[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3621":3,"related-3621":113},{"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":38,"search_phrases":82,"slug":85,"view_count":86,"doi":87,"paper":88,"created_at":112},3621,"Beyond Precision Agriculture: AI and Data-driven Information Systems as Catalysts for Sustainable Agribusiness Transformation","https:\u002F\u002Fdoi.org\u002F10.56557\u002Fjgembr\u002F2026\u002Fv18i311156","The dominant narrative in digital agriculture centres on discrete technologies, including sensors, unmanned aerial vehicles, global positioning systems, remote sensing, and machine learning, evaluated primarily for their capacity to increase yields and improve input-use efficiency. This narrows the analytical lens to the farm gate and understates a more consequential shift: individual technologies generate value for agribusiness only when information systems capture, integrate, and route the data they produce into real operational and strategic decisions. This review critically examines the AI- and data-driven agribusiness transformation literature through a single organising pathway: data, analytics, artificial intelligence, information systems, decisions, sustainable agribusiness outcomes; it asks whether the field's technical achievements have matched comparable progress in systems integration and governance. A transparent narrative search and appraisal protocol synthesised evidence across ten thematic domains: agricultural big-data integration, predictive and prescriptive analytics, farm management information systems (FMIS), AI-enabled supply-chain analytics, climate-risk adaptation, sustainability and ESG analytics, data governance and cybersecurity, digital infrastructure and adoption, circular agriculture, and decision-support systems. The evidence shows that sensing and algorithmic capabilities have matured faster than the organisational and governance capacity required to convert those capabilities into adopted agribusiness decisions. Data-governance barriers, rather than technological ones, now dominate constraints on information-system integration in the literature published since 2023. Research, investment, and policy attention should therefore shift further toward the interoperability, governance, and organisational design layers of digital agriculture, rather than continuing to concentrate primarily on sensor accuracy or model performance.","数字农业的主流叙事聚焦于各类离散技术，包括传感器、无人机、全球定位系统、遥感和机器学习，并主要依据其提高产量和改善投入品利用效率的能力加以评价。这种视角将分析视野收窄至农场大门，低估了一个更具深远意义的转变：只有当信息系统能够捕获、整合单个技术所产生的数据，并将其导入实际运营和战略决策时，这些技术才能为农业企业创造价值。本综述通过一条单一的组织路径——数据、分析、人工智能、信息系统、决策、可持续农业企业成果——批判性地审视人工智能和数据驱动的农业企业转型文献，并追问该领域的技术成就是否与系统集成和治理方面的相应进展相匹配。一项透明的叙述性检索与评价方案综合了十个主题领域的证据：农业大数据集成、预测性与规范性分析、农场管理信息系统(FMIS)、人工智能赋能的供应链分析、气候风险适应、可持续性与ESG分析、数据治理与网络安全、数字基础设施与采用、循环农业以及决策支持系统。证据表明，感知和算法能力的成熟速度快于将这些能力转化为已被采用的农业企业决策所需的组织和治理能力。在2023年以来发表的文献中，数据治理障碍而非技术障碍，已成为信息系统集成的主要制约因素。