OPTIMIZING E-COMMERCE SALES VOLUME AND EVALUATING EFFICIENCY USING ARTIFICIAL INTELLIGENCE-BASED MODELS
Keywords:
artificial intelligence, e-commerce, machine learning, XGBoost, LSTM, sales forecasting, demand forecasting, digital economy, Uzbekistan, sales optimization.Abstract
The article examines the potential of artificial intelligence-based models for optimizing sales volumes and improving the economic efficiency of e-commerce enterprises in Uzbekistan. The study analyzes recent trends in the development of Uzbekistan’s e-commerce market and identifies the main factors influencing online sales. Particular attention is paid to machine learning models such as XGBoost and LSTM for demand forecasting, inventory optimization, personalized marketing, and pricing decisions. The research also considers the institutional and infrastructural characteristics of Uzbekistan’s developing digital economy. Based on the analysis, practical recommendations are proposed for integrating AI technologies into e-commerce management and improving the competitiveness of Uzbek digital businesses.
References
1. National Statistics Committee of the Republic of Uzbekistan. E-commerce Trade Volume (Annual), SIAT Database, 2016–2024. Updated July 8, 2025.
2. Uzbekistan National Statistics Committee. The Volume of Retail Trade Turnover Exceeded 251 Trillion Soums, July 29, 2026.
3. Uzbekistan Agency for Perspective Projects. Data on the development of e-commerce in Uzbekistan, 2025–2026. Reported e-commerce sales of approximately 22.7 trillion soums and 115 e-commerce platforms in 2025.
4. UzA. E-commerce is growing. April 14, 2026. Information on Uzbekistan’s e-commerce development policy and the 2026–2030 target of 30 trillion soums in e-commerce volume.
5. Zhang, M., & Wu, L. (2025). Sales Forecasting and Data-Driven Marketing Strategies for E-Commerce Platforms Using XGBoost. International Journal of Intelligent Information Technologies, 21(1), 1–21. DOI: 10.4018/IJIIT.382563.
6. Du, C., & Li, Y. (2025). E-commerce sales forecasting based on deep learning algorithm. Procedia Computer Science, 261, 1157–1164. DOI: 10.1016/j.procs.2025.04.699.
7. Wang, C., & Wang, J. (2025). Research on E-Commerce Inventory Sales Forecasting Model Based on ARIMA and LSTM Algorithm. Mathematics, 13(11), 1838. DOI: 10.3390/math13111838.
8. Zhang, H., & Lu, X. (2026). Sales Forecasting in E-commerce Scenarios Based on Improved Bidirectional Long Short-Term Memory Network. Proceedings of the 2025 6th International Conference on Big Data Economy and Information Management. DOI: 10.1145/3800000.3800207.
