[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-1888":3,"related-1888":139},{"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":23,"tags":25,"primary_topic":31,"secondary_topics":37,"classification":50,"search_phrases":120,"slug":123,"view_count":20,"doi":124,"paper":125,"created_at":138},1888,"Smart Technologies for Precision Nutrient Management in Smallholder Farming: A Comprehensive Review","https:\u002F\u002Fdoi.org\u002F10.4038\u002Fjur.v14i1.8134","Smallholder farmers with less than 2 hectares contribute more than 1\u002F3rd of the world food supply but continue to suffer the challenges of low nutrient use efficiency (30-40%), reduced soil fertility and lack of access to precision agriculture technologies. The aim of the review is to synthesize the evidence of 235 peer-reviewed articles (2010-2025) to assess the precision nutrient management (PNM) technologies, namely, in terms of their applicability, economic feasibility, and adoption capacity within the smallholder farming systems. Systematic literature review according to PRISMA, the analysis of articles on satellite and UAV-based remote sensing, IoT-enabled soil sensors, variable rate application systems, machine learning algorithms, and case studies worldwide were summarized. The application of variable rates can cut fertilizer use by 10-25 % and boost yields by 7-15 % but only small holders adopt it globally because of capital limitations, fragmented land holdings (1-2 ha), poor rural connectivity and insufficient extension services. Multi-source data forecasts nutrient needs with 85-92 % accuracy using machine learning algorithms (Random Forests, Support Vector Machines. Open-source sensor development, offline mobile decision support systems, equipment sharing cooperatives, blended finance (30–50% tech subsidies), and farmer involvement must all be part of the democratization of PNM for smallholders. Institutional adjustments, unified extension services, public-private cooperation, and supportive nutrient management regulations are all necessary for effective scaling. PNM adoption produces 15-30% nutrient runoff reduction, 10-20% greenhouse gas emission reductions, and 0.4-1.2 tons CO2-equivalent per hectare of soil carbon sequestration. PNM adoption produces 15-30% nutrient runoff reduction, 10-20% greenhouse gas emission reductions, and 0.4-1.2 tons CO2-equivalent per hectare of soil carbon sequestration. Economic analyses indicated that farms over 10 ha have positive returns, though smallholders required special support to overcome initial costs and limitations.","拥有不足2公顷土地的小农户贡献了全球粮食供应总量的三分之一以上，却持续面临养分利用效率低下（30-40%）、土壤肥力下降以及缺乏精准农业技术获取途径等挑战。本综述旨在综合235篇同行评审论文（2010-2025年）的证据，评估精准养分管理（PNM）技术在小农农业系统中的适用性、经济可行性和采纳能力。依据PRISMA方法进行系统性文献综述，总结了基于卫星和无人机遥感、物联网土壤传感器、变量施肥系统、机器学习算法及全球案例研究的分析结果。变量施肥技术的应用可减少化肥使用量10-25%，提高产量7-15%，但由于资金限制、土地碎片化（1-2公顷）、农村网络覆盖不足和推广服务欠缺，全球仅有少数小农户采纳该技术。利用机器学习算法（随机森林、支持向量机）进行多源数据预测，养分需求预测准确率可达85-92%。实现小农户PNM技术的普及化，必须包括开源传感器开发、离线移动决策支持系统、设备共享合作社、混合融资（30-50%技术补贴）以及农户参与等多方面措施。有效的规模化推广还需要制度调整、统一的推广服务、公私合作以及支持性养分管理法规的配套。PNM技术的采纳可减少养分径流15-30%，降低温室气体排放10-20%，并实现每公顷0.4-1.2吨二氧化碳当量的土壤碳封存。经济分析表明，10公顷以上的农场可获得正收益，但小农户需要特殊支持以克服初始成本和限制因素。",null,"Journal of the University of Ruhuna","2026-09-07T00:00:00Z","论文",10,false,72,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,12,2,1,"系统综述小农户精准养分管理技术，数据详实，但时效性低，适合专题参考。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","精准施肥","农业遥感","机器学习","小农户",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"G2","farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[38,44],{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},"F3","fertilizers","肥料与水肥一体化","\u002Ftopic\u002Finputs-equipment\u002Ffertilizers","F",{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",{"status":51,"locked":14,"version":52,"method":53,"evidence":54},"classified","2026-10-22.2","rules",{"candidates":55,"agricultural_evidence":58},[56,63,70,76,84,91,99,106,113],{"code":32,"score":57,"supported":58,"title":59,"abstract":61,"tags":62},15,true,[60],"smallholder",[30,60],[30],{"code":39,"score":13,"supported":58,"title":64,"abstract":65,"tags":69},[],[66,67,68],"施肥","养分管理","fertilizer",[27],{"code":45,"score":13,"supported":58,"title":71,"abstract":72,"tags":75},[],[29,73,74],"machine learning","decision