因此，研究、投资和政策关注应进一步转向数字农业的互操作性、治理和组织设计层面，而非继续主要集中于传感器精度或模型性能。",null,"Journal of Global Economics Management and Business Research","2026-09-26T00:00:00Z","论文",10,false,80,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":20,"relevant":21,"comment":22},18,22,13,9,1,"该综述指出数据治理与系统集成而非传感技术本身才是数字农业落地的关键瓶颈，观点有实质增量，但属学术综述、产业影响偏间接。",[24],{"name":10,"url":6},[26,27,28,29,30],"智慧农业","农业人工智能","农业大数据","数据治理","农业信息体系",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":39,"locked":14,"version":40,"method":41,"evidence":42},"classified","2026-10-22.2","rules",{"candidates":43,"agricultural_evidence":47},[44,54,61,69,75],{"code":45,"score":46,"supported":47,"title":48,"abstract":50,"tags":53},"I1",15,true,[49],"precision agriculture",[51,52],"数字农业","digital agriculture",[26],{"code":32,"score":13,"supported":47,"title":55,"abstract":56,"tags":60},[],[57,58,59],"机器学习","machine learning","artificial intelligence",[27],{"code":62,"score":63,"supported":47,"title":64,"abstract":65,"tags":68},"I3",6,[],[66,67],"遥感","remote sensing",[],{"code":70,"score":71,"supported":47,"title":72,"abstract":73,"tags":74},"I6",5,[],[28],[28,29],{"code":76,"score":77,"supported":47,"title":78,"abstract":79,"tags":81},"H8",3,[],[80],"可持续农业",[],[83,84],"农业信息体系 数据治理","农场管理信息系统 FMIS","农业信息体系数据治理-3621",4,"10.56557\u002Fjgembr\u002F2026\u002Fv18i311156",{"doi":87,"openalex_id":89,"authors":90,"venue":10,"cited_by_count":103,"oa_url":6,"card":104,"direction":110,"ingested_from":111},"W7214471201",[91,93,95,97,99,101],{"name":92,"orcid":9},"Timilehin Fawibe",{"name":94,"orcid":9},"Gbenga Samuel Ernest",{"name":96,"orcid":9},"Joshua Emmanuel",{"name":98,"orcid":9},"Simon Jacob",{"name":100,"orcid":9},"Olawale Akinleye",{"name":102,"orcid":9},"Tosin Olorunwo",0,{"tldr":105,"method":106,"finding":107,"direction":108,"opportunity":109},"综述指出农业数字化瓶颈已从技术转向数据治理与系统集成，呼吁研究重心转向互操作与组织设计。","采用透明叙事检索与评价协议，综合十大主题领域的AI与数据驱动农业文献。","2023年后数据治理障碍而非技术障碍，成为农业信息系统集成的主要制约。","数字乡村与农业信息化","可研究农业数据互操作标准、治理机制与组织采纳路径如何影响可持续农业商业决策。","农业遥感与作物表型","openalex","2026-09-27T23:30:46.092194Z",{"total":63,"page":21,"page_size":63,"items":114},[115,239,325,395,456,513],{"id":116,"title":117,"url":118,"summary":119,"summary_zh":120,"content":9,"source_name":121,"source_url":118,"published_at":122,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":123,"score_detail":124,"sources":129,"tags":131,"primary_topic":134,"secondary_topics":135,"classification":146,"search_phrases":213,"slug":216,"view_count":103,"doi":217,"paper":218,"created_at":238},4302,"BIG DATA IN AGRONOMY: TECHNOLOGIES, APPLICATIONS, BENEFITS AND CHALLENGES","https:\u002F\u002Fdoi.org\u002F10.4238\u002Fk6hcgw77","Big Data is becoming an important component of modern agronomy by enabling the integration, processing and analysis of large, diverse and continuously generated datasets for improved agricultural decision-making. Agricultural Big Data is generated from multiple sources, including remote sensing satellites, Unmanned Aerial Vehicles (UAVs), Internet of Things (IoT) sensors, weather stations, farm machinery, field observations and historical databases. The integration of these data sources with Geographic Information Systems (GIS), cloud computing, Artificial Intelligence (AI) and Machine Learning (ML) enables the identification of spatial and temporal patterns that are difficult to detect using conventional approaches. Big Data applications in agronomy include climate and weather prediction, pest and disease management, soil health and nutrient management, irrigation scheduling, crop yield prediction, precision agriculture, crop insurance and weed management. Examples from India and other regions demonstrate the application of satellite imagery, UAV-based sensing, IoT systems, crop-growth models and machine-learning