Introduction

Abstract

The idea of extracting knowledge from sets of data emerged back in the 90s, motivated by the decision support problem faced by several retail organizations that due to several technological advances, were able to store massive amounts of sales data. At that point, the research field of knowledge discovery started as an active area of investigation, and data mining, one of the most challenging steps of the process, was meant to provide efficient algorithms and techniques to automatize the exploratory analysis of the data. In its simplest form, such data is viewed as a set of transactions where each transaction is a set of items (attributes of the database), that is, a simple binary relation. This representation is known popularly as market-basket data.