A Novel Approach to Enhance the Efficiency of Apriori Algorithm

Abstract

Data mining is the process of obtaining valuable or significant information from a large-scale database. One significant
area of research in the field of data mining is association rules mining. Apriori algorithm is one of the classical algorithms
in the association rule mining field. This research analyses the basic ideas and shortcomings of the Apriori algorithm and
compares several different styles of its major improvement strategies. Then it suggests an improved version of the Apriori
algorithm that utilizes a dataset summarization method, an optimized database mapping technique, an intersection operation,
and a joined optimization strategy. These enhancements aim to address the low performance and efficiency by reducing the
generation of candidate itemsets and minimizing the execution time. This addresses the issues of generating numerous candidate itemsets and repeatedly scanning the transaction database. After implementing the optimized algorithm, to verify its effectiveness, it has been applied to a groceries dataset, which is for market basket analysis. The improved Apriori algorithm demonstrated significant enhancements over the original algorithm in terms of reduced candidate itemsets and running time, leading to improved algorithm efficiency.