Consumer Purchasing Pattern Analysis Using The Fp-Growth Algorithm To Support Business Decision-Making

Abstract

Customer purchasing pattern analysis is an important aspect of improving marketing effectiveness and optimizing sales strategies. One approach that can be employed is Market Basket Analysis (MBA), which aims to identify relationships among products that are frequently purchased together. This study applies the FP-Growth algorithm to analyze customer purchasing patterns at CV Aksara IT Solution, a company engaged in computer equipment sales. The study aims to identify customer purchasing patterns and utilize these patterns to optimize marketing strategies, enhance cross-selling opportunities, and improve the customer shopping experience. The FP-Growth algorithm was selected due to its efficiency in identifying frequent itemsets without requiring candidate generation, as in the Apriori algorithm. The dataset consisted of 722 transaction records collected over a period of two years and three months. The analysis identified 17 association rules. The strongest association rule showed that customers who purchased a BNC Connector also purchased a CCTV RG59 Cable + Power, with a support of 78.6% and a confidence of 100%. Similarly, customers who purchased a DC Male Connector also purchased a CCTV RG59 Cable + Power, with a support of 76.6% and a confidence of 100%. Another significant pattern indicated that customers who purchased a DC Male Connector also purchased a BNC Connector, with a support of 76.6% and a confidence of 100%. These findings can support CV Aksara IT Solution in improving marketing strategies through strategic product placement, product bundling, promotional discounts, and data-driven recommendation systems. Therefore, the application of the FP-Growth algorithm in Market Basket Analysis can provide valuable insights into customer purchasing behavior and contribute to more effective sales and marketing strategies.