support",[29],{"code":77,"score":78,"supported":58,"title":79,"abstract":80,"tags":83},"I1",7,[],[81,82],"精准农业","precision agriculture",[26],{"code":85,"score":78,"supported":58,"title":86,"abstract":87,"tags":90},"I3",[],[88,89],"遥感","remote sensing",[28],{"code":92,"score":93,"supported":58,"title":94,"abstract":95,"tags":98},"B7",6,[],[96,97],"温室","greenhouse",[],{"code":100,"score":93,"supported":58,"title":101,"abstract":102,"tags":105},"H2",[],[103,104],"土壤","soil",[],{"code":107,"score":93,"supported":58,"title":108,"abstract":109,"tags":112},"I5",[],[110,111],"物联网","IoT",[],{"code":114,"score":115,"supported":58,"title":116,"abstract":117,"tags":119},"H4",3,[],[118],"carbon sequestration",[],[121,122],"农业遥感 智慧农业 机器学习 精准施肥","农业遥感 智慧农业","农业遥感智慧农业机器学习精准施肥-1888","10.4038\u002Fjur.v14i1.8134",{"doi":124,"openalex_id":126,"authors":127,"venue":10,"cited_by_count":130,"oa_url":6,"card":131,"direction":135,"ingested_from":137},"W7211915396",[128],{"name":129,"orcid":9},"Anuga Liyanage",0,{"tldr":132,"method":133,"finding":134,"direction":135,"opportunity":136},"综述小农户精准养分管理技术，分析其适用性、经济可行性与采纳障碍。","PRISMA系统综述235篇文献，分析遥感、物联网、变量施肥及机器学习案例。","变量施肥可减肥10-25%并增产7-15%，但小农户因资金、土地细碎等采纳率低。","智慧农业 \u002F 农业物联网","小农户精准养分管理技术民主化路径，如低成本传感器、离线决策支持系统及合作社共享模式。","openalex","2026-09-08T23:30:08.156887Z",{"total":93,"page":21,"page_size":93,"items":140},[141,200,252,324,393,474],{"id":142,"title":143,"url":144,"summary":145,"summary_zh":146,"content":9,"source_name":147,"source_url":144,"published_at":148,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":149,"score_detail":150,"sources":154,"tags":156,"primary_topic":160,"secondary_topics":161,"classification":163,"search_phrases":179,"slug":182,"view_count":130,"doi":183,"paper":184,"created_at":199},5592,"Evaluation of Concept-Rank Collapse Audits for Computer Vision Agriculture Disease Classifiers Under Low-Label Regimes: An Interpretability Capacity Bound for Smallholder Fungicide Selectivity","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545193","Concept-based interpretability methods for plant-disease classifiers are typically evaluated by whether their concept activations correlate with human-annotated labels, but this criterion is silent on a prior question: whether the classifier's internal representation retains enough concept-rank structure for any audit to be informative. the present analysis formalises that prior question as an interpretability capacity bound. A classifier is modelled as a concept-rank operator whose singular values govern how much concept-level signal survives the projection from pixel space to class logits, and a collapse audit is defined as the ratio of retained concept-rank energy to total concept energy. Under a low-label regime, where concept annotations are scarce and the audit must be estimated from few labelled examples, the bound predicts that audit reliability degrades as a function of the concept-rank spectrum and the label budget. The analysis is a theoretical and methodological contribution: it derives the bound, states the conditions under which a collapse audit is identifiable, and characterises the operating characteristics of the audit as a decision procedure under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64). Power was estimated across 2000 design cells for cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points; 52 clusters were required for 5 percentage points at 80% power. The bound reframes concept audits as capacity tests rather than fidelity tests, with consequences for how smallholder fungicide-selectivity decisions should be supported by interpretable disease classifiers.","面向植物病害分类器的基于概念的可解释性方法，通常以其概念激活是否与人工标注标签相关作为评价标准，但该标准对一个更先前的问题保持沉默：分类器的内部表征是否保留了足够的概念秩结构，从而使任何审计具有信息量。本分析将该先前问题形式化为一个可解释性容量上界。分类器被建模为概念秩算子，其奇异值支配着从像素空间到类别logits的投影中概念级信号得以保留的程度，并将坍缩审计定义为保留的概念秩能量与总概念能量之比。在低标签情形下，概念标注稀缺且审计必须由少量标注样本估计，该上界预测审计可靠性会随概念秩谱和标签预算的函数而退化。本分析是一项理论和方法论贡献：推导出该上界，陈述坍缩审计可识别的条件，并在声明的方差结构（簇间SD = 6.0，残差SD = 4.5，组内相关系数 = 0.64）下刻画审计作为决策程序的操作特征。在2000个设计单元中，针对簇数为8、16、24和32以及效应量为3、5和8个百分点的情形估计了功效；在80%功效下达到5个百分点需要52个簇。该上界将概念审计重新框定为容量检验而非保真度检验，并对可解释病害分类器应如何支持小农杀菌剂选择性决策具有启示意义。