Using Frequent Pattern Growth Algorithm in Analyzing Sudanese Shopping Behavior

Abstract

Data mining is an important technique to discover frequent items in customer shopping basket. Such information can be used as a basis for decisions about marketing activities such as promotional support, inventory control and cross-sale campaigns. The main objective of this research is analyzing Sudanese shopping behavior: a case study Aldooma supermarket and figuring out the commodities that are sold together, the data source is ORACLE database backup file which are collected from Aldooma supermarket’s sales points system, the results performed by using Frequent Pattern Growth Algorithm in Rapidminer tool. The researcher conducted many experiments and selected the best results which contained the longest frequent itemsets sold together. The best results represented in 12 Association rules with confidence 0.8 and support 0.004. The techniques which applied in this research are useful for the supermarket owner or the decision maker who can use them to grow their customer base and build stronger customer relationships to turn inventory into cash.