An overview of association rule mining (arm) algorithms for market basket analysis (mba)

Data mining is a technique that has become a widely accepted procedure for organizations in sourcing for data and processing it for decision making. Association rule mining is an aspect of data mining that has revolutionized the area of predictive modelling paving way for data mining technique to become the recommended method for business owners to evaluate organisational performance. Association rule mining (ARM) give top managers the opportunity to make informed business decisions by anticipating future movements and behaviours of customers. Market basket analysis (MBA) is paving the path in business as it has become the most widely used areas of data mining in marketing. This study defines association rule mining as a technique used to extract important patterns from existing information which enables better decision making in an establishment. MBA is a marketing strategy used by various organizations to find the optimal environments to advertise merchandise. A market basket comprises of products picked by a customer during the visit to a superstore. These work specifically focus on association rule mining algorithms and its application to MBA. This paper presents a critical review of various ARM algorithms, comparing each of the algorithms, and considering the merit and demerit of each. The outcome of the study shows that choosing an ARM algorithm for MBA depends on the data set size and the application area of MBA that the algorithm will be used, this is because according to the no free lunch theorem which state that no algorithm is guaranteed to outperform others in all domains hence the need for this study, to determine the performance of the algorithms. The study concluded by recommending a hybrid algorithm to be used for ARM in MBA systems.