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",69,{"impact":151,"substance":152,"depth":17,"authority":19,"freshness":13,"relevant":21,"comment":153},8,21,"理论方法论文，提出可解释性容量上界并给出审计功效分析，对智慧农业病害识别可信度有参考价值，但属小众方法学研究、缺乏田间实证，暂不宜进入每日精选。",[155],{"name":147,"url":144},[26,157,30,158,159],"农业人工智能","植物病害识别","可解释性",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[162],{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},{"status":51,"locked":14,"version":52,"method":53,"evidence":164},{"candidates":165,"agricultural_evidence":58},[166,170,175],{"code":32,"score":19,"supported":58,"title":167,"abstract":168,"tags":169},[60],[60],[30],{"code":45,"score":13,"supported":58,"title":171,"abstract":173,"tags":174},[172],"computer vision",[],[157,158],{"code":77,"score":21,"supported":14,"title":176,"abstract":177,"tags":178},[],[],[26],[180,181],"农业人工智能 植物病害识别 可解释性 智慧农业","农业人工智能 植物病害识别","农业人工智能植物病害识别可解释性智慧农业-5592","10.5281\u002Fzenodo.23545193",{"doi":183,"openalex_id":185,"authors":186,"venue":147,"cited_by_count":130,"oa_url":144,"card":193,"direction":197,"ingested_from":137},"W7223137323",[187,189,191],{"name":188,"orcid":9},"Akshay Ramgoolam",{"name":190,"orcid":9},"Priya Bundhoo Appadoo",{"name":192,"orcid":9},"Jean-Noël LEspérance",{"tldr":194,"method":195,"finding":196,"direction":197,"opportunity":198},"提出可解释性容量上界，评估低标签下概念秩崩溃审计的可靠性。","将分类器建模为概念秩算子，推导容量上界并做2000格功效模拟。","审计可靠性随概念秩谱与标签预算下降；5个百分点效应需52个聚类达80%功效。","农业人工智能与决策模型","可将容量上界用于小农户杀菌剂选择性决策的可解释模型审计与标签预算设计。","2026-10-22T11:21:10.425290Z",{"id":201,"title":202,"url":203,"summary":204,"summary_zh":205,"content":9,"source_name":147,"source_url":203,"published_at":148,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":206,"score_detail":207,"sources":209,"tags":211,"primary_topic":214,"secondary_topics":215,"classification":217,"search_phrases":232,"slug":235,"view_count":130,"doi":236,"paper":237,"created_at":251},5590,"Evaluation of Concept-Probe Temporal-Stability Audits for Computer Vision Agriculture Disease Classifiers Under Seasonal Phenology Drift: An Interpretability Decay Bound for Smallholder Fungicide Timing","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23545234","Concept-probe audits are increasingly used to certify that a crop-disease classifier's internal representations track the pathological concepts a smallholder would act on, but a probe certified at one point in the season may silently decouple from those concepts as phenology shifts the visual statistics of the field. the present analysis treats that decoupling as a formal object. It defines a temporal-stability audit as a sequence of concept-probe accuracies measured across defined phenological windows, and it derives an interpretability decay bound: the rate at which probe accuracy can fall before the fungicide-timing decision it is meant to support becomes unreliable. The bound is expressed as a function of the probe's initial margin, the drift in concept prevalence between windows, and the classifier's sensitivity to the drifted features, and it yields a certified audit interval rather than a single-point certificate. The paper then characterises the operating characteristics of the audit design under a declared variance structure, using a replication-based evaluation of the before-after-control-impact estimator with between-cluster SD = 6.0, residual SD = 4.5 and intraclass correlation = 0.64. Power was estimated across 2000 design cells over cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points; 54 clusters were required for 5 percentage points at 80% power. The contribution is a decision-theoretic account of when a concept-probe certificate expires, and a design rule for scheduling re-audits across the season.","概念探针审计（concept-probe audit）日益被用于验证作物病害分类器的内部表征是否追踪小农户会据以采取行动的病理学概念，但在生长季某一时点获得认证的探针，可能随着物候变化改变田间的视觉统计特征而悄然与这些概念脱钩。本分析将这种脱钩视为一个形式化对象。它将时间稳定性审计（temporal-stability audit）定义为在既定物候窗口内测量的一系列概念探针准确率，并推导出一个可解释性衰减界限（interpretability decay bound）：即探针准确率在其所支持的杀菌剂施用时机决策变得不可靠之前所能下降的速率。该界限表示为探针初始裕度、窗口间概念流行率漂移以及分类器对漂移特征敏感度的函数，并给出一个经认证的审计区间而非单点认证。随后，本文在声明的方差结构下刻画该审计设计的运行特征，采用基于重复的前后对照影响估计量评估，其中组间标准差 = 6.0、残差标准差 = 4.5、组内相关系数 = 0.64。功效在 2000 个设计单元上估计，聚类数取 8、16、24 和 32，效应量取 3、5 和 8 个百分点；在 80% 功效下达到 5 个百分点需要 54 个聚类。