 6

An itemset is maximal frequent if it is frequent and none of its immediate supersets is frequent.

Question 7

Closed Itemset

Answer 7

An itemset X is closed if none of its immediate supersets has the same support as the itemset X.