Explainable Ensemble Learning for Accurate Hydroponic Stock and Sales Prediction
Tạp chí Khoa học Lạc Hồng · 25.06.2026 · 1 (26) · 25-29
DOI: 10.61591/jslhu.26.1085 ↗Kurz gefasst
Fachartikel in der Zeitschrift Tạp chí Khoa học Lạc Hồng, online erschienen am 25.06.2026. Verfasst von Dang Dang Khoa und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Growing global food demand calls for more sustainable agricultural practices. While hydroponic farming uses resources efficiently, predicting stock levels and sales remains difficult due to fluctuating conditions. This study proposes an optimized ensemble framework that integrates linear regression, random forest, and XGBoost, with weights refined through an evolutionary algorithm. Evaluated using root mean square error (RMSE) and mean absolute error (MAE), the model outperforms individual approaches, reducing RMSE by 43.82% for stock prediction and 55.3% for sales forecasting. A model-agnostic local explanation method is used to interpret the results, highlighting harvested crop volume and historical sales as key predictors. Future work will focus on integrating real-time data to improve adaptability and forecasting performance.
Zitieren
APA
Khoa, D. D., Diem, P. T. H., Triet, P. M., & Nam, P. V. (2026). Explainable Ensemble Learning for Accurate Hydroponic Stock and Sales Prediction. Tạp chí Khoa học Lạc Hồng, 1(26), 25-29. https://doi.org/10.61591/jslhu.26.1085
Vancouver
Khoa DD, Diem PTH, Triet PM, Nam PV. Explainable Ensemble Learning for Accurate Hydroponic Stock and Sales Prediction. Tạp chí Khoa học Lạc Hồng. 2026;1(26):25-29. doi:10.61591/jslhu.26.1085
BibTeX
@article{Khoa2026,
author = {Khoa, Dang Dang and Diem, Pham Thi Hong and Triet, Pham Minh and Nam, Phan Van},
title = {Explainable Ensemble Learning for Accurate Hydroponic Stock and Sales Prediction},
journal = {Tạp chí Khoa học Lạc Hồng},
year = {2026},
volume = {1},
number = {26},
pages = {25--29},
doi = {10.61591/jslhu.26.1085}
}
EUVIMED