How to Determine Minimum Support in Association Rule

Abstract

The growth of increasingly complex data now raises new challenges in the world of technology. Large volumes of data store a lot of knowledge that can help in the decision-making process. One way to find knowledge in big data is by the Association Rule. The association rule is a technique in data mining that can produce rules based on the frequency of items appearing from a transactional database. One thing that is critical in the association rule is the determination of the minimum support value used to determine which items will be included in the formation of rules. If the minimum support value that is set is too small, it causes too many items to be involved in establishing the rules. Conversely, if the minimum support is too large, the number of items involved in forming the rule is too small. The problem in determining the minimum support value greatly affects the accuracy of the resulting rule. In this paper, various methods will be discussed to determine the minimum value of support through study literature based on related research. In addition, it explains research opportunities that can be done in the future about the minimum value of support determination in the association rule.