[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-2045":3,"related-2045":64},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":6,"published_at":11,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":18,"tags":20,"primary_topic":9,"secondary_topics":22,"classification":23,"search_phrases":42,"slug":43,"view_count":44,"doi":45,"paper":46,"created_at":63},2045,"Adaptive Ensemble Machine Learning Framework for Robust Prediction, Classification, and Intelligent Decision Support Across Heterogeneous Data Domains","https:\u002F\u002Fdoi.org\u002F10.17148\u002Fijarcce.2026.15911","The increasing diversity and complexity of data across different application domains create significant challenges for conventional machine learning models in achieving robust prediction, classification, and intelligent decision support.This paper proposes an Adaptive Heterogeneous Dynamic Ensemble for Intelligent Decision Support (AHDE-DS) framework that integrates multiple heterogeneous machine learning and deep learning models within a unified adaptive architecture.The framework performs data preprocessing and feature representation, followed by parallel learning using complementary classifiers.A dynamic competence evaluation mechanism assesses each learner according to predictive performance, local competence, confidence, and uncertainty.Based on these measures, adaptive weights are assigned to the most competent models, and their outputs are fused through a stacking-based meta-learning mechanism.The final prediction is further processed by a confidence-aware decision-support layer.Experimental evaluation demonstrates that AHDE-DS achieves 98.73% accuracy, 98.61% precision, 98.54% recall, and 98.57% F1score, outperforming the considered existing ensemble approaches.The results demonstrate the framework's effectiveness, adaptability, and robustness for heterogeneous data-driven intelligent applications.","不同应用领域数据的日益多样化和复杂化，给传统机器学习模型实现稳健的预测、分类和智能决策支持带来了重大挑战。本文提出了一种面向智能决策支持的自适应异构动态集成框架（AHDE-DS），该框架在统一的自适应架构中集成了多种异构机器学习和深度学习模型。该框架首先进行数据预处理和特征表示，随后利用互补分类器进行并行学习。一种动态能力评估机制根据预测性能、局部能力、置信度和不确定性对每个学习器进行评估。基于这些度量，为最具能力的模型分配自适应权重，并通过基于堆叠的元学习机制融合其输出。最终预测结果进一步由置信度感知的决策支持层进行处理。实验评估表明，AHDE-DS达到了98.73%的准确率、98.61%的精确率、98.54%的召回率和98.57%的F1分数，优于所考虑的现有集成方法。结果表明，该框架对于异构数据驱动的智能应用具有有效性、适应性和稳健性。",null,"IJARCCE","2026-09-09T00:00:00Z","论文",10,false,0,{"impact":15,"substance":15,"depth":15,"authority":15,"freshness":15,"relevant":15,"comment":17},"通用机器学习集成框架论文，未涉及农业场景与数据，与三农及农业信息化无直接关联，不建议进入每日精选。",[19],{"name":10,"url":6},[21],"农业人工智能",[],{"status":24,"locked":14,"version":25,"method":26,"evidence":27},"needs_review","2026-10-22.2","rules",{"candidates":28,"agricultural_evidence":14,"reason":41},[29],{"code":30,"score":31,"supported":32,"title":33,"abstract":36,"tags":40},"I2",26,true,[34,35],"machine learning","decision support",[37,38,34,39,35],"深度学习","机器学习","deep