algorithms for field-scale and regional agricultural decision making. Big Data analytics can improve resource-use efficiency, support predictive and real-time monitoring, strengthen agricultural risk assessment and facilitate site-specific management. However, effective implementation is constrained by challenges such as high infrastructure and operational costs, data quality and interoperability issues, data privacy and ownership concerns, limited rural digital infrastructure, inadequate technical skills and increasing requirements for data storage and computing. Therefore, the effective integration of Big Data with agronomic knowledge and decision-support systems can contribute to more precise, efficient and sustainable agricultural production while addressing emerging challenges related to climate variability, resource scarcity and food security.","大数据正成为现代农学的重要组成部分，它能够整合、处理和分析大规模、多样化且持续生成的数据集，从而改进农业决策。农业大数据来源于多种渠道，包括遥感卫星、无人机（UAV）、物联网（IoT）传感器、气象站、农业机械、田间观测和历史数据库。将这些数据源与地理信息系统（GIS）、云计算、人工智能（AI）和机器学习（ML）相结合，能够识别传统方法难以发现的空间和时间模式。大数据在农学中的应用包括气候与天气预测、病虫害管理、土壤健康与养分管理、灌溉调度、作物产量预测、精准农业、作物保险和杂草管理。来自印度及其他地区的案例展示了卫星影像、基于无人机的遥感、物联网系统、作物生长模型和机器学习算法在田块尺度和区域农业决策中的应用。大数据分析可以提高资源利用效率，支持预测性和实时监测，加强农业风险评估，并促进因地制宜的管理。然而，有效实施受到诸多挑战的制约，如高昂的基础设施和运营成本、数据质量和互操作性问题、数据隐私和所有权担忧、农村数字基础设施有限、技术技能不足以及数据存储和计算需求不断增加。因此，将大数据与农学知识和决策支持系统有效整合，有助于实现更精准、高效和可持续的农业生产，同时应对气候变异性、资源稀缺和粮食安全等方面的新兴挑战。","Genetics and Molecular Research","2026-10-05T00:00:00Z",76,{"impact":17,"substance":125,"depth":126,"authority":19,"freshness":127,"relevant":21,"comment":128},20,17,8,"系统综述农业大数据技术、应用与挑战，信息密度高，对智慧农业决策有参考价值，但属综述性论文，非突破性成果。",[130],{"name":121,"url":118},[26,27,28,132,133],"精准农业","遥感监测",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[136,141],{"code":137,"slug":138,"name":139,"path":140,"parent_code":36},"I4","gis","GIS与空间分析","\u002Ftopic\u002Fdigital-agriculture\u002Fgis",{"code":142,"slug":143,"name":144,"path":145,"parent_code":36},"I5","agricultural-iot","物联网传感与边缘计算","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-iot",{"status":39,"locked":14,"version":40,"method":41,"evidence":147},{"candidates":148,"agricultural_evidence":47},[149,153,160,167,171,176,183,190,197,201,207],{"code":32,"score":13,"supported":47,"title":150,"abstract":151,"tags":152},[],[57,58,59],[27],{"code":137,"score":20,"supported":47,"title":154,"abstract":155,"tags":159},[],[156,157,158],"地理信息","GIS","geographic information",[],{"code":142,"score":20,"supported":47,"title":161,"abstract":162,"tags":166},[],[163,164,165],"物联网","internet of things","IoT",[],{"code":45,"score":127,"supported":47,"title":168,"abstract":169,"tags":170},[],[132,49],[26,132],{"code":62,"score":172,"supported":47,"title":173,"abstract":174,"tags":175},7,[],[66,67],[133],{"code":177,"score":63,"supported":47,"title":178,"abstract":179,"tags":182},"A8",[],[180,181],"粮食安全","food security",[],{"code":184,"score":63,"supported":47,"title":185,"abstract":186,"tags":189},"H2",[],[187,188],"土壤","soil",[],{"code":191,"score":63,"supported":47,"title":192,"abstract":193,"tags":196},"H3",[],[194,195],"灌溉","irrigation",[],{"code":70,"score":86,"supported":47,"title":198,"abstract":199,"tags":200},[],[28],[28],{"code":202,"score":77,"supported":47,"title":203,"abstract":204,"tags":206},"F3",[],[205],"养分管理",[],{"code":208,"score":77,"supported":47,"title":209,"abstract":210,"tags":212},"F5",[],[211],"农业机械",[],[214,215],"农业大数据 