Association Rules
Daniel T. Larose, Chantal D. Larose · 2014
This chapter considers some basic concepts and notation for association rule mining. The a priori algorithm for mining association rules, however, takes advantage of structure within the rules themselves to reduce the search problem to a more manageable size. Methods for affinity analysis, also known as market basket analysis, seek to uncover associations among these attributes. It seeks to uncover rules for quantifying the relationship between two or more attributes. The chapter examines association rules using flag data types. The generalized rule induction (GRI) methodology can handle either categorical or numerical variables as inputs, but still requires categorical variables as outputs. Association rule mining, however, can be applied in either a supervised or an unsupervised manner. Association rules could be used to help classify the voting preferences of citizens with certain demographic characteristics, in a supervised learning process.