learning",[21],"农业相关性需要复核",[],"AdaptiveEnsembleMachineLearningFramework-2045",1,"10.17148\u002Fijarcce.2026.15911",{"doi":45,"openalex_id":47,"authors":48,"venue":10,"cited_by_count":15,"oa_url":6,"card":55,"direction":61,"ingested_from":62},"W7212075043",[49,51,53],{"name":50,"orcid":9},"M. Sasikumar",{"name":52,"orcid":9},"Addepalli Keerthika",{"name":54,"orcid":9},"N Saranya",{"tldr":56,"method":57,"finding":58,"direction":59,"opportunity":60},"提出自适应异构动态集成框架AHDE-DS，实现跨异构数据的鲁棒预测、分类与智能决策支持。","融合多种机器学习与深度学习模型，采用动态能力评估、自适应加权与堆叠元学习。","实验准确率达98.73%，优于现有集成方法，验证了框架的适应性与鲁棒性。","农业人工智能与决策模型","可将该自适应集成框架迁移至农业多源异构数据（气象、土壤、遥感）的病害预警与决策场景。","智慧农业 \u002F 农业物联网","openalex","2026-09-10T23:30:09.623326Z",{"total":65,"page":44,"page_size":65,"items":66},6,[67,123,178,234,286,338],{"id":68,"title":69,"url":70,"summary":71,"summary_zh":9,"content":9,"source_name":72,"source_url":9,"published_at":73,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":74,"score_detail":75,"sources":80,"tags":82,"primary_topic":9,"secondary_topics":87,"classification":88,"search_phrases":110,"slug":113,"view_count":15,"doi":9,"paper":114,"created_at":122},5658,"基于混合卷积Transformer与知识自蒸馏的香蕉品种跨等级分类泛化增强","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.29008","预印本，未同行评审。针对香蕉品种分类中训练等级与实际品质等级的域偏移，用31678张图像、三个品种三个等级数据并采用留一等级交叉验证，比较卷积网络、视觉Transformer与双流架构；提出的混合模型加自蒸馏平均准确率达81.25%，优于两类基线。","arXiv 2026-10-21","2026-10-20T16:00:00Z",56,{"impact":76,"substance":77,"depth":78,"authority":65,"freshness":76,"relevant":44,"comment":79},8,18,16,"预印本提出混合卷积Transformer与知识自蒸馏方法，在香蕉品种跨等级分类上取得81.25%准确率，方法有增量但属细分技术进展，未同行评审，公共影响有限。",[81],{"name":72,"url":70},[83,21,84,85,86],"智慧农业","作物分类","香蕉","领域泛化",[],{"status":24,"locked":14,"version":25,"method":26,"evidence":89},{"candidates":90,"agricultural_evidence":14,"reason":41},[91,96,101,105],{"code":92,"score":44,"supported":14,"title":93,"abstract":94,"tags":95},"B2",[],[],[85],{"code":97,"score":44,"supported":14,"title":98,"abstract":99,"tags":100},"I1",[],[],[83],{"code":30,"score":44,"supported":14,"title":102,"abstract":103,"tags":104},[],[],[21],{"code":106,"score":44,"supported":14,"title":107,"abstract":108,"tags":109},"I3",[],[],[84],[111,112],"香蕉品种 跨等级分类","混合卷积Transformer 知识自蒸馏","香蕉品种跨等级分类-5658",{"doi":9,"openalex_id":9,"authors":115,"venue":9,"cited_by_count":15,"oa_url":9,"card":116,"direction":59,"ingested_from":121},[],{"tldr":117,"method":118,"finding":119,"direction":59,"opportunity":120},"提出混合卷积Transformer加知识自蒸馏，提升香蕉品种跨等级分类泛化能力。","31678张图像、三品种三等级，留一等级交叉验证，对比CNN、ViT与双流架构。","混合模型加自蒸馏平均准确率81.25%，优于卷积网络和视觉Transformer基线。","可探索知识自蒸馏在更多农产品跨域分级中的应用，并研究轻量化部署与真实分拣场景验证。","agent","2026-10-23T00:32:42.976104Z",{"id":124,"title":125,"url":126,"summary":127,"summary_zh":9,"content":9,"source_name":128,"source_url":9,"published_at":129,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":130,"score_detail":131,"sources":134,"tags":136,"primary_topic":141,"secondary_topics":147,"classification":148,"search_phrases":166,"slug":169,"view_count":15,"doi":9,"paper":170,"created_at":177},5656,"农业操作中基于视觉语言动作模型的机器人学习","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.26717","预印本，未同行评审。