遥感 物联网","机器学习 作物产量预测","农业大数据遥感物联网-4302","10.4238\u002Fk6hcgw77",{"doi":217,"openalex_id":219,"authors":220,"venue":121,"cited_by_count":103,"oa_url":118,"card":231,"direction":237,"ingested_from":111},"W7220374836",[221,223,225,227,229],{"name":222,"orcid":9},"Kodavali Aswini",{"name":224,"orcid":9},"Raju Marimuthu",{"name":226,"orcid":9},"Guru Gurusamy",{"name":228,"orcid":9},"Kannan Pandian",{"name":230,"orcid":9},"Anandham Rangasamy",{"tldr":232,"method":233,"finding":234,"direction":235,"opportunity":236},"综述农学大数据的技术、应用、效益与挑战，强调多源数据融合支撑精准农业决策。","整合遥感、无人机、物联网、GIS、云计算与AI\u002FML的多源农业数据分析。","大数据可提升资源效率与实时监测，但受成本、数据质量、隐私和农村数字基建制约。","农业人工智能与决策模型","可研究低成本边缘计算与数据互操作标准，破解小农户场景下大数据落地瓶颈。","智慧农业 \u002F 农业物联网","2026-10-06T23:30:11.880721Z",{"id":240,"title":241,"url":242,"summary":243,"summary_zh":9,"content":9,"source_name":244,"source_url":9,"published_at":245,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":246,"score_detail":247,"sources":252,"tags":254,"primary_topic":257,"secondary_topics":258,"classification":259,"search_phrases":312,"slug":315,"view_count":127,"doi":9,"paper":316,"created_at":324},3326,"Research on the Application of Agricultural Big Data in Plant Growth Prediction——基于多源数据同化与混合智能（MDA-HI）框架","https:\u002F\u002Fwww.icck.org\u002Ffilebob\u002Fuploads\u002Fstorage\u002FDIA_ANrbHOHS4uBBpnVQ6.pdf","《Digital Intelligence in Agriculture》2026年第2卷第2期。Wei Yongqiang等提出Multi-source Data Assimilation and Hybrid Intelligence（MDA-HI）框架，将基于过程的作物模型与集成机器学习算法（包括基于Transformer的架构和物理信息神经网络）相结合。在2023—2025年中国多生态区主要作物（水稻、小麦、玉米、番茄）的实证验证中：MDA-HI模型实现了产量预测RMSE平均减少42.7%、关键物候期预测减少38.1%。水稻-小麦轮作系统大规模案例研究显示数据驱动处方可将氮肥使用减少22.5%、灌溉水减少18.3%，同时产量增加5.1%。","《Digital Intelligence in Agriculture》2026; 2(2):54-67","2026-09-17T00:00:00Z",85,{"impact":18,"substance":248,"depth":249,"authority":250,"freshness":63,"relevant":21,"comment":251},24,19,14,"多源数据同化与混合智能框架在四大作物上验证，减肥节水增产数据扎实，方法新颖且具产业推广价值。",[253],{"name":244,"url":242},[26,27,255,28,256],"精准施肥","作物生长预测",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":260},{"candidates":261,"agricultural_evidence":47},[262,266,272,278,284,290,296,300,304,308],{"code":32,"score":86,"supported":47,"title":263,"abstract":264,"tags":265},[],[57],[27],{"code":267,"score":77,"supported":47,"title":268,"abstract":269,"tags":271},"A1",[],[270],"水稻",[],{"code":273,"score":77,"supported":47,"title":274,"abstract":275,"tags":277},"A2",[],[276],"小麦",[],{"code":279,"score":77,"supported":47,"title":280,"abstract":281,"tags":283},"A3",[],[282],"玉米",[],{"code":285,"score":77,"supported":47,"title":286,"abstract":287,"tags":289},"A7",[],[288],"轮作",[],{"code":291,"score":77,"supported":47,"title":292,"abstract":293,"tags":295},"B1",[],[294],"番茄",[],{"code":191,"score":77,"supported":47,"title":297,"abstract":298,"tags":299},[],[194],[],{"code":202,"score":21,"supported":14,"title":301,"abstract":302,"tags":303},[],[],[255],{"code":45,"score":21,"supported":14,"title":305,"abstract":306,"tags":307},[],[],[26],{"code":70,"score":21,"supported":14,"title":309,"abstract":310,"tags":311},[],[],[28],[313,314],"MDA-HI 多源数据同化 作物模型","水稻小麦轮作 氮肥减量 