Mamdani Fuzzy on Improving the Minimum Confidence Final Value Accuracy in the Apriori Algorithm

EasyChair Preprint no. 11140

Abstract

Finding rules from data is an active area of ​​research in artificial intelligence. Active research is carried out based on regular exploration of item sets, eventually leading to the creation of rules. Many algorithms provide suggestions to fix different problems. Many optimization methods aim to minimize frequent itemset generation and analysis time for data reduction. The apriori algorithm is a data mining algorithm  used to analyze databases based on their frequency, based on an association rule learning system. In this paper, we modified the apriori algorithm in such with fuzzy Mamdani for generate package items within a minimum support value. The generated item sets can also help the decision maker to forming new packages for the customers. At the same time, the improved Apriori algorithm has a great improvement in operating efficiency when it conducts association rule analysis on the data set with a large amount of item indexes. Fuzzy Mamdani is here as a supporting method that can increase the accuracy of confidence values and produce a model with the hope of being able to eliminate itemset based on the a priori algorithm. The proposed procedure can reduce the time consumption of Apriori based association rule mining by up to 50% while still maintaining substantial similarity in output, contingent upon user input.

Keyphrases: Apriori algorithm, confidence value, Fuzzy Mamdani

@Booklet{EasyChair:11140,
  author = {Erman Sibarani and Erna Budhiarti Nababan and Syahril Efendi},
  title = {Mamdani Fuzzy on Improving the Minimum Confidence Final Value Accuracy in the Apriori Algorithm},
  howpublished = {EasyChair Preprint no. 11140},

  year = {EasyChair, 2023}}