FULL PDF
Abstract
The rapid evolution of data-driven decision-making necessitates robust methodologies capable of performing well even when data availability or consistency is a challenge. Transfer learning (TL) offers a powerful solution by enabling the adaptation of predictive models from one domain to another, especially when the target domain has limited labeled data. This paper explores the strategic application of transfer learning in cross-domain predictive analytics, with a focus on enhancing business intelligence (BI) and strategic forecasting. By analyzing key developments from prior literature and identifying use cases across finance, retail, and supply chain domains, the paper outlines the strengths and challenges of implementing TL frameworks for actionable forecasting. The paper concludes by offering pathways for future research and practical deployment in enterprise systems.
View more >>