产量预测","MDA-HI多源数据同化作物模型-3326",{"doi":9,"openalex_id":9,"authors":317,"venue":9,"cited_by_count":103,"oa_url":9,"card":318,"direction":235,"ingested_from":323},[],{"tldr":319,"method":320,"finding":321,"direction":235,"opportunity":322},"提出MDA-HI框架，融合过程模型与混合智能，用于作物生长与产量预测。","多源数据同化结合Transformer与物理信息神经网络，在中国多生态区验证。","产量预测RMSE降42.7%，氮肥减22.5%、灌溉水减18.3%，产量增5.1%。","可探索轻量化MDA-HI在边缘设备部署及跨区域迁移能力，降低小农户应用门槛。","agent","2026-09-24T00:04:02.947080Z",{"id":326,"title":327,"url":328,"summary":329,"summary_zh":330,"content":9,"source_name":331,"source_url":328,"published_at":332,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":333,"score_detail":334,"sources":336,"tags":338,"primary_topic":342,"secondary_topics":343,"classification":344,"search_phrases":370,"slug":373,"view_count":77,"doi":374,"paper":375,"created_at":394},2292,"A governance-first framework integrates artificial intelligence and blockchain to advance sustainable and inclusive horticultural value chains","https:\u002F\u002Fdoi.org\u002F10.3389\u002Ffsufs.2026.1894057","Horticultural value chains face persistent agronomic challenges including production inefficiencies, postharvest losses, poor resource management, and limited supply-chain transparency, particularly in smallholder-dominated systems across the Global South. These constraints undermine agricultural productivity, market access, and the long-term sustainability of horticultural farming. Despite growing research interest in digital agriculture, artificial intelligence (AI) and blockchain technology are increasingly explored together, including in emerging integrated AI-blockchain-IoT architectures for agricultural traceability. However, these existing integrations remain predominantly technology-centric, prioritizing traceability, automation, and data verification, without an explicit governance, equity, or data-sovereignty architecture designed specifically for smallholder-dominated horticultural production and distribution systems. This study proposes a comprehensive conceptual framework that integrates AI’s predictive and decision-support capabilities with blockchain’s data integrity, traceability, and automation features to enhance sustainability, inclusivity, and governance in horticultural value chains. Drawing on an integrative review of peer-reviewed sources identified through Scopus, Web of Science, IEEE Xplore, ScienceDirect, and Google Scholar, and grounded in the Technology-Organisation-Environment (TOE) framework, Sociotechnical Systems Theory, and Sustainable Digital Transformation Theory, the study addresses the agronomic, governance, and equity dimensions of digital transformation in horticulture. The framework is structured around five interlinked operational layers, data collection, AI-driven analytics and decision-making, blockchain validation and traceability, smart contracts and decision support, and user interface and stakeholder feedback, all operating within a governance architecture and ethical design layer that functions as a structural precondition for the framework rather than a downstream feature. This governing layer embeds algorithmic transparency, data sovereignty through community-governed data cooperatives, and inclusive digital access for smallholder farmers as non-negotiable design principles. The framework explicitly aligns with Sustainable Development Goals 2, 9, 10, 12, 13, and 16, and includes a three-phase empirical validation pathway comprising expert Delphi elicitation, simulation modeling, and pilot implementation in horticultural