Association Rule Mining

Association Rule Mining is a Rule-Based Machine Learning method, to analyze dataset for patterns or co-occurrences to create list of Association Rules and is a method for discovering relations between data items, creating Frequent Itemsets. Apriori Algorithm is the most famous example of this method.


It's the discovering of statistically significant associations in data.



  • Support and confidence are common metrics used to evaluate association rules. Support measures the frequency of a particular itemset, while confidence measures the likelihood of the consequent in the rule being true, given that the antecedent is true.
  • The Apriori algorithm uses a level-wise search strategy to discover frequent item sets and generate association rules. It prunes the search space by applying the downward closure property to avoid generating unnecessary candidate item sets.
  • Association rule mining is widely used in various domains including market basket analysis in retail, Recommender Systems, and web usage mining to discern patterns in user behavior.