Association Rules

Association Analysis is a method for discovering interesting relationships hidden in large datasets. Given a set of transactions, it finds rules that will predict the occurrence of an item based on the occurrences of other items in the transaction.

Rules are of the form A -> B (e.g., {onions, potatoes} - > {burger}).

The concepts of support and confidence are used to identify how strong the discovered rules are.

Support is the fraction of transactions that contain both A and B:
Support(A,B) = P(A,B)

Confidence is the fraction of transactions, where items in B appear in transactions that contain A:
Confidence(A,B) = P(B|A)

We are using Apriori algorithm to identify frequent item sets. It proceeds by identifying the frequent individual items in the database and extending them to larger item sets while the items satisfy the minimum support requirement (frequency of items in the database). The frequent item sets determined by Apriori are then used to determine association rules.