11.7. Summary

Association analysis is a set of tools used to find interesting relationships in a large set of data. There are two ways you can quantify the interesting relationships. The first way is a frequent itemset, which shows items that commonly appear in the data together. The second way of measuring interesting relationships is association rules. Association rules imply an if..then relationship between items.

Finding different combinations of items can be a time-consuming task and prohibitively expensive in terms of computing power. More intelligent approaches are needed to find frequent itemsets in a reasonable amount of time. One such approach is the Apriori algorithm, which uses the Apriori principle to reduce the number of sets that are checked against the database. The Apriori principle states that if an item is infrequent, then supersets containing that item will also be infrequent. The Apriori algorithm starts from single itemsets and creates larger sets by combining sets that meet the minimum support measure. Support is used to measure how often a set appears in the original data.