本文的贡献在于从决策论角度阐明概念探针认证何时失效，并给出在整个生长季内安排重新审计的设计规则。",70,{"impact":151,"substance":18,"depth":17,"authority":19,"freshness":13,"relevant":21,"comment":208},"该论文提出概念探针审计的时序稳定性框架与可解释性衰减界，方法新颖、统计设计扎实，但属小众方法论研究，产业影响有限，适合作为专业精选而非头条。",[210],{"name":147,"url":203},[26,157,30,212,213],"作物病害识别","模型审计",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[216],{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},{"status":51,"locked":14,"version":52,"method":53,"evidence":218},{"candidates":219,"agricultural_evidence":58},[220,224,228],{"code":32,"score":57,"supported":58,"title":221,"abstract":222,"tags":223},[60],[30,60],[30],{"code":45,"score":13,"supported":58,"title":225,"abstract":226,"tags":227},[172],[],[212,157],{"code":77,"score":21,"supported":14,"title":229,"abstract":230,"tags":231},[],[],[26],[233,234],"概念探针 时序稳定性 作物病害分类器","杀菌剂 施药时机 可解释性衰减","概念探针时序稳定性作物病害分类器-5590","10.5281\u002Fzenodo.23545234",{"doi":236,"openalex_id":238,"authors":239,"venue":147,"cited_by_count":130,"oa_url":203,"card":246,"direction":197,"ingested_from":137},"W7223088030",[240,242,244],{"name":241,"orcid":9},"Abdi-Basid Idriss Farah",{"name":243,"orcid":9},"Amina Abdillahi Guedi",{"name":245,"orcid":9},"Moussa Omar Youssouf",{"tldr":247,"method":248,"finding":249,"direction":197,"opportunity":250},"提出概念探针时间稳定性审计与可解释性衰减界，判断何时需重新审计。","定义跨物候窗口探针精度序列，推导衰减界，用BACI复制评估统计功效。","5个百分点效应需54个聚类达80%功效，审计证书会随物候漂移过期。","可将衰减界与再审计调度规则嵌入小农户杀菌剂时机决策支持系统，并验证跨作物泛化。","2026-10-22T11:21:08.978348Z",{"id":253,"title":254,"url":255,"summary":256,"summary_zh":257,"content":9,"source_name":258,"source_url":255,"published_at":259,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":260,"score_detail":261,"sources":264,"tags":266,"primary_topic":269,"secondary_topics":270,"classification":276,"search_phrases":303,"slug":306,"view_count":130,"doi":9,"paper":307,"created_at":323},4670,"SMART FERTILIZER DISTRIBUTION SYSTEM: A COMPREHENSIVE REVIEW","https:\u002F\u002Fijmert.com\u002Findex.php\u002Fijmert\u002Farticle\u002Fview\u002F334","Agriculture is an important sector for food production and economic development, and the efficient use of fertilizers is essential for achieving better crop growth and productivity. Conventional fertilizer distribution methods largely depend on manual application, which can lead to uneven distribution, excessive fertilizer usage, increased labor requirements, and wastage of agricultural resources. The development of automation and Internet of Things (IoT) technologies has provided new opportunities for improving fertilizer distribution and management in agricultural fields. This review paper presents a comprehensive study of Smart Fertilizer Distribution Systems, focusing on the technologies, control methods, communication techniques, and approaches reported in existing research. Various systems using microcontrollers, IoT devices, wireless communication, web-based platforms, sensors, motor or pump control, driver circuits, and solenoid valves are reviewed. The working principles of automated fertilizer distribution systems and their role in controlling fertilizer flow and reducing manual intervention are discussed. Different communication technologies, including Wi-Fi and other wireless methods, are compared based on their applicability, cost, operating range, reliability, and control capability. The review also examines the major advantages and limitations of existing systems, such as automation, remote operation, improved fertilizer management, reduced labor, network dependency, power requirements, sensor accuracy, and system complexity. A comparative analysis of selected research works is presented to identify the techniques used, their contributions, and the limitations that remain to be addressed. The review further identifies research gaps and discusses future developments such as automatic fertilizer scheduling, multi-zone fertilizer distribution, real-time monitoring, cloud-based data storage, artificial intelligence, machine learning, and integration with renewable energy sources. The study concludes that smart fertilizer distribution can improve the efficiency, accuracy, and sustainability of fertilizer application and has significant potential for future precision agriculture applications.","农业是粮食生产和经济发展的重要部门，而肥料的高效利用对于实现作物更好生长和生产力提升至关重要。传统的肥料分配方法在很大程度上依赖人工施用，这可能导致分配不均、肥料过量使用、劳动力需求增加以及农业资源浪费。自动化和物联网（IoT）技术的发展为改善农田肥料分配与管理提供了新的机遇。本文综述对智能肥料分配系统进行了全面研究，重点关注现有研究中报道的技术、控制方法、通信技术和方法。本文回顾了使用微控制器、物联网设备、无线通信、基于网络的平台、传感器、电机或泵控制、驱动电路和电磁阀的各种系统。讨论了自动化肥料分配系统的工作原理及其在控制肥料流量和减少人工干预方面的作用。基于适用性、成本、工作范围、可靠性和控制能力，比较了包括Wi-Fi和其他无线方法在内的不同通信技术。