cooperatives. The integration of AI and blockchain within an institutionally governed, equity-oriented architecture represents a paradigm shift from technology-first to governance-first digital transformation in horticulture, providing a theoretically robust and practically actionable foundation for future research, systems design, and policy development, with direct application to resource-constrained smallholder systems in sub-Saharan Africa and comparable Global South contexts.","园艺价值链面临持续存在的农艺挑战，包括生产效率低下、采后损失、资源管理不善以及供应链透明度有限，在全球南方以小农为主的体系中尤为突出。这些制约因素削弱了农业生产力、市场准入以及园艺种植的长期可持续性。尽管数字农业的研究兴趣日益增长，人工智能（AI）与区块链技术正越来越多地被共同探索，包括用于农业溯源的新兴AI-区块链-物联网集成架构。然而，现有这些集成仍以技术为中心，优先考虑溯源、自动化和数据验证，缺乏专门针对以小农为主的园艺生产和分销体系而设计的明确的治理、公平或数据主权架构。本研究提出一个综合性概念框架，将AI的预测与决策支持能力与区块链的数据完整性、溯源性和自动化特征相结合，以增强园艺价值链的可持续性、包容性和治理水平。基于对通过Scopus、Web of Science、IEEE Xplore、ScienceDirect和Google Scholar识别的同行评审文献的整合性综述，并以技术-组织-环境（TOE）框架、社会技术系统理论和可持续数字化转型理论为基础，本研究探讨了园艺数字转型中的农艺、治理和公平维度。该框架围绕五个相互关联的操作层构建：数据采集、AI驱动的分析与决策、区块链验证与溯源、智能合约与决策支持，以及用户界面与利益相关者反馈，所有这些都在治理架构与伦理设计层内运作，该层作为框架的结构性前提而非下游功能发挥作用。这一治理层将算法透明度、通过社区治理的数据合作社实现的数据主权，以及小农的包容性数字接入作为不可协商的设计原则。该框架明确与可持续发展目标2、9、10、12、13和16相一致，并包含一个三阶段实证验证路径，包括专家德尔菲征询、仿真建模以及在","Frontiers in Sustainable Food Systems","2026-09-11T00:00:00Z",78,{"impact":17,"substance":18,"depth":17,"authority":19,"freshness":172,"relevant":21,"comment":335},"该研究提出治理优先的AI-区块链融合框架，聚焦小农园艺价值链的可持续与包容性，方法新颖、理论扎实，对数字农业政策设计有参考价值，但属概念框架尚未实证，时效性一般。",[337],{"name":331,"url":328},[26,27,29,339,340,341],"小农户","农业区块链","园艺价值链",{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},[],{"status":39,"locked":14,"version":40,"method":41,"evidence":345},{"candidates":346,"agricultural_evidence":47},[347,352,356,360,366],{"code":32,"score":46,"supported":47,"title":348,"abstract":349,"tags":351},[59],[59,350],"decision support",[27],{"code":45,"score":172,"supported":47,"title":353,"abstract":354,"tags":355},[],[51,52],[26],{"code":142,"score":63,"supported":47,"title":357,"abstract":358,"tags":359},[],[163,165],[],{"code":361,"score":86,"supported":47,"title":362,"abstract":363,"tags":365},"G2",[],[364],"smallholder",[339],{"code":70,"score":21,"supported":14,"title":367,"abstract":368,"tags":369},[],[],[29],[371,372],"农业人工智能 农业区块链 园艺价值链 数据治理","农业人工智能 农业区块链","农业人工智能农业区块链园艺价值链数据治理-2292","10.3389\u002Ffsufs.2026.1894057",{"doi":374,"openalex_id":376,"authors":377,"venue":331,"cited_by_count":103,"oa_url":328,"card":389,"direction":237,"ingested_from":111},"W7212272522",[378,380,383,386],{"name":379,"orcid":9},"Olaoluwa Olarewaju",{"name":381,"orcid":382},"Tobi Fadiji","https:\u002F\u002Forcid.org\u002F0000-0001-5525-4495",{"name":384,"orcid":385},"Olaniyi Amos Fawole","https:\u002F\u002Forcid.org\u002F0000-0001-5591-4633",{"name":387,"orcid":388},"Lembe Samukelo Magwaza","https:\u002F\u002Forcid.org\u002F0000-0001-9809-2254",{"tldr":390,"method":391,"finding":392,"direction":108,"opportunity":393},"提出治理优先框架，整合AI与区块链提升园艺价值链可持续性与包容性。","整合性文献综述，基于TOE、社会技术系统与可持续数字化转型理论。","治理架构应作为结构性前提，嵌入算法透明、数据主权与包容性数字接入。","可实证检验社区数据合作社与治理层对小型农户采纳和公平收益的影响。","2026-09-13T23:30:07.102172Z",{"id":396,"title":397,"url":398,"summary":399,"summary_zh":400,"content":9,"source_name":401,"source_url":398,"published_at":402,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":403,"score_detail":404,"sources":408,"tags":410,"primary_topic":413,"secondary_topics":418,"classification