Real Market Basket Analysis using Apriori and Frequent Pattern Tree Algorithm

Recently, data mining has been implemented in various fields, including business and telecommunications. Data mining is a technique for extracting and detecting patterns in massive data sets that combines machine learning, statistics, and database systems. One of the most important use-cases in data mining is finding the high-frequency patterns between the set of itemset called association rules. Association rule mining is a well-researched technique for finding some relations between variables in large databases. This paper aims to measure the performance of the Apriori and Frequent Pattern Tree algorithms by comparing them using several points of comparison. Then we compared the outputs, whether they produce the same or different rules, to find out whether the way the two algorithms work is similar or not. After that, we looked for the itemsets that best match the reality in the market by giving them to a user who had transaction data from his spare parts shop.