In this work we discuss a supervised learning approach for identification of frequent itemsets and association rules from transac-tional data. This task is typically encountered in market basket analysis, where the goal is to find subsets of products that are frequently purchased in combination. In this work we compare the traditional approach and the supervised learning approach to find association rules in a real-world retail data set using two well known algorithm, namely Apriori and PRIM.
展开▼