Comparative Analysis of Machine Learning Algorithms to Predict Municipal Solid Waste

This repository provides the curated dataset and full reproducible workflow for the study Comparative Analysis of Machine Learning Algorithms to Predict Municipal Solid Waste. It includes the cleaned source--day panel dataset, preprocessing, feature engineering, model training, temporal backtesting, baselines, uncertainty estimation, and scripts/notebooks to reproduce the reported tables and figures. The study compares fourteen machine learning algorithms for predicting daily organic and inorganic waste generation in La Joya de los Sachas, Ecuador.本仓库为《预测城市生活垃圾的机器学习算法对比分析》研究提供了经整理的数据集与完整可复现的工作流程。其中涵盖经清洗后的源数据集——逐日面板数据集、数据预处理、特征工程、模型训练、时序回测、基准模型、不确定性估计相关内容，以及可复现该研究报告中所列表格与图表的脚本/Notebook文件。本研究针对厄瓜多尔拉霍亚德洛斯萨查斯市的每日有机与无机生活垃圾生成量预测任务，对比了十四种机器学习算法的性能表现。