Application of Apriori Algorithm in Data Mining to Find Consumer Purchasing Patterns in Supermarkets

Abstract

The development of information technology has encouraged the use of transaction data in the retail world to gain deeper business insights. One method used in data mining is the Apriori algorithm, which is able to identify consumer purchasing patterns through association analysis between products. This study aims to apply the Apriori algorithm in finding product combination patterns that are often purchased together by consumers in supermarkets. The data used are sales transactions that have gone through a preprocessing process, including product category classification and transformation into a basket format. The results of the analysis show that products such as biscuits, detergents, and household appliances have the highest support values ​​individually, while product combinations such as (milk, drinks, soap & shampoo, cosmetics) also appear consistently in transactions. The application of the Apriori algorithm with a certain minimum support threshold is able to produce frequent itemsets that represent consumer shopping habits. These findings can be used to develop promotional strategies, product arrangement, and category-based recommendation systems. Thus, this study proves that the Apriori algorithm can be used effectively in the context of data mining to support business decision making in the retail sector, especially supermarkets.