[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2219":3},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":8,"source_name":9,"source_url":8,"published_at":10,"category":11,"cover_url":8,"hotness":12,"is_selected":13,"score":14,"score_detail":15,"sources":22,"tags":24,"view_count":30,"doi":8,"paper":31,"created_at":40},2219,"《面向部署的中国南方水稻产量预测阶段特异性信息评估:跨年份、地点及地点-年份环境》","https:\u002F\u002Fwww.ebiotrade.com\u002Fnewsf\u002F2026-9\u002F20260909000242394.htm","研究人员利用中国南方10个地点、7年、67个地点-年份环境下168个品种的3204条产量记录,开发场景感知的机器学习框架。研究在随机五折交叉验证、留一年验证、前向年份预测、留一地点验证和留一地点-年份验证中比较五组建模特征集。随机五折交叉验证产生最高表观精度,最优田间调查增强模型达到R²=0.844,RMSE为340.59 kg\u002Fha;面向部署的性能较低,在前向年份预测中宏平均R²为0.359,留一地点验证为0.290,留一地点-年份验证为0.131。",null,"Frontiers in Plant Science \u002F 生物通 2026年9月9日","2026-09-09T00:00:00Z","论文",10,false,79,{"impact":16,"substance":17,"depth":16,"authority":18,"freshness":19,"relevant":20,"comment":21},18,22,13,8,1,"基于中国南方10地7年3204条产量记录的场景感知机器学习框架，系统揭示交叉验证精度与真实部署性能的巨大落差，对农业AI落地评估有实质参考价值。",[23],{"name":9,"url":6},[25,26,27,28,29],"智慧农业","农业人工智能","水稻","产量预测","模型部署",0,{"doi":8,"openalex_id":8,"authors":32,"venue":8,"cited_by_count":30,"oa_url":8,"card":33,"direction":37,"ingested_from":39},[],{"tldr":34,"method":35,"finding":36,"direction":37,"opportunity":38},"基于中国南方多环境水稻数据，评估机器学习产量预测在不同部署场景下的真实精度。","用10地点7年3204条记录，比较五组特征集在多种交叉验证下的表现。","随机交叉验证R²达0.844，但前向年份、留一地点等部署场景R²仅0.13-0.36。","农业人工智能与决策模型","需开发跨年份地点鲁棒建模与迁移学习，缩小随机验证与真实部署间的精度差距。","agent","2026-09-12T00:06:40.813584Z"]