Association is one of the methods in data mining, which is looking for a similar pattern in data transactions or certain items that often appear. So that the supermarket manager can determine in making decisions on the pattern of items purchased together and often appears in the database. In order to generate patterns from association rules, there are several algorithms that can be used. First pioneered by the AIS Algorithm, then increased memory performance by the Apriori algorithm. The Apriori algorithm process involves several steps. Such as determining the minimum support, confidence, candidate itemset. The process continues until the pattern is obtained according to the support and confidence criteria of the minimum threshold value. Most of the algorithms used to find patterns of association rules start with the Apriori algorithm, then progress to the Apriori series algorithm. But there are two obstacles to the Apriori algorithm. One of them is the process of making complex candidates that mostly require time, space, and memory. Another problem is the repeated scanning of large-scale databases.
Copyrights © 2020