[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3564":3,"related-3564":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":24,"tags":26,"search_phrases":32,"slug":35,"view_count":36,"doi":37,"paper":38,"created_at":63},3564,"Research Progress on Intelligent Seeding Technology and Equipment: The Development of Seeders from Multi-Functional Integration to Agricultural Intelligent Agents","https:\u002F\u002Fdoi.org\u002F10.3390\u002Fagronomy16191884","Seeding constitutes a key crop-production operation that governs seed spatial arrangement, crop population structure, and potential yield formation, and forms the foundation of precise, efficient, and eco-friendly farming. However, field soil properties, regional climate, and crop agronomic requirements exhibit strong spatio-temporal heterogeneity. Conventional seeding operations based on manual experience and fixed preset parameters cannot meet the demands of large-scale precision agriculture. Enabled by progress in precision agriculture, intelligent sensing, artificial intelligence, and autonomous machinery, modern intelligent seeding systems integrate precision seed metering, high-precision environmental perception, and closed-loop dynamic self-regulation. Such systems can improve plant-spacing uniformity and enable precise seeding-depth control under standard open-field conditions, yet face noticeable performance limitations in GNSS-denied complex environments including dense crop canopies and greenhouses. This review outlines the evolutionary trajectory of seeding machinery and summarizes research progress regarding precision seeding, multi-functional equipment integration, multi-source information perception, and intelligent decision-making. Integrated design principles covering mechanical optimization, electronic control, and perception-driven decision systems are elaborated. Four developmental phases of seeding equipment are identified: mechanical precision operation, electronic intelligent regulation, multi-functional module integration, and intelligent cognitive integration. Current intelligent seeding technologies are constrained by limited adaptability to complex farmland conditions, unstable multi-source data fusion, insufficient long-term operational reliability, and high deployment costs across diverse scenarios, restricting their broad field-scale adoption. Future research should combine agronomic knowledge with artificial intelligence to improve environmental awareness and autonomous decision-making capability, develop low-cost, high-reliability integrated seeding equipment, and support the construction of intelligent agricultural machinery systems.","播种是决定种子空间分布、作物群体结构和潜在产量形成的关键作物生产环节，也是精准、高效、绿色农业的基础。然而，田间土壤特性、区域气候和作物农艺要求具有强烈的时空异质性。基于人工经验和固定预设参数的传统播种作业无法满足大规模精准农业的需求。在精准农业、智能感知、人工智能和自主机械等领域的进步推动下，现代智能播种系统集成了精密排种、高精度环境感知和闭环动态自适应调节。此类系统可在标准露地条件下提高株距均匀性并实现精量播种深度控制，但在全球导航卫星系统（GNSS）拒止的复杂环境中，包括密植作物冠层和温室，仍面临明显的性能局限。本文综述了播种机械的演进轨迹，总结了精量播种、多功能装备集成、多源信息感知和智能决策方面的研究进展。阐述了涵盖机械优化、电子控制和感知驱动决策系统的集成设计原则。识别出播种装备的四个发展阶段：机械精量作业、电子智能调控、多功能模块集成和智能认知集成。当前智能播种技术受限于对复杂农田条件的适应性不足、多源数据融合不稳定、长期运行可靠性不够以及多场景部署成本高昂，制约了其在田间的广泛规模化应用。未来研究应将农艺知识与人工智能相结合，提升环境感知和自主决策能力，开发低成本、高可靠性的集成播种装备，支撑智能农机体系建设。",null,"Agronomy","2026-09-25T00:00:00Z","论文",10,true,82,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},22,21,18,13,8,1,"系统梳理智能播种装备从机械化到智能体四阶段演进，指出GNSS受限环境与多源数据融合瓶颈，对智慧农业装备研发有参考价值。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","智能农机","精准农业","智能播种",[33,34],"智能播种 装备","Agronomy 智能播种 装备","智能播种装备-3564",0,"10.3390\u002Fagronomy16191884",{"doi":37,"openalex_id":39,"authors":40,"venue":10,"cited_by_count":36,"oa_url":6,"card":55,"direction":61,"ingested_from":62},"W7214297818",[41,43,45,47,49,52],{"name":42,"orcid":9},"Yuting Dong",{"name":44,"orcid":9},"Yapeng Wu",{"name":46,"orcid":9},"Shiguo Wang",{"name":48,"orcid":9},"Xiaohu Guo",{"name":50,"orcid":51},"Xin Lu","https:\u002F\u002Forcid.org\u002F0000-0003-4462-3472",{"name":53,"orcid":54},"Zhong Tang","https:\u002F\u002Forcid.org\u002F0000-0002-2724-115X",{"tldr":56,"method":57,"finding":58,"direction":59,"opportunity":60},"综述智能播种技术装备从多功能集成到农业智能体的四阶段演进及瓶颈。","文献综述，梳理精量播种、多源感知与智能决策的集成设计。","智能播种在开阔农田表现良好，但复杂环境下适应性、数据融合与成本仍受限。","智慧农业 \u002F 农业物联网","GNSS拒止的冠层与温室环境下低成本高可靠感知与自主决策播种装备是研究空白。","农业人工智能与决策模型","openalex","2026-09-26T23:30:47.821171Z",{"total":65,"page":22,"page_size":65,"items":66},6,[67,110,135,186,205,239],{"id":68,"title":69,"url":70,"summary":71,"summary_zh":9,"content":9,"source_name":72,"source_url":70,"published_at":11,"category":12,"cover_url":9,"hotness":73,"is_selected":74,"score":75,"score_detail":76,"sources":79,"tags":84,"search_phrases":87,"slug":90,"view_count":36,"doi":91,"paper":92,"created_at":109},3565,"An Automata-Driven Cognitive Explainable Artificial Intelligence Framework for Climate-Adaptive Precision Agriculture and Environmental Sustainability","https:\u