构建含多样番茄结构的仿真农业环境，用遥操作采集专家演示，先在仿真中适配两个预训练通用策略，再在机械臂真机演示上微调。较新策略任务得分更高且分布偏移下退化更小；真机实验规模有限。","arXiv 2026-10-19","2026-10-18T16:00:00Z",55,{"impact":13,"substance":78,"depth":132,"authority":76,"freshness":65,"relevant":44,"comment":133},15,"预印本方法有新意但真机规模有限，属细分技术进展，暂不宜进每日精选。",[135],{"name":128,"url":126},[21,137,138,139,140],"农业机器人","番茄","视觉语言动作模型","仿真到现实",{"code":142,"slug":143,"name":144,"path":145,"parent_code":146},"B1","vegetables","蔬菜","\u002Ftopic\u002Fhorticulture\u002Fvegetables","B",[],{"status":149,"locked":14,"version":25,"method":26,"evidence":150},"classified",{"candidates":151,"agricultural_evidence":32},[152,157,162],{"code":142,"score":153,"supported":32,"title":154,"abstract":155,"tags":156},4,[],[138],[138],{"code":158,"score":44,"supported":14,"title":159,"abstract":160,"tags":161},"F6",[],[],[137],{"code":30,"score":44,"supported":14,"title":163,"abstract":164,"tags":165},[],[],[21],[167,168],"视觉语言动作模型 农业机器人","番茄 仿真 机械臂","视觉语言动作模型农业机器人-5656",{"doi":9,"openalex_id":9,"authors":171,"venue":9,"cited_by_count":15,"oa_url":9,"card":172,"direction":59,"ingested_from":121},[],{"tldr":173,"method":174,"finding":175,"direction":59,"opportunity":176},"在仿真番茄环境中适配并微调视觉语言动作模型，实现农业操作机器人学习。","构建仿真番茄环境，遥操作采集演示，仿真适配预训练策略后真机微调。","较新策略任务得分更高，分布偏移下退化更小；真机实验规模有限。","可扩大真机实验规模，研究视觉语言动作模型在复杂农田分布偏移下的鲁棒性。","2026-10-23T00:32:42.411245Z",{"id":179,"title":180,"url":181,"summary":182,"summary_zh":9,"content":9,"source_name":183,"source_url":9,"published_at":184,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":185,"score_detail":186,"sources":191,"tags":193,"primary_topic":197,"secondary_topics":198,"classification":199,"search_phrases":222,"slug":225,"view_count":15,"doi":9,"paper":226,"created_at":233},5653,"蔬菜田半自动驾驶精准喷药系统多速度田间性能评价","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.24893","预印本，未同行评审。用五个生长季10599张图、84311个标注实例训练目标检测模型，结合多目标跟踪与12喷头装置在萝卜田开展三档速度田间试验。最低速平均精度达94.6%，提速后杂草命中率由78.3%降至35.0%，未达预设商业门槛。","arXiv 2026-10-17","2026-10-16T16:00:00Z",64,{"impact":187,"substance":188,"depth":189,"authority":76,"freshness":65,"relevant":44,"comment":190},12,21,17,"数据规模扎实、结论明确的预印本田间评价研究，但提速后命中率大幅下降且未达商业门槛，属细分技术进展，可作候选而非必选。",[192],{"name":183,"url":181},[83,21,194,195,196],"杂草识别","田间试验","精准喷药",{"code":142,"slug":143,"name":144,"path":145,"parent_code":146},[],{"status":149,"locked":14,"version":25,"method":26,"evidence":200},{"candidates":201,"agricultural_evidence":32},[202,208,213,218],{"code":142,"score":203,"supported":32,"title":204,"abstract":205,"tags":207},11,[144],[206],"萝卜",[],{"code":30,"score":153,"supported":32,"title":209,"abstract":210,"tags":212},[],[211],"目标检测",[21],{"code":214,"score":44,"supported":14,"title":215,"abstract":216,"tags":217},"F4",[],[],[194],{"code":97,"score":44,"supported":14,"title":219,"abstract":220,"tags":221},[],[],[83],[223,224],"萝卜田 