Summary

Market basket analysis is used to analyze and extract insights from transaction or transaction-like data that can be used to help drive growth in many industries, most famously the retail industry. These decisions can include how to lay out the retail space, what products to discount, and how to price products. One of the central pillars of market basket analysis is the establishment of association rules. Association rule learning is a machine learning approach to uncovering the associations between the products individuals purchase that are strong enough to be leveraged for business decisions. Association rule learning relies on the Apriori algorithm to find frequent item sets in a computationally efficient way. These models are atypical of machine learning models because no prediction is being done, the results cannot really be evaluated using any one metric, and the parameter values are selected not by grid search, but by domain requirements specific to the question of interest...