Association Rules Analysis Using Algorithm Apriori And Fuzzy Normalization

Apriori algorithm is an algorithm to find a combination of items in a dataset that has a relationship value between items by paying attention to the minimum support and minimum confidence values. A common problem in the a priori algorithm is the formation of an item set when getting a large number of items from a dataset. The more items in a dataset, the more itemset will be formed. This will lead to increased storage space and poor algorithm performance. Some techniques can be applied to help correct these deficiencies, namely fuzzy logic. Fuzzy logic can normalize the items in the dataset by grouping the data according to the given rules. The formation of association rules using an a priori algorithm with fuzzy normalization obtained a total of 63 rules, a support value of 6.80%, and a confidence value of 88.4%. While the conventional a priori algorithm obtained a total of 13 rules, a support value of 6.80%, and a confidence value of 85%. Based on the results obtained, it can be concluded that the fuzzy normalization process on the a priori algorithm has a fairly good impact. The results obtained are not much different, but the difference can be seen from the number of rules and itemset formed.