A Blockchain Based Framework for Improving Data Quality and Predictive Accuracy in Business Intelligence Systems

The increasing dependence on predictive analytics within Business Intelligence (BI) systems has underscored critical issues concerning data quality, integrity, and security, which in turn affect the reliability of analytical outcomes. This research introduces a blockchain-based framework designed to enhance data quality and predictive accuracy in BI environments. The study examines the integration of blockchain’s core attributes transparency, immutability, and decentralized consensus as mechanisms for strengthening the validity and traceability of data used in predictive modeling. A conceptual model was developed and validated through Structural Equation Modeling (SEM) using SmartPLS, based on survey data collected from professionals with expertise in blockchain and data mining. Empirical findings demonstrate that blockchain integration significantly improves both data quality and the efficiency of data mining processes, ultimately leading to enhanced predictive accuracy in BI applications. Notably, the analysis confirms the mediating role of data quality, emphasizing its foundational importance in achieving reliable predictive outcomes. The study contributes to existing literature by offering a novel perspective on blockchain’s utility in data-centric systems and provides actionable insights for organizations aiming to optimize their BI infrastructure. The proposed framework underscores blockchain’s potential as a strategic enabler of trustworthy and high-performance predictive analytics, paving the way for broader implementation in enterprise intelligence solutions.