":420,"search_phrases":436,"slug":439,"view_count":103,"doi":440,"paper":441,"created_at":455},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":127,"substance":405,"depth":17,"authority":406,"freshness":13,"relevant":21,"comment":407},21,12,"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。",[409],{"name":401,"url":398},[26,27,339,411,412],"植物病害识别","可解释性",{"code":361,"slug":414,"name":415,"path":416,"parent_code":417},"farm-services","社会化服务与规模经营","\u002Ftopic\u002Frural-services\u002Ffarm-services","G",[419],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":421},{"candidates":422,"agricultural_evidence":47},[423,427,432],{"code":361,"score":406,"supported":47,"title":424,"abstract":425,"tags":426},[364],[364],[339],{"code":32,"score":13,"supported":47,"title":428,"abstract":430,"tags":431},[429],"computer vision",[],[27,411],{"code":45,"score":21,"supported":14,"title":433,"abstract":434,"tags":435},[],[],[26],[437,438],"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":440,"openalex_id":442,"authors":443,"venue":401,"cited_by_count":103,"oa_url":398,"card":450,"direction":235,"ingested_from":111},"W7223137323",[444,446,448],{"name":445,"orcid":9},"Akshay Ramgoolam",{"name":447,"orcid":9},"Priya Bundhoo Appadoo",{"name":449,"orcid":9},"Jean-Noël LEspérance",{"tldr":451,"method":452,"finding":453,"direction":235,"opportunity":454},"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。","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":457,"title":458,"url":459,"summary":460,"summary_zh":461,"content":9,"source_name":401,"source_url":459,"published_at":402,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":403,"score_detail":462,"sources":464,"tags":466,"primary_topic":470,"secondary_topics":474,"classification":476,"search_phrases":495,"slug":498,"view_count":103,"doi":499,"paper":500,"created_at":512},5591,"Hierarchical Conformal Risk Control for Artificial Intelligence Agriculture Irrigation Scheduling Under Unmodeled Infiltration: A Computer Safety-Guarantee Evaluation","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23543665","Irrigation scheduling algorithms that rely on soil-moisture forecasts are vulnerable to unmodeled infiltration, the process by which water bypasses the root zone through preferential pathways that resistance-based soil models cannot represent. The framework is evaluated through a replication-based design study that characterises the operating characteristics of the before-after-control-impact estimator used to audit scheduler performance, under a declared variance structure with between-cluster standard deviation 6.0, residual standard deviation 4.5 and intraclass correlation 0.64. Power was estimated across 2000 design cells spanning cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The analysis indicates that 54 clusters are required to detect a 5-percentage-point effect at 80% power, and that the minimum detectable effect declines steeply as cluster count rises from 8 to 32. The contribution is a formal safety guarantee for irrigation decision support that remains valid when the infiltration model is wrong, together with an audit design that makes the guarantee empirically checkable.","依赖土壤湿度预报的灌溉调度算法容易受到未建模入渗的影响，即水分通过优先路径绕过根区的过程，而基于阻力的土壤模型无法表征这一过程。该框架通过一项基于重复的设计研究进行评估，该研究刻画了用于审计调度器性能的前后对照影响估计量的运行特性，其所依据的声明方差结构为：组间标准差6.0、残差标准差4.5、组内相关系数0.64。