MACHINE LEARNING IN BUSINESS INTELLIGENCE: FROM DATA MINING TO STRATEGIC INSIGHTS IN MIS

Abstract

The convergence of machine learning (ML) and business intelligence (BI) has transformed the landscape of management information systems (MIS), enabling a shift from static reporting and descriptive analytics to predictive and prescriptive decision support. This study examines the critical factors influencing the adoption and effectiveness of ML-driven BI systems within MIS frameworks, focusing on organizations in Bangladesh. Grounded in the Technology Acceptance Model (TAM), the Technology-Organization-Environment (TOE) framework, strategic alignment theory, and sociotechnical systems theory, the research adopts a quantitative approach using data collected from 312 professionals across various industries. Structural equation modeling (SEM) was employed to analyze hypothesized relationships among eight key variables.The findings reveal that all hypothesized relationships were statistically significant (p < .001). Perceived usefulness was the strongest predictor of adoption (β = 0.63), confirming that users are more likely to adopt ML tools when they believe those tools enhance decision-making and work performance. Strategic alignment between ML initiatives and business goals also had a strong positive effect on system effectiveness (β = 0.57), while leadership support (β = 0.61) and a data-driven culture (β = 0.52) emerged as critical enablers of system usage and impact. Interpretability of ML models (β = 0.54) significantly influenced user trust and system acceptance, and ethical governance (β = 0.48) contributed meaningfully to organizational readiness. On the technical side, infrastructure readiness (β = 0.59) and integration capability (β = 0.62) had the most substantial effects on system performance and decision-making efficiency.These results suggest that the success of ML-BI implementation is determined by a blend of technical robustness, strategic alignment, ethical oversight, and organizational maturity. The study contributes to the literature by validating a comprehensive, multidimensional model for ML-BI integration within MIS in a developing economy context. It also provides practical guidance for organizations seeking to deploy intelligent decision-support systems. Ultimately, this research affirms that achieving value from ML-enhanced BI requires more than algorithms—it requires leadership, infrastructure, trust, and strategic clarity.