In this chapter, we discussed association rule learning, or the approach of finding frequent sets of items in a transactional database and relating them to one another via probabilities. We learned that association rule learning was invented for market basket analysis but has applications in many fields, since the underlying probability theory and the concept of transactional databases are both broadly applicable.
We then discussed the mathematics of association rule learning in depth, and explored the canonical algorithmic approach to frequent itemset mining: the Apriori algorithm. We looked at other possible applications of association rule learning before trying out our own example on a retail dataset.
