A Systematic Review of Artificial Intelligence-Driven Insight Recommendation Systems in Power BI: Enhancing Business Intelligence and Decision-Making

Abstract

This study examines how Artificial Intelligence (AI)–driven insight recommendation systems can enhance Business Intelligence (BI) and organizational decision-making within Microsoft Power BI environments. Traditional BI dashboards are largely descriptive and struggle to deliver proactive, actionable insights as enterprise data becomes increasingly complex and heterogeneous. To address this limitation, the study explores the integration of AI techniques specifically machine learning (ML), natural language processing (NLP), and predictive analytics into BI platforms. A systematic literature review was conducted following PRISMA-guided systematic review principles, analyzing 84 peer-reviewed articles retrieved from major academic databases using keywords such as AI-driven analytics, Power BI recommendation systems, intelligent dashboards, and predictive business intelligence. The selection focused on studies addressing AI-enabled BI tools, automated insight generation, dashboard intelligence, and enterprise decision-support systems (DSS). The analysis reveals several key trends: AI-enhanced BI systems improve decision accuracy, reduce user cognitive load, and support real-time operational intelligence. NLP facilitates conversational analytics, ML techniques uncover latent data patterns, and predictive analytics enables forward looking recommendations. Despite these benefits, challenges remain in terms of system integration, data quality, model interpretability, and user trust. The study concludes that AI-driven insight recommendation systems represent a significant evolution of BI in Power BI, transforming dashboards from descriptive reporting tools into prescriptive decision-support platforms. The primary contribution of this research is the synthesis of existing evidence into an integrated conceptual framework for designing AI-enhanced BI systems, highlighting both their transformational potential and key implementation challenges. Keywords: p