Apriori Algorithm

Updated 02 Oct 2026

•
What it is
- •
  miens frequenct itemsets using the property that every subset of a frequent itemset must also be frequent
•
Procedure
- •
  count frequent single items.
- •
  generate larger candidates from frequent smaller itemsets.
- •
  prune candidates having an infrequent subset, count support, and repeat.
- •
  generate association rules from the frequent itemsets and filter using chosen thresholds.
•
Checked example
- •
  a minimum support of $50%$ across five transactions requires at least $\lceil0.5(5)\rceil=3$ occurrences.
- •
  Bread–Milk occurs $3$ times; Bread–Butter and Butter–Milk each occur $2$ times.
- •
  only Bread–Milk survives as a frequent pair, so no frequent triple can be generated.
•
Corrections
- •
  Minimum support is a chosen parameter, not a threshold uniquely derived by the support formula.
- •
  Standard Apriori scans are associated with candidate levels; it need not perform a separate database scan for every individual candidate.
•
Related
- •
  Unavailable reference
•

Linked references 3