Skip to main navigation Skip to search Skip to main content

Abstract

—Accurate sales forecasting is crucial for e-commerce businesses to optimize inventory, improve customer satisfaction, and maintain competitiveness. Motivated by the need for data-driven decision-making in small and medium-sized enterprises, this study integrates machine learning (ML) and Business Intelligence (BI) to enhance short-term sales prediction and strategic planning for a real-world clothing store. Using transactional data from October 2022 to April 2025, three supervised models—Linear Regression, XGBoost, and Random Forest—were applied to predict daily sales trends. Linear Regression outperformed the others in RMSE, MAE, and R2, providing accurate forecasts with minimal overfitting. An interactive Power BI dashboard with a star schema design offered insights into customer behavior, product performance, and sales projections, enabling informed strategic adjustments. The research demonstrates the value of combining ML and BI for data-driven forecasting and operational decision-making in e-commerce.

Original languageEnglish
Title of host publication2025 10th International Conference on Information Technology Trends, ITT 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages152-157
Number of pages6
ISBN (Electronic)9798331545758
DOIs
StatePublished - 2025
Event10th International Conference on Information Technology Trends, ITT 2025 - Dubai, United Arab Emirates
Duration: 6 Nov 20257 Nov 2025

Publication series

Name2025 10th International Conference on Information Technology Trends, ITT 2025

Conference

Conference10th International Conference on Information Technology Trends, ITT 2025
Country/TerritoryUnited Arab Emirates
CityDubai
Period6/11/257/11/25

Keywords

  • Business Intelligence
  • E-commerce Analytics
  • Machine Learning
  • Power BI Dashboard
  • Sales Forecasting

Fingerprint

Dive into the research topics of 'Data-Driven Sales Prediction and Strategic Insights for E-Commerce Through ML and BI Tools'. Together they form a unique fingerprint.

Cite this