本综述还考察了现有系统的主要优势和局限性，如自动化、远程操作、改善肥料管理、减少劳动力、网络依赖、功率需求、传感器精度和系统复杂性。本文对选定的研究工作进行了比较分析，以识别所采用的技术、其贡献以及仍有待解决的局限性。本综述进一步指出了研究空白，并讨论了未来发展方向，如自动肥料调度、多区域肥料分配、实时监测、基于云的数据存储、人工智能、机器学习以及与可再生能源的集成。该研究得出结论，智能肥料分配可以提高肥料施用的效率、准确性和可持续性，并在未来精准农业应用中具有巨大潜力。","International Journal of Mechanical Engineering Research and Technology","2026-10-08T00:00:00Z",61,{"impact":19,"substance":262,"depth":57,"authority":19,"freshness":93,"relevant":21,"comment":263},16,"综述性论文系统梳理智能施肥系统的技术路线与研究空白，对智慧农业领域有参考价值，但属文献综述而非原创突破，时效性一般。",[265],{"name":258,"url":255},[26,157,27,267,268],"农业物联网","智能农机",{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},[271,272],{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},{"code":107,"slug":273,"name":274,"path":275,"parent_code":49},"agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot",{"status":51,"locked":14,"version":52,"method":53,"evidence":277},{"candidates":278,"agricultural_evidence":58},[279,284,289,294,298],{"code":39,"score":57,"supported":58,"title":280,"abstract":281,"tags":283},[68],[282,68],"肥料",[27],{"code":45,"score":13,"supported":58,"title":285,"abstract":286,"tags":288},[],[29,73,287],"artificial intelligence",[157],{"code":107,"score":13,"supported":58,"title":290,"abstract":291,"tags":293},[],[110,292,111],"internet of things",[267],{"code":77,"score":78,"supported":58,"title":295,"abstract":296,"tags":297},[],[81,82],[26],{"code":299,"score":21,"supported":14,"title":300,"abstract":301,"tags":302},"F5",[],[],[268],[304,305],"智能施肥系统 物联网","精准施肥 自动化","智能施肥系统物联网-4670",{"doi":9,"openalex_id":308,"authors":309,"venue":258,"cited_by_count":130,"oa_url":255,"card":318,"direction":135,"ingested_from":137},"W7221394563",[310,312,314,316],{"name":311,"orcid":9},"Anjali Manoj Kumbhar",{"name":313,"orcid":9},"Shruti Prashant Kulkarni",{"name":315,"orcid":9},"Snehal Santosh Khune",{"name":317,"orcid":9},"B. B. Godbole",{"tldr":319,"method":320,"finding":321,"direction":135,"opportunity":322},"综述智能施肥系统的技术、控制与通信方法，并指出研究空白与未来方向。","综述基于微控制器、物联网、无线通信、传感器与电磁阀的自动施肥系统研究。","智能施肥可提升施肥效率、精度与可持续性，但存在网络依赖、功耗与传感器精度等局限。","可探索AI驱动的自动施肥调度、多区域分配与云边协同的实时精准施肥系统。","2026-10-11T23:30:16.128775Z",{"id":325,"title":326,"url":327,"summary":328,"summary_zh":329,"content":9,"source_name":330,"source_url":327,"published_at":331,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":332,"score_detail":333,"sources":337,"tags":339,"primary_topic":342,"secondary_topics":343,"classification":345,"search_phrases":373,"slug":376,"view_count":93,"doi":377,"paper":378,"created_at":392},3512,"AI-Driven Agricultural Advisory and Diagnostic Systems for Smallholder Farming: Technical Architectures, Evidence and Deployment Priorities for North-East India","https:\u002F\u002Fdoi.org\u002F10.9734\u002Farja\u002F2026\u002Fv19i4919","Artificial intelligence (AI) is being introduced into agricultural advisory services through machine learning, computer vision, conversational large language models, retrieval-augmented generation and multimodal interfaces. For smallholder farming, the central question is not whether these technologies can produce technically plausible outputs, but whether they can provide locally correct, actionable and safe recommendations under heterogeneous agronomic, linguistic and connectivity conditions. This critical narrative review integrates evidence on digital extension, AI-enabled agricultural advice, image-based diagnosis and responsible digital agriculture, with particular reference to North-East India. Literature published from 1 January 2010 to 17 July 2026 was considered, with emphasis on peer-reviewed field evaluations, technical validation studies, reviews and regionally relevant research. Evidence from digital extension provides the strongest causal baseline: mobile and personalised advisory services frequently improve information recall, agronomic knowledge and adoption of recommended practices, yet effects on yield, profit and welfare are inconsistent. Recent generative-AI studies show that large language models can produce useful agricultural responses, but site-specific rates, timing and local practice remain recurrent failure points. Retrieval grounding and expert feedback improve local relevance, although multi-season farm-level effectiveness evidence remains scarce. Image-based plant-disease systems achieve