A Machine Learning–Driven Web Platform for Business Analytics, Revenue Forecasting, and Inventory Optimization
International Journal of Advanced Research in Science Communication and Technology · 27.07.2026 · 343
DOI: 10.48175/ijarsct-37836 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Research in Science Communication and Technology, online erschienen am 27.07.2026. Verfasst von Tannu and Priyanka. Die Arbeit nennt 18 Literaturangaben.
Abstract (Originalsprache)
The rapid digitization of commerce has generated unprecedented volumes of transactional, behavioral, and operational data, creating both an opportunity and a challenge for small and medium-sized enterprises (SMEs) that lack dedicated data-science teams. This paper presents the design, architecture, and implementation of a machine learning–driven web platform that unifies three core business functions—descriptive analytics, revenue forecasting, and inventory optimization—within a single, accessible interface. The platform ingests heterogeneous data from point-of-sale systems, enterprise resource planning (ERP) logs, and web analytics, and processes it through an automated extract-transform-load (ETL) pipeline before feeding it to a suite of machine learning models. A hybrid forecasting module combining Long Short-Term Memory (LSTM) networks with the Prophet decomposition model is used to predict short- and medium-term revenue, while a gradient-boosted ensemble (XGBoost) coupled with a reorder-point heuristic drives inventory optimization. A React-based dashboard, backed by a Django REST Framework API, exposes these insights through interactive charts, alerts, and what-if simulations. Experimental evaluation on a twenty-four-month retail dataset shows that the hybrid forecasting model achieves a Mean Absolute Percentage Error (MAPE) of 6.8%, outperforming standalone ARIMA (11.4%) and standalone LSTM (8.9%) baselines, while the inventory module reduces simulated stock-out incidents by 27% and excess holding costs by 19% relative to a static reorder-point baseline. These results indicate that an integrated, web-accessible machine learning platform can materially improve data-driven decision-making for resource-constrained organizations.
Zitieren
APA
Priyanka, T. A. (2026). A Machine Learning–Driven Web Platform for Business Analytics, Revenue Forecasting, and Inventory Optimization. International Journal of Advanced Research in Science Communication and Technology, 343. https://doi.org/10.48175/ijarsct-37836
Vancouver
Priyanka TA. A Machine Learning–Driven Web Platform for Business Analytics, Revenue Forecasting, and Inventory Optimization. International Journal of Advanced Research in Science Communication and Technology. 2026:343. doi:10.48175/ijarsct-37836
BibTeX
@article{Priyanka2026,
author = {Priyanka, Tannu and},
title = {A Machine Learning–Driven Web Platform for Business Analytics, Revenue Forecasting, and Inventory Optimization},
journal = {International Journal of Advanced Research in Science Communication and Technology},
year = {2026},
pages = {343},
doi = {10.48175/ijarsct-37836}
}
EUVIMED