精准喷药 田间试验","半自动驾驶 喷药 杂草检测","萝卜田精准喷药田间试验-5653",{"doi":9,"openalex_id":9,"authors":227,"venue":9,"cited_by_count":15,"oa_url":9,"card":228,"direction":61,"ingested_from":121},[],{"tldr":229,"method":230,"finding":231,"direction":61,"opportunity":232},"在萝卜田测试半自动精准喷药系统，评估三档速度下的检测与喷施性能。","五个生长季10599张图训练目标检测，结合多目标跟踪与12喷头装置开展田间试验。","最低速精度94.6%，提速后杂草命中率由78.3%降至35.0%，未达商业门槛。","高速下跟踪与喷施协同失效是瓶颈，可研究速度自适应跟踪与喷头时序补偿算法。","2026-10-23T00:32:41.788532Z",{"id":235,"title":236,"url":237,"summary":238,"summary_zh":9,"content":9,"source_name":128,"source_url":9,"published_at":129,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":239,"score_detail":240,"sources":245,"tags":247,"primary_topic":251,"secondary_topics":256,"classification":257,"search_phrases":274,"slug":277,"view_count":15,"doi":9,"paper":278,"created_at":285},5652,"Bombadil：面向植物电生理信号的基础模型","https:\u002F\u002Farxiv.org\u002Fabs\u002F2610.26597","预印本，未同行评审。以100个植物物种、十类环境（实验室、温室、商业果园、天然林）超400万小时植物电信号无标签预训练基础模型，可从同一信号追踪七项胁迫生理指标变化，并覆盖训练中无标签的物种与胁迫类型；跨场景田间验证仍有限。",74,{"impact":77,"substance":241,"depth":242,"authority":243,"freshness":65,"relevant":44,"comment":244},22,19,9,"预印本尚未经同行评审，但以400万小时、100个物种的植物电信号预训练基础模型，在方法新颖性与数据规模上具有实质增量，值得关注但田间验证有限。",[246],{"name":128,"url":237},[83,21,248,249,250],"基础模型","植物电生理","胁迫监测",{"code":252,"slug":253,"name":254,"path":255,"parent_code":146},"B7","protected-horticulture","设施园艺与植物工厂","\u002Ftopic\u002Fhorticulture\u002Fprotected-horticulture",[],{"status":149,"locked":14,"version":25,"method":26,"evidence":258},{"candidates":259,"agricultural_evidence":32},[260,266,270],{"code":252,"score":261,"supported":32,"title":262,"abstract":263,"tags":265},3,[],[264],"温室",[],{"code":97,"score":44,"supported":14,"title":267,"abstract":268,"tags":269},[],[],[83],{"code":30,"score":44,"supported":14,"title":271,"abstract":272,"tags":273},[],[],[21],[275,276],"Bombadil 植物电信号 基础模型","植物电生理 胁迫监测","Bombadil植物电信号基础模型-5652",{"doi":9,"openalex_id":9,"authors":279,"venue":9,"cited_by_count":15,"oa_url":9,"card":280,"direction":59,"ingested_from":121},[],{"tldr":281,"method":282,"finding":283,"direction":59,"opportunity":284},"构建植物电生理信号基础模型，用超400万小时无标签数据预训练。","100个物种、十类环境超400万小时植物电信号无标签预训练。","同一信号可追踪七项胁迫生理指标，并泛化到未见物种与胁迫类型。","跨场景田间验证不足，可研究多环境迁移与胁迫早期预警的落地评估。","2026-10-23T00:32:41.455307Z",{"id":287,"title":288,"url":289,"summary":290,"summary_zh":9,"content":9,"source_name":291,"source_url":9,"published_at":292,"category":293,"cover_url":9,"hotness":13,"is_selected":32,"score":294,"score_detail":295,"sources":299,"tags":301,"primary_topic":305,"secondary_topics":311,"classification":312,"search_phrases":333,"slug":336,"view_count":15,"doi":9,"paper":9,"created_at":337},5628,"2026智慧农业博览会10月23日南宁启幕，200余家单位参展对接东盟","https:\u002F\u002Fnynct.gxzf.gov.cn\u002Fxwdt\u002Fywkb\u002Ft28253108.shtml","由农业农村部对外经济合作中心主办的2026智慧农业博览会10月23日至25日在南宁举办，展览总面积10700平方米，汇聚200余家国内外参展单位、专业观众超2万人次，设智能装备、人工智能加农业服务、数字乡村三大展区，同期举办中国—东盟数字农业论坛。","广西壮族自治区农业农村厅 