Data Mining: Mining Frequent Patterns, Associations Rules, and Correlations

In this article we provide a survey of frequent pattern mining, a fundamental data mining task that deals with the search of recurring regularities in large data sets. Frequent patterns can take different forms depending on the type of data analyzed, for example, frequent itemsets (set of items), frequent sequences, or frequent sub-graphs. We focus here on frequent itemsets and associations between itemsets. We start with a short overview on data mining and the area of frequent pattern mining, then, after reviewing the basic concepts and definitions underlying the problem of frequent itemsets, we introduce the main interestingness metrics used to evaluate the goodness of a mined association and the most important classical algorithms to tackle this mining task.