high accuracy in curated datasets, but performance can deteriorate sharply under field domain shift, class novelty and variable image quality. North-East Indian studies of mobile advisory systems in Meghalaya, Nagaland and Tripura demonstrate a valuable institutional foundation based on interactive voice response, local expert networks and user-centred service design; they do not, however, establish the effectiveness of autonomous AI. The most defensible deployment model is therefore an offline-tolerant, multilingual, multimodal and human-supervised architecture that grounds recommendations in curated regional knowledge, represents uncertainty, preserves provenance and escalates high-risk or out-of-distribution cases. Future research should prioritise prospective district- and season-spanning evaluations that connect model quality to farmer decisions, agronomic outcomes, equity, safety and cost-effectiveness.","人工智能（AI）正通过机器学习、计算机视觉、对话式大语言模型、检索增强生成和多模态界面被引入农业咨询服务。对于小农户而言，核心问题不在于这些技术能否产生技术上看似合理的输出，而在于它们能否在异质的农艺、语言和网络连接条件下提供本地正确、可操作且安全的建议。本批判性叙事综述整合了数字推广、AI赋能的农业建议、基于图像的诊断和负责任数字农业方面的证据，并特别关注印度东北部。本文考察了2010年1月1日至2026年7月17日期间发表的文献，重点关注同行评议的田间评估、技术验证研究、综述及区域相关研究。来自数字推广的证据提供了最强的因果基线：移动化和个性化咨询服务经常改善信息记忆、农艺知识和对推荐措施的采纳，但对产量、利润和福利的影响并不一致。近期生成式AI研究表明，大语言模型能够产生有用的农业回答，但针对具体地点的用量、时机和本地实践仍是反复出现的失败点。检索 grounding 和专家反馈可提高本地相关性，但多季农场层面的有效性证据仍然稀缺。基于图像的植物病害系统在精选数据集上达到高准确率，但在田间域偏移、类别新颖性和图像质量多变的情况下，性能可能急剧下降。印度东北部在梅加拉亚邦、那加兰邦和特里普拉邦开展的移动咨询系统研究展示了基于交互式语音应答、本地专家网络和以用户为中心的服务设计的宝贵制度基础；然而，这些研究并未确立自主AI的有效性。因此，最可辩护的部署模式是一种容忍离线、多语言、多模态且有人工监督的架构，该架构将建议建立在精选的区域知识之上，表征不确定性，保留来源信息，并对高风险或分布外案例进行升级处理。未来研究应优先开展前瞻性的跨区县和跨季节评估，将模型质量与农户决策、农艺结果、公平性、安全性和成本效益联系起来。","Asian Research Journal of Agriculture","2026-09-24T00:00:00Z",80,{"impact":17,"substance":18,"depth":17,"authority":334,"freshness":335,"relevant":21,"comment":336},13,9,"系统综述AI农业咨询与诊断系统在印度东北小农场景的技术架构与落地证据，指出人机协同、离线多语言与检索增强是可行路径，对智慧农业落地有参考价值。",[338],{"name":330,"url":327},[340,26,157,341,30],"数字乡村","农业技术推广",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[344],{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},{"status":51,"locked":14,"version":52,"method":53,"evidence":346},{"candidates":347,"agricultural_evidence":58},[348,352,357,363,368],{"code":32,"score":57,"supported":58,"title":349,"abstract":350,"tags":351},[60],[30,60],[30],{"code":45,"score":13,"supported":58,"title":353,"abstract":354,"tags":356},[],[29,355,73,287,172],"计算机视觉",[157],{"code":77,"score":78,"supported":58,"title":358,"abstract":359,"tags":362},[],[360,361],"数字农业","digital agriculture",[26],{"code":364,"score":21,"supported":14,"title":365,"abstract":366,"tags":367},"G1",[],[],[341],{"code":369,"score":21,"supported":14,"title":370,"abstract":371,"tags":372},"I6",[],[],[340],[374,375],"印度东北部 农业AI 小农户","农业智能诊断 多语言 离线","印度东北部农业AI小农户-3512","10.9734\u002Farja\u002F2026\u002Fv19i4919",{"doi":377,"openalex_id":379,"authors":380,"venue":330,"cited_by_count":130,"oa_url":327,"card":386,"direction":391,"ingested_from":137},"W7214205238",[381,383],{"name":382,"orcid":9},"Pravangkar Boruah",{"name":384,"orcid":385},"Rubul Kumar Bania","https:\u002F\u002Forcid.org\u002F0000-0001-6294-0231",{"tldr":387,"method":388,"finding":389,"direction":197,"opportunity":390},"综述AI农业咨询与诊断系统，聚焦印度东北小农，提出人监督多模态部署架构。","批判性叙述综述，整合2010-2026年数字推广、生成式AI与图像诊断证据。","AI输出技术可行但本地化、安全与田间效果证据不足，需人监督与检索增强。","可开展跨区跨季前瞻评估，连接模型质量与农户决策、产量、公平及成本效益。","数字乡村与农业信息化","2026-09-25T23:30:39.745514Z",{"id":394,"title":395,"url":396,"summary":397,"summary_zh":398,"content":9,"source_name":399,"source_url":396,"published_at":331,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":332,"score_detail":400,"sources":402,"tags":404,"primary_topic":407,"secondary_topics":408,"classification":410,"search_phrases":443,"slug":446,"view_count":447,"doi":448,"paper":449,"created_at":473},3469,"ARTIFICIAL INTELLIGENCE FOR AGRICULTURE: A SYSTEMATIC REVIEW OF FARMERS' PERCEPTIONS, ACCEPTANCE, ADOPTION AND BARRIERS","https:\u002F\u002Fdoi.org\u002F10.64013\u002Fbbasrjlifess.v2026i1.70","Artificial intelligence (AI) is rapidly reshaping agriculture, enabling the application of data for efficient decision-making, precision farming, crop monitoring, pest management, disease detection, smart irrigation, yield prediction, and various aspects of farm automation. Yet, to successfully gain a foothold