2026-10-22","2026-10-21T16:00:00Z","报道",85,{"impact":31,"substance":296,"depth":132,"authority":297,"freshness":13,"relevant":44,"comment":298},20,14,"农业农村部主办的国家级智慧农业展会，规模与东盟对接定位明确，时效性强，值得进入每日精选。",[300],{"name":291,"url":289},[302,83,21,303,304],"数字乡村","智能装备","中国—东盟",{"code":306,"slug":307,"name":308,"path":309,"parent_code":310},"I6","digital-village","数字乡村与数据要素","\u002Ftopic\u002Fdigital-agriculture\u002Fdigital-village","I",[],{"status":149,"locked":14,"version":25,"method":26,"evidence":313},{"candidates":314,"agricultural_evidence":32},[315,320,324,329],{"code":97,"score":132,"supported":32,"title":316,"abstract":317,"tags":319},[83],[83,318],"数字农业",[83],{"code":306,"score":153,"supported":32,"title":321,"abstract":322,"tags":323},[],[302],[302],{"code":325,"score":44,"supported":14,"title":326,"abstract":327,"tags":328},"F5",[],[],[303],{"code":30,"score":44,"supported":14,"title":330,"abstract":331,"tags":332},[],[],[21],[334,335],"智慧农业博览会 南宁","中国—东盟 数字农业论坛","智慧农业博览会南宁-5628","2026-10-23T00:31:53.301869Z",{"id":339,"title":340,"url":341,"summary":342,"summary_zh":343,"content":9,"source_name":344,"source_url":341,"published_at":345,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":346,"score_detail":347,"sources":350,"tags":352,"primary_topic":356,"secondary_topics":360,"classification":366,"search_phrases":383,"slug":386,"view_count":15,"doi":387,"paper":388,"created_at":405},5613,"ExplainableCropAI: Interpretable Deep Learning for Agricultural Disease Diagnosis and Decision Support","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs44196-026-01654-9","Deep learning models classify crop diseases accurately, but their black-box nature limits practical use: farmers and extension officers need to understand a diagnosis before acting on it. We present ExplainableCropAI, a decision-support system for paddy disease diagnosis that attaches three complementary explanation channels to a ConvNeXt V2-Tiny classifier: Grad-CAM++ visual attribution, Monte Carlo Dropout uncertainty with temperature-scaled calibration, and natural-language explanations generated by Phi-3-mini under a grounding constraint that restricts the model to typed fields drawn from the prediction, the attribution map and a curated agronomic knowledge base. None of the components is new; the contribution lies in their integration into a single pipeline and in measuring the properties that integration is meant to deliver. On a leakage-controlled, held-out test split of 1,540 images from the Paddy Doctor dataset (10,407 images, nine diseases and healthy), the classifier reaches 97.5% accuracy (95% CI 96.7–98.3) and macro-F1 of 96.9%. Six attribution methods are compared under identical settings with model-based faithfulness metrics; the calibrated confidence separates correct from incorrect predictions with an AUROC of 0.93, and referring the 10% least-confident cases to an expert removes 74% of all errors. Under eight image