Machine learning integration in serverless ERP systems for financial forecasting and E-Commerce Applications

Onteddu, Koti Reddy (2025) Machine learning integration in serverless ERP systems for financial forecasting and E-Commerce Applications. World Journal of Advanced Research and Reviews, 26 (2). pp. 4301-4312. ISSN 2581-9615

This article examines the transformative integration of machine learning technologies with serverless computing architectures in enterprise resource planning systems, focusing specifically on financial forecasting and e-commerce personalization applications. The article explores how serverless frameworks enable organizations to deploy sophisticated ML models that analyze historical financial data, detect spending patterns, and predict revenue trends without the burden of infrastructure management. The article investigates system architectures that facilitate seamless integration with existing ERP modules while enabling dynamic scalability during peak processing periods. Through detailed case analyses across multiple industry sectors, the article documents the implementation approaches, algorithm selection criteria, and performance outcomes of these systems. The article's findings reveal that ML-powered financial forecasting delivers significant improvements in prediction accuracy while reducing infrastructure costs compared to traditional forecasting methods. Similarly, personalized recommendation systems implemented on serverless platforms demonstrate substantial enhancements in customer engagement metrics and conversion rates. The article addresses implementation challenges, including technical integration barriers, organizational adoption factors, and specialized skill requirements, while providing a framework for business value assessment. The article concludes by identifying promising research directions, including emerging ML techniques, integration with complementary serverless technologies, and potential cross-domain applications that could further extend the business impact of these implementations.