Enhancing Retail Strategies through Apriori, ECLAT& FP Growth Algorithms in Market Basket Analysis

Abstract

"Market basket analysis" is a method employed in data mining to discover items that are commonly bought together by customers in a retail store. It is a crucial tool for retailers to understand consumer purchasing behavior and to improve their sales and marketing strategies. In this research paper, we present a comprehensive study on market basket analysis using three popular algorithms: Apriori, ECLAT, and FPGrowth. The paper begins with a brief synopsis of market basket analysis and the techniques adopted for itemset mining. We then introduce the dataset used in this study, which consists of real-life transaction data collected from a retail store. Next, we perform a thorough evaluation of the Apriori, ECLAT, and FPGrowth algorithms in terms of their computational time and the quality of the association rules generated. The results show that the FPGrowth algorithm is the fastest of the three algorithms, while the Apriori algorithm generates the most comprehensive and high-quality association rules.In addition, we also present a comparison of the performance of these algorithms that involve different assessment criteria like support, confidence, and lift. Our study highlights the importance of selecting the appropriate algorithm for market basket analysis depending on the specific requirements and constraints of the task. The paper concludes with an analysis on the limitations and future directions of research in this area. Overall, our study provides insights into the strengths and weaknesses of the Apriori, ECLAT, and FPGrowth algorithms and functions as a valuable resource for professionals and researchers in the field of market basket analysis.