From Business Intelligence to Learning Intelligence: A Big Data Analytics Framework for Educational Data Mining in Higher Education

Universities generate vast volumes of operational and behavioral data through learning management systems, student information systems, and digital assessment tools, yet many institutions lack a coherent architecture for converting this data into decisions. Enterprise Business Intelligence (BI), by contrast, has spent two decades developing mature, risk-aware pipelines for exactly this problem in commercial settings. This paper argues that Learning Analytics (LA) and Educational Data Mining (EDM) can benefit substantially from the direct transfer of enterprise BI architecture, cloud-security baselines, and Enterprise Risk Management (ERM) governance principles, rather than developing analytics infrastructure in isolation from established enterprise practice. Drawing on enterprise-sector research on big data analytics for business intelligence, cloud computing security within an ERM framework, and the strategic integration of ERM for competitive advantage, this paper proposes a four-layer framework — data, cloud/storage, analytics, and decision layers, overlaid with a risk-and-security layer and a governance layer — for institutional learning-analytics deployment. The framework maps enterprise BI components onto their EDM/LA equivalents, translates baseline cloud-security requirements into a student-data context, and reframes ERM's competitive-advantage rationale as an institutional rationale for treating learning analytics as a strategically governed capability rather than an ad hoc IT initiative. The paper concludes with adoption barriers, cost and scalability considerations, and recommendations for institutions beginning or scaling learning-analytics programs.