An Approach to Explore Consumer Behavior Patterns in Retail Markets Using Market Basket Analysis

This is the authors' abstract. We don't add key points for this paper.

Big data is becoming an integral part of everyday life. Data can be generated through various sources, and analyzing this data to generate revenue is the biggest challenge. The growth of grocery stores in the online and offline market requires retailers to analyze customer purchase behavior. Effective analysis can improve service quality, profitability, and customer satisfaction. This paper focuses on Market Basket Analysis (MBA), an efficient technique that identifies customer purchase behaviors. AAGI an Apriori‐based algorithm is developed and used to analyze categorical data collected from grocery stores. The dataset consists of 10,233 transactions collected from the Raigad region Maharashtra, India and includes 125 unique items across dairy products, fruits, processed food, and vegetables. Association rules were generated using support values of 2%, 3%, 3.5%, and 4% and confidence values of 20%, 30%, and 40%. The best results were observed with the support of 3.5% and confidence of 30%, which produced 18 profound association rules. These findings inform targeted marketing strategies and cross‐selling opportunities.