in farms, factors like farmers' perceptions, their acceptance, desire, and readiness to adopt, as well as the capability to surmount socioeconomic, technological, and institutional challenges, must also make a difference. This article systematically collates the available information on farmers' perspectives on, acceptance of, readiness to adopt, and the obstacles related to AI in the agricultural context. The PRISMA 2020 guideline has been followed to select studies that are related, evaluated for inclusion, and finally, per the criteria, combined into a systematic synthesis. The review mainly discusses the aspects that influence the use of AI, like perceived usefulness, ease of use, trust, digital literacy, affordability, farm size, socioeconomic characteristics, availability of digital infrastructure, and access to agricultural technical advisory services. Main challenges identified involve the high cost of implementing AI, poor connectivity, weak rural infrastructure, low levels of technological knowledge, unavailability of support services, concerns about data privacy, distrust of algorithmic bias, language barriers, and inequalities impacting smallholder farmers. Besides barriers, the review indicates the potential of leveraging AI through extension services, agricultural mobile apps, precision farming, climate-smart agriculture, early-warning systems, and tailored farm advisories. By drawing out technical, behavioral, socioeconomic, and institutional perspectives, the review pinpoints important research questions and presents a farmholder-oriented setup to explain the process of taking up AI in farming. The outcomes can be used by scientific experts, policymakers, extension personnel, and software developers to create low-cost, reliable, and accessible AI systems aimed at promoting sustainable agricultural development.","人工智能（AI）正迅速重塑农业，使数据得以应用于高效决策、精准农业、作物监测、病虫害管理、病害检测、智能灌溉、产量预测以及农场自动化的各个方面。然而，要在农场中成功立足，农民的认知、接受度、采用意愿和准备程度，以及克服社会经济、技术和制度挑战的能力，同样至关重要。本文系统梳理了现有关于农民对农业领域人工智能的看法、接受度、采用准备程度及相关障碍的信息。研究遵循PRISMA 2020指南，筛选相关研究，评估其纳入资格，并最终依据标准进行系统性综合。综述主要探讨了影响人工智能使用的因素，包括感知有用性、易用性、信任、数字素养、可负担性、农场规模、社会经济特征、数字基础设施的可用性以及农业技术咨询服务的获取。识别出的主要挑战包括人工智能实施成本高昂、网络连接不佳、农村基础设施薄弱、技术水平低下、支持服务缺乏、数据隐私担忧、对算法偏见的疑虑、语言障碍以及影响小农户的不平等问题。除障碍外，综述还指出了通过推广服务、农业移动应用、精准农业、气候智慧型农业、预警系统和定制化农场咨询来利用人工智能的潜力。通过梳理技术、行为、社会经济和制度层面的视角，本综述指出了重要的研究问题，并提出了一个以农场主为导向的框架，以解释在农业中采用人工智能的过程。研究结果可供科学专家、政策制定者、推广人员和软件开发者用于创建低成本、可靠且可及的人工智能系统，以促进可持续农业发展。","Journal of Life and Social Sciences",{"impact":17,"substance":18,"depth":17,"authority":334,"freshness":335,"relevant":21,"comment":401},"基于PRISMA的系统综述，系统梳理农户对AI的认知、接受度与采纳障碍，对智慧农业推广与政策设计有实质参考价值。",[403],{"name":399,"url":396},[26,157,405,406,30],"数字素养","农户采纳",{"code":45,"slug":46,"name":47,"path":48,"parent_code":49},[409],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":51,"locked":14,"version":52,"method":53,"evidence":411},{"candidates":412,"agricultural_evidence":58},[413,417,421,426,433,439],{"code":45,"score":19,"supported":58,"title":414,"abstract":415,"tags":416},[287],[287],[157],{"code":32,"score":151,"supported":58,"title":418,"abstract":419,"tags":420},[],[30,60],[406,30],{"code":77,"score":78,"supported":58,"title":422,"abstract":423,"tags":425},[],[81,424],"smart agriculture",[26],{"code":427,"score":93,"supported":58,"title":428,"abstract":429,"tags":432},"H3",[],[430,431],"灌溉","irrigation",[],{"code":434,"score":115,"supported":58,"title":435,"abstract":436,"tags":438},"H8",[],[437],"可持续农业",[],{"code":369,"score":21,"supported":14,"title":440,"abstract":441,"tags":442},[],[],[405],[444,445],"农民 AI 采纳 障碍","农业人工智能 系统综述","农民AI采纳障碍-3469",5,"10.64013\u002Fbbasrjlifess.v2026i1.70",{"doi":448,"openalex_id":450,"authors":451,"venue":399,"cited_by_count":130,"oa_url":396,"card":468,"direction":135,"ingested_from":137},"W7214189410",[452,454,456,458,460,462,464,466],{"name":453,"orcid":9},"MM JAMEEL",{"name":455,"orcid":9},"M SAEED",{"name":457,"orcid":9},"SA SHER",{"name":459,"orcid":9},"Z ALI",{"name":461,"orcid":9},"Q HAYYAT",{"name":463,"orcid":9},"S KIRBAG",{"name":465,"orcid":9},"S KHAN",{"name":467,"orcid":9},"HN AHMAD",{"tldr":469,"method":470,"finding":471,"direction":391,"opportunity":472},"系统综述农民对农业AI的感知、接受度、采纳意愿及障碍。","遵循PRISMA 