An Integrated Advanced Pattern Mining Framework for Strategic Retail Decision Intelligence

Abstract

Retail businesses increasingly rely on data-driven intelligence to support strategic decision-making; however, existing approaches often focus on isolated pattern mining or predictive models, limiting their practical impact. This paper proposes an Integrated Advanced Pattern Mining Framework that unifies frequent, temporal, and utility-driven patterns extracted from large-scale transactional data. The mined patterns are consolidated and ranked using an Integrated Pattern Score, enabling the identification of actionable and business-relevant knowledge. To operationalize these insights, a hybrid recommendation mechanism combining interpretable rule-based reasoning and convolutional neural network–based predictive learning is introduced. Experiments conducted on a real-world retail dataset comprising one year of transactional records from a Bangladeshi supermarket demonstrate that the proposed hybrid approach achieves an F1-score of 0.84. Under offline evaluation, the framework achieved the highest projected revenue contribution (10.6%) compared with the classical association rule mining baseline (7.5%), while also outperforming standalone recommendation models. The results confirm that integrating advanced pattern mining with hybrid learning models provides an effective and scalable solution for strategic retail decision intelligence.

Received: 30 April 2026 | Revised: 15 June 2026 | Accepted: 2 July 2026 

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

The data that support the findings of this study are openly available in SupershopBd1-Retail-Market-Dataset at https://github.com/Sohel-Rana-2/SupershopBd1-Retail-Market-Dataset.

Author Contribution Statement

Sohel Rana: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Funding acquisition. Md. Nazrul Islam Mondal: Conceptualization, Methodology, Validation, Writing – review & editing, Supervision. Md. Rabiul Islam: Conceptualization, Methodology, Validation, Writing – review & editing, Supervision.