功效估计覆盖2000个设计单元，涵盖聚类数8、16、24和32以及效应量3、5和8个百分点。分析表明，在80%功效下检测5个百分点效应需要54个聚类，且最小可检测效应随聚类数从8增至32而急剧下降。本文的贡献在于为灌溉决策支持提供了一种正式的安全保证，该保证在入渗模型错误时仍然有效，同时提出了一种使该保证可经实证检验的审计设计。",{"impact":127,"substance":18,"depth":17,"authority":406,"freshness":20,"relevant":21,"comment":463},"该研究提出在入渗模型失配条件下仍有效的灌溉决策安全保证框架，并给出可审计的试验设计，方法新颖、结论可靠，但属细分方法学研究，产业影响有限，适合作为专业精选而非头条。",[465],{"name":401,"url":459},[26,27,467,468,469],"灌溉决策","安全保证","试验设计",{"code":191,"slug":195,"name":471,"path":472,"parent_code":473},"水资源与灌溉","\u002Ftopic\u002Fenvironment\u002Firrigation","H",[475],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":477},{"candidates":478,"agricultural_evidence":47},[479,483,487,491],{"code":191,"score":46,"supported":47,"title":480,"abstract":481,"tags":482},[195],[194,195],[467],{"code":32,"score":406,"supported":47,"title":484,"abstract":485,"tags":486},[59],[350],[27],{"code":184,"score":63,"supported":47,"title":488,"abstract":489,"tags":490},[],[187,188],[],{"code":45,"score":21,"supported":14,"title":492,"abstract":493,"tags":494},[],[],[26],[496,497],"灌溉调度 安全保证","农业人工智能 安全保证 智慧农业 灌溉决策","灌溉调度安全保证-5591","10.5281\u002Fzenodo.23543665",{"doi":499,"openalex_id":501,"authors":502,"venue":401,"cited_by_count":103,"oa_url":459,"card":507,"direction":235,"ingested_from":111},"W7223131205",[503,505],{"name":504,"orcid":9},"Nassima Benkhelifa",{"name":506,"orcid":9},"Yacine Aït Ahmed",{"tldr":508,"method":509,"finding":510,"direction":235,"opportunity":511},"提出分层保形风险控制，为未建模入渗下的灌溉调度提供安全保证。","分层保形风险控制与基于重复设计的BACI审计，2000个设计单元功效分析。","检测5个百分点效应需54个集群，最小可检测效应随集群数增加陡降。","可将保形风险控制扩展到其他未建模扰动下的农业决策安全保证与审计设计。","2026-10-22T11:21:10.283561Z",{"id":514,"title":515,"url":516,"summary":517,"summary_zh":518,"content":9,"source_name":401,"source_url":516,"published_at":402,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":519,"score_detail":520,"sources":522,"tags":524,"primary_topic":527,"secondary_topics":528,"classification":530,"search_phrases":545,"slug":548,"view_count":103,"doi":549,"paper":550,"created_at":564},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":127,"substance":18,"depth":17,"authority":406,"freshness":13,"relevant":21,"comment":521},"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。",[523],{"name":401,"url":516},[26,27,339,525,526],"作物病害识别","模型审计",{"code":361,"slug":414,"name":415,"path":416,"parent_code":417},[529],{"code":32,"slug":33,"name":34,"path":35,"parent_code":36},{"status":39,"locked":14,"version":40,"method":41,"evidence":531},{"candidates":532,"agricultural_evidence":47},[533,537,541],{"code":361,"score":46,"supported":47,"title":534,"abstract":535,"tags":536},[364],[339,364],[339],{"code":32,"score":13,"supported":47,"title":538,"abstract":539,"tags":540},[429],[],[525,27],{"code":45,"score":21,"supported":14,"title":542,"abstract":543,"tags":544},[],[],[26],[546,547],"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":549,"openalex_id":551,"authors":552,"venue":401,"cited_by_count":103,"oa_url":516,"card":559,"direction":235,"ingested_from":111},"W7223088030",[553,555,557],{"name":554,"orcid":9},"Abdi-Basid Idriss Farah",{"name":556,"orcid":9},"Amina Abdillahi Guedi",{"name":558,"orcid":9},"Moussa Omar Youssouf",{"tldr":560,"method":561,"finding":562,"direction":235,"opportunity":563},"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"]