General Advance Review of Borgelt's Apriori RuleMining Algorithm Implementation : BusinessIntelligence and Data Science Perspective

In this paper we studied Finding Frequent Item Sets and Association Rules with the Apriori Algorithm. This comparative study refers to Borgelt's Apriori implementation version 6.12 in particular and original Apriori by R. Agrawal and R. Srikant. This paper covers introduction to Apriori, an Association Rule Mining algorithm and discussed related issues like Basic Notions of Items and Transactions, Support of an Item Set, Confidence of an Association Rule, Support of an Association Rule, and also discussed Target Types like Frequent Item Sets, Closed Item Sets, Maximal Item Sets, Generators/Free Item Sets, Association Rules, and Extended Rule Selection and their variations, for confidence, lift, conviction, χ2-Measure, p-value, and importance of Fisher's Exact test for table probability, χ2-Measure, information gain, support, and in the end item set selection methods are discussed. All these measures are very important to invent new data mining algorithms and they all have very strong utilization and application in data science and business intelligence. This paper also discuss The possible application of association rule mining from the perspective of Business Intelligence and Data Science. This review is important as this implementation is useful for Business Analysts, Data Scientists and data miners who want to exploit association rules to take decisions using R programming language to find solution for business problems.