Enhancing Retail Decision-Making through Market Basket Analysis: An Apriori Algorithm Approach
Prof Nilesh Avinash Joshi*1, Dr Soumitra Das*2, Miss Pranali More*3 ,Miss Anjali Ghangurde*4, Mr Kamalesh Bhosale*5,Mr Shreyas Bansode*6 ,Mr Nilay Sharma *7
*1 Assistant Professor Artificial Intelligence & Data Science Enginnering Department, Indira College of Engineering & management, Pune. India
*2 Head of Computer Engineering Department & Vice Principal Indira College of Engineering & Management, Pune,India
*3 Student,Third Year Artifical Intelligence & Data Science Engineering Indira College of Engineering & Management ,Pune ,India
*4 Student,Third Year Artifical Intelligence & Data Science Engineering , Indira College of Engineering & Management ,Pune ,India
*5 Student,Third Year Artifical Intelligence & Data Science Engineering , Indira College of Engineering & Management ,Pune ,India
*6 Student,Third Year Artifical Intelligence & Data Science Engineering , Indira College of Engineering & Management ,Pune ,India
*7 Student,Third Year Artifical Intelligence & Data Science Engineering , Indira College of Engineering & Management ,Pune ,India
Abstract
Retail businesses constantly seek data-driven insights to optimize sales strategies and enhance customer experience. Market Basket Analysis (MBA), a key data mining technique, uncovers hidden patterns in consumer purchasing behavior by identifying associations between products. This study employs the Apriori algorithm, a widely used approach for frequent itemset mining, to analyze transactional data and extract meaningful correlations. By leveraging real-world retail datasets, this research highlights how retailers can optimize product placement, crossselling, and promotional strategies. The findings demonstrate the effectiveness of Apriori in improving decision-making by providing actionable insights into consumer buying patterns. Additionally, the study discusses computational efficiency challenges and potential enhancements to the algorithm for large-scale data processing. The results offer valuable implications for retailers, enabling data-driven inventory management, pricing strategies, and personalized recommendations to enhance customer satisfaction and profitability.
