FP-Growth Frequent Itemset Miner

Description

Mine bounded transaction itemsets with actual FP-trees and weighted conditional pattern bases, counting each item once per transaction and retaining empty transactions in support denominators.

Frequent-itemset mining counts groups of items occurring together in transactions. FP-growth stores shared frequent-item prefixes in a compact tree and explores conditional pattern bases. Transaction support differs from counting repeated item tokens or measuring a rule’s confidence.^[1]

Mine bounded transaction itemsets with actual FP-trees and weighted conditional pattern bases, counting each item once per transaction and retaining empty transactions in support denominators.

When to use this tool

Use this tool when you need the stated calculation or classification with explicit inputs, intermediate evidence, and boundary conventions.

How to use the tool and how it works

Supply transactions, minimumCount, maximumSize in the field shapes below.
Run the calculation. Correct format, shape, range, or undefined-statistic errors before interpreting a result.
Read transactionCount, initialTreeNodeCount, frequentItemsets. Use the conventions below.

Deduplicate items within each transaction, count frequent items, and build a frequency-ordered prefix tree. Mine weighted conditional pattern bases recursively up to maximumSize, then return lexically sorted itemsets and transaction-based support with the original tree's nonroot node count.^[2]

Input field	Meaning and units	Accepted input
transactions	Transactions	Required. Array of Array of Text; minimum UTF-16 length 1; maximum UTF-16 length 64.; maximum items 16.; minimum items 1; maximum items 512.
minimumCount	Minimum number of transactions containing the itemset.	Required. Safe integer; minimum 1; maximum 512.
maximumSize	Largest returned itemset cardinality.	Required. Safe integer; minimum 1; maximum 16.

Output field	Interpretation	Returned shape
transactionCount	Transaction Count	Required. Safe integer; minimum 1; maximum 512.
initialTreeNodeCount	Nonroot nodes in the original FP-tree, excluding conditional trees.	Required. Safe integer; minimum 0; maximum 8192.
frequentItemsets	Frequent Itemsets	Required. Array of Object; no unknown keys.; maximum items 4096.
frequentItemsets[].items	Items	Required. Array of Text; minimum UTF-16 length 1; maximum UTF-16 length 64.; minimum items 1; maximum items 16.
frequentItemsets[].count	Count	Required. Safe integer; minimum 1; maximum 512.
frequentItemsets[].support	Itemset count divided by all input transactions, including empty baskets.	Required. Finite number; maximum 1; strictly above 0.

json

{
  "input": {
    "transactions": [
      [
        "bread",
        "milk"
      ],
      [
        "bread",
        "milk",
        "eggs"
      ],
      [
        "bread"
      ],
      [
        "eggs",
        "milk"
      ]
    ],
    "minimumCount": 2,
    "maximumSize": 2
  },
  "output": {
    "transactionCount": 4,
    "initialTreeNodeCount": 5,
    "frequentItemsets": [
      {
        "items": [
          "bread"
        ],
        "count": 3,
        "support": 0.75
      },
      {
        "items": [
          "eggs"
        ],
        "count": 2,
        "support": 0.5
      },
      {
        "items": [
          "milk"
        ],
        "count": 3,
        "support": 0.75
      },
      {
        "items": [
          "bread",
          "milk"
        ],
        "count": 2,
        "support": 0.5
      },
      {
        "items": [
          "eggs",
          "milk"
        ],
        "count": 2,
        "support": 0.5
      }
    ]
  }
}

Limitations and assumptions

Repeated items in a basket count once, while duplicate transactions count separately. Empty transactions remain in the support denominator. At most sixteen distinct items, 4,096 output patterns, and a bounded work budget are supported; association rules are not generated here.
The field tables state structural bounds; cross-field constraints and bounded search or output budgets also apply. Unknown keys are rejected where the object contract disallows them. Input errors are distinct from null fields and deliberate invalid controls in a successful output.

Alternative or Complementary methods

Choose the statistic according to the question and data conventions. Inspect the supplied series or matrix and the stated assumptions before comparing outputs from implementations with different defaults.

References

Han, Pei and Yin (2000): Mining Frequent Patterns without Candidate Generation — Simon Fraser University
Tool behavior and example: FP-Growth Frequent Itemset Mining Tool — tool.io

Similar or alternative tools

Apriori
Find item combinations whose support meets a selected threshold.
Association rules
Derive association rules and calculate support, confidence, and lift.