Abstract

Optimizing inventory management, refining marketing strategies, and enhancing personalized recommendations remain critical challenges in e-commerce. This study introduces a novel hybrid approach that integrates association rule mining with data-driven visualization techniques to extract and interpret consumer purchasing patterns. Unlike conventional studies that focus solely on frequent item set mining, the presented approach enhances interpretability by employing graph-based visual analytics alongside the Apriori algorithm, enabling more intuitive insights into product correlations. By leveraging publicly available transactional datasets, the study systematically applies association rule mining to identify frequently co-purchased items, evaluating their relationships through key metrics—support, confidence, and lift. The implementation, conducted using Python- based frameworks, demonstrates how interactive visual tools, such as heatmaps and network graphs, facilitate better decision-making for e- commerce businesses. Furthermore, this research explores the potential integration of machine learning models to enhance predictive accuracy, offering a foundation for real-time recommendation systems. The findings highlight the significance of combining traditional data mining with visualization-driven analytics to improve customer engagement and revenue generation.

Authors

Twinkle Shaileshbhai Panchal
Veer Narmad South Gujarat University, India

Keywords

Association Rule Mining, Apriori Algorithm, Data-Driven, Machine Learning