2026-10-10T00:14:34Z https://www.scidb.cn/oai

10.57760/sciencedb.010qw 2026-09-07T09:30:12Z

2026-09-07 XGBoost, LightGBM, and CatBoost. doi:10.57760/sciencedb.010qw en Accurate forecasting of building energy consumption is important for improving energy management and supporting more efficient building operation. In this study, we investigate a sequential metaheuristic hyperparameter optimization framework for three widely used gradient-boosting models, namely XGBoost, LightGBM, and CatBoost. Two optimization pathways, Grey Wolf Optimizer–Particle Swarm Optimization (GWO–PSO) and Ant Lion Optimizer–Moth-Flame Optimization (ALO–MFO), are evaluated under consistent experimental conditions. XGBoost; LightGBM; and CatBoost.; GWO–PSO; ALO–MFO YUKOKHWA WANGYANG EMBARGO https://creativecommons.org/licenses/by-nc-nd/4.0/ dataset Science Data Bank