THE INFLUENCE OF MARKETING MIX ON CONSUMER PURCHASE PATTERNS USING THE APRIORI DATA MINING ALGORITHM

Abstract

The increasing volume of sales transaction data in retail and marketplace environments presents an opportunity to extract valuable insights for decision-making; however, such data are often underutilized. This study aims to analyze consumer purchasing patterns using the Apriori algorithm and to examine the influence of the marketing mix (product, price, place, and promotion) on purchasing decisions that shape these patterns. This research employs a quantitative approach by integrating data mining and statistical analysis. Transaction data are processed using the Apriori algorithm through RapidMiner to generate association rules and identify frequent itemsets. In addition, questionnaire data are analyzed using multiple linear regression to evaluate the effect of marketing mix variables on purchasing decisions. The results show that product, price, place, and promotion simultaneously have a significant effect on purchasing decisions. Partially, product (t = 2.622; p = 0.011), price (t = 4.738; p = 0.000), and place/distribution (t = 2.239; p = 0.029) have a significant positive effect, while promotion does not have a significant effect (t = 1.486; p = 0.143). The Apriori analysis reveals dominant purchasing patterns that can be translated into practical marketing strategies, such as product bundling and layout optimization. This study contributes by integrating association rule mining with marketing mix analysis to provide both predictive patterns and explanatory insights. However, the findings should be interpreted with caution due to data limitations, including a relatively small sample size (n = 148) and a short observation period of three months during peak season, which may limit generalizability. Despite these constraints, the results offer practical implications for optimizing marketing strategies and contribute theoretically to interdisciplinary research in data mining and consumer behavior.