corruptions at five severities the model retains 89.7% mean accuracy against 72–78% for six baselines trained identically, and its referral rate rises with the degree of degradation. Zero-shot evaluation on two independently collected rice-disease datasets from Bangladesh and India shows that accuracy roughly halves under cross-dataset shift and that confidence no longer identifies errors, while fifty local images per class recover accuracy above 90%. The code, split files, trained weights and evaluation scripts are released.","深度学习模型能够准确分类作物病害，但其黑箱性质限制了实际应用：农民和推广人员需要在理解诊断结果后才能采取行动。我们提出ExplainableCropAI，一个用于水稻病害诊断的决策支持系统，它在ConvNeXt V2-Tiny分类器上附加了三条互补的解释通道：Grad-CAM++视觉归因、经温度缩放校准的蒙特卡洛Dropout不确定性，以及由Phi-3-mini在接地约束下生成的自然语言解释——该约束将模型限制为从预测结果、归因图和精选农学知识库中提取的类型化字段。各组件均非新创；贡献在于将其集成到单一流水线中，并测量该集成旨在提供的特性。在来自Paddy Doctor数据集（10,407张图像，九种病害及健康）的1,540张图像上，经泄漏控制的留出测试划分显示，分类器达到97.5%的准确率（95% CI 96.7–98.3）和96.9%的宏F1值。在相同设置下，使用基于模型的忠实性指标比较了六种归因方法；校准后的置信度以0.93的AUROC区分正确与错误预测，将置信度最低的10%案例转交专家可消除全部错误的74%。在五种严重程度的八种图像损坏下，模型保持89.7%的平均准确率，而相同训练的六个基线为72–78%，且其转交率随退化程度上升。在来自孟加拉国和印度的两个独立采集的水稻病害数据集上进行零样本评估显示，跨数据集偏移下准确率大约减半，置信度不再能识别错误，而每类五十张本地图像可将准确率恢复至90%以上。代码、划分文件、训练权重和评估脚本均已发布。","International Journal of Computational Intelligence Systems","2026-10-22T00:00:00Z",84,{"impact":77,"substance":348,"depth":242,"authority":297,"freshness":13,"relevant":44,"comment":349},23,"可解释AI与作物病害诊断的集成系统，方法组合有工程价值，数据规模与评测较扎实，但组件本身非新，属细分领域重要进展而非重大突破。",[351],{"name":344,"url":341},[83,21,353,354,355],"水稻","可解释AI","作物病害诊断",{"code":30,"slug":357,"name":358,"path":359,"parent_code":310},"agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai",[361],{"code":362,"slug":363,"name":353,"path":364,"parent_code":365},"A1","rice","\u002Ftopic\u002Ffield-crops\u002Frice","A",{"status":149,"locked":14,"version":25,"method":26,"evidence":367},{"candidates":368,"agricultural_evidence":32},[369,374,379],{"code":30,"score":370,"supported":32,"title":371,"abstract":372,"tags":373},24,[39,35],[37,39],[21,354],{"code":362,"score":13,"supported":32,"title":375,"abstract":376,"tags":378},[],[353,363,377],"paddy",[353],{"code":97,"score":44,"supported":14,"title":380,"abstract":381,"tags":382},[],[],[83],[384,385],"ExplainableCropAI 水稻病害","Paddy Doctor 数据集","ExplainableCropAI水稻病害-5613","10.1007\u002Fs44196-026-01654-9",{"doi":387,"openalex_id":389,"authors":390,"venue":344,"cited_by_count":15,"oa_url":341,"card":400,"direction":59,"ingested_from":62},"W7166554515",[391,393,394,397,399],{"name":392,"orcid":9},"Santosh Kumar",{"name":392,"orcid":9},{"name":395,"orcid":396},"Durgesh Nandan","https:\u002F\u002Forcid.org\u002F0000-0002-9762-3559",{"name":398,"orcid":9},"Sumit Kumar",{"name":398,"orcid":9},{"tldr":401,"method":402,"finding":403,"direction":59,"opportunity":404},"构建可解释水稻病害诊断系统，融合归因、不确定性与自然语言解释。","ConvNeXt V2-Tiny分类器，Grad-CAM++、MC Dropou","测试准确率97.5%，低置信转诊消除74%错误，跨数据集精度减半。","跨数据集置信失效，少量本地样本微调可恢复精度，值得研究田间自适应与主动学习。","2026-10-22T23:31:14.582261Z"]