The rapid digital transformation of the banking industry has highlighted the critical role of data mining (DM) an‎d business intelligence (BI) dashboards in supporting efficient, accurate, an‎d transparent decision-making. This study develops a hybrid framework that integrates predictive models, clustering techniques, graph-based embeddings, an‎d fuzzy logic into a scalable dashboard system tailored for banking applications. Experimental results demonstrate significant improvements in fraud detection, customer segmentation, an‎d churn prediction, with F1-scores exceeding 0.85 an‎d decision-making time reduced by more than 60% compared to traditional BI systems. Moreover, the integration of explainable AI techniques, such as SHAP an‎d LIME, enhanced user trust an‎d regulatory compliance. By synthesizing insights from eight foundational studies, this research bridges theoretical advancements with practical implementation, establishing intelligent dashboards as indispensable infrastructures for modern digital banking.

10/28/2025 12:00:00 AM

87984

1404/08/07

Study on Data Mining for Dashboard Design for Banking Systems

(داده‌كاوي، هوش تجاري، سيستم‌هاي بانكي، طراحي داشبورد، جاسازي گراف، هوش مصنوعي قابل توضيح، مدل‌سازي اعتماد.)

(Data Mining, Business Intelligence, Banking Systems, Dashboard Design, Graph Embedding, Explainable AI, Trust Modeling.)