In association rules mining, the aprori help to find all frequent itemsets in a dataset and generate strong association rules.
Algorithm: Apriori
Input: Dataset (D), Minimum Support (min_support), Minimum Confidence (min_confidence)
Output: Frequent Itemsets, Strong Rules
1. Initialize: L1 = {frequent 1-itemsets}
2. For (k = 2; Lk-1 is not empty; k++):
a. Generate Ck, candidate k-itemsets, from Lk-1
b. For each transaction t in D:
i. Increment count of all candidates in Ck that are contained in t
c. Lk = {c in Ck | support(c) >= min_support}
3. Frequent Itemsets = Union of all Lk
4. For each frequent itemset l in Frequent Itemsets:
a. Generate all non-empty subsets of l
b. For every non-empty subset s of l:
i. Rule = s -> (l - s)
ii. If Calculate_Confidence(D, Rule) >= min_confidence:
a. Add Rule to Strong Rules
5. Return Frequent Itemsets, Strong Rules