Apriori (Levelwise)
Scan database multiple times
For the ith scan, find all large itemsets of size i with min support
Use the large itemsets from scan i as input to scan i+1
Generate candidates by creating subsets of size i+1 which contain only large itemsets as subsets
Notation: Large k-itemset,
Set of candidate large itemsets of size k,
Note: If {A,B} is not a large itemset, then no superset of it can be either.
Hash tree used to store candidate itemsets (optimizes subset testing)