西北师范大学学报（自然科学版）2025，Vol.61Issue(4)：91-99,9.DOI:10.16783/j.cnki.nwnuz.2025.04.011
基于堆叠模型的多特征融合碳排放预测
Multi-feature fusion carbon emission prediction based on stacked models
摘要
Abstract
Accurate prediction of carbon emissions is crucial to help governments and enterprises formulate effective carbon reduction strategies and promote the realization of sustainable development goals.However,it is difficult for traditional prediction methods to effectively deal with the nonlinear and dynamic characteristics of carbon emission data.To address this problem,this paper applies a carbon emission prediction method based on the integrated learning stacked model,which integrates gradient boosting decision tree,random forest,and support vector regression as the base learner.Ridge regression as a meta-learner to further improve the accuracy and generalization of the model.The optimization and integration of each learner enables the stacked model to effectively capture the complex nonlinear relationships in carbon emission data.By optimizing and integrating the learning tools,the stacked model can effectively capture the complex nonlinear relationships in carbon emission data.The experimental results show that the proposed stacked model method is better than the single prediction model in multiple evaluation indexes,which significantly improves the prediction accuracy of carbon emissions,and is an efficient and reliable solution for carbon emissions prediction.关键词
碳排放预测/多特征融合/堆叠模型/岭回归Key words
carbon emission prediction/multi-feature fusion/stacked models/ridge regression分类
信息技术与安全科学引用本文复制引用
曹静,段博文,黄羿博..基于堆叠模型的多特征融合碳排放预测[J].西北师范大学学报（自然科学版）,2025,61(4):91-99,9.基金项目
甘肃省科技计划资助项目(21JR7RA120 （）
23JDKA0008) （）