2020指南，系统筛选并综合相关文献。","成本、基础设施、数字素养与信任是主要障碍，小农户受影响最大。","可研究低成本、本地化AI采纳模型及小农户数字包容机制。","2026-09-25T23:30:09.670568Z",{"id":475,"title":476,"url":477,"summary":478,"summary_zh":479,"content":9,"source_name":480,"source_url":477,"published_at":481,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":482,"score_detail":483,"sources":485,"tags":487,"primary_topic":489,"secondary_topics":490,"classification":493,"search_phrases":528,"slug":531,"view_count":20,"doi":532,"paper":533,"created_at":547},3010,"A Multi-Task Stacked Ensemble and IoT-Enabled Decision Support System for Precision Fertigation in Smallholder Agriculture","https:\u002F\u002Fdoi.org\u002F10.5120\u002Fijca9a6e56b81235","Nigerian agriculture's fixed-schedule fertigation causes low efficiency and nutrient leaching.A stacked-ensemble model is developed for precision fertigation that jointly predicts fertigation need, rate (kg\u002Fha) and timing (Early\u002FOptimal\u002FLate).To train and evaluate the ensemble, a unified dataset was integrated, comprising 12,840 records and 42 variables from a Nigerian soil-weather-yield dataset, a locally sourced Nigerian IoT sensor series and historical weather\u002FNDVI feeds.An LSTM soil-dynamics model, an XGBoost rate regressor and Random Forest need\u002Ftiming classifiers are fused through an XGBoost meta-learner trained on out-of-fold predictions.On held-out partitions, the ensemble reduced rate MAE from 0.55 to 0.49 kg\u002Fha (-10.9%) and RMSE from 0.68 to 0.61 kg\u002Fha (-10.3%;R² 0.88→0.92),raised need F1 from 0.83 to 0.86 (accuracy 0.87→0.89;AUC 0.89→0.93)and timing macro-F1 from 0.84 to 0.86, with well-calibrated probabilities (Brier 0.082).The trained ensemble was deployed through a RESTful API and responsive dashboard; under concurrent load, the system recorded 0% request errors with 1.88s median API latency, demonstrating practical deployability for Nigerian smallholder agriculture.","尼日利亚农业的固定日程水肥一体化导致效率低下和养分淋失。本研究开发了一种堆叠集成模型用于精准水肥管理，可联合预测灌溉施肥需求、施用量（kg\u002Fha）和时机（早\u002F最佳\u002F晚）。为训练和评估该集成模型，整合了一个统一数据集，包含来自尼日利亚土壤-天气-产量数据集、本地尼日利亚物联网传感器序列及历史天气\u002FNDVI数据的12,840条记录和42个变量。通过基于折外预测训练的XGBoost元学习器，将LSTM土壤动力学模型、XGBoost施用量回归器和随机森林需求\u002F时机分类器进行融合。在留出集上，该集成模型将施用量MAE从0.55降至0.49 kg\u002Fha（-10.9%），RMSE从0.68降至0.61 kg\u002Fha（-10.3%；R² 0.88→0.92），需求F1从0.83提升至0.86（准确率0.87→0.89；AUC 0.89→0.93），时机宏平均F1从0.84提升至0.86，且概率校准良好（Brier 0.082）。训练后的集成模型通过RESTful API和响应式仪表板部署；在并发负载下，系统录得0%请求错误，API延迟中位数为1.88秒，展示了在尼日利亚小农农业中的实际可部署性。","International Journal of Computer Applications","2026-09-18T00:00:00Z",79,{"impact":17,"substance":18,"depth":17,"authority":334,"freshness":151,"relevant":21,"comment":484},"面向小农户的精准水肥一体化多任务集成模型与物联网决策支持系统，数据规模与方法验证扎实，对智慧农业落地有参考价值。",[486],{"name":480,"url":477},[26,157,267,30,488],"精准灌溉",{"code":39,"slug":40,"name":41,"path":42,"parent_code":43},[491,492],{"code":107,"slug":273,"name":274,"path":275,"parent_code":49},{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":51,"locked":14,"version":52,"method":53,"evidence":494},{"candidates":495,"agricultural_evidence":58},[496,503,507,511,515,519,524],{"code":39,"score":497,"supported":58,"title":498,"abstract":500,"tags":502},17,[499],"fertigation",[66,501,499],"水肥一体化",[],{"code":107,"score":57,"supported":58,"title":504,"abstract":505,"tags":506},[111],[110,111],[267],{"code":32,"score":19,"supported":58,"title":508,"abstract":509,"tags":510},[60],[60],[30],{"code":45,"score":335,"supported":58,"title":512,"abstract":513,"tags":514},[74],[],[157],{"code":100,"score":93,"supported":58,"title":516,"abstract":517,"tags":518},[],[103,104],[],{"code":427,"score":520,"supported":58,"title":521,"abstract":522,"tags":523},4,[],[430],[488],{"code":77,"score":21,"supported":14,"title":525,"abstract":526,"tags":527},[],[],[26],[529,530],"尼日利亚 精准灌溉 物联网","堆叠集成 施肥决策 小农户","尼日利亚精准灌溉物联网-3010","10.5120\u002Fijca9a6e56b81235",{"doi":532,"openalex_id":534,"authors":535,"venue":480,"cited_by_count":130,"oa_url":477,"card":542,"direction":135,"ingested_from":137},"W7213663876",[536,538,540],{"name":537,"orcid":9},"Awojide S.",{"name":539,"orcid":9},"Ikpotokin F.O.",{"name":541,"orcid":9},"Sadiq F.I.",{"tldr":543,"method":544,"finding":545,"direction":135,"opportunity":546},"构建多任务堆叠集成模型与物联网决策支持系统，实现小农户精准水肥一体化。","LSTM、XGBoost、随机森林堆叠集成，融合尼日利亚土壤气象、IoT与NDV","集成模型将施肥量MAE降低10.9%，需求与时机分类F1提升，系统部署零错误。","可探索多任务集成模型在非洲小农户不同作物与气候区的迁移能力及低成本IoT部署。","2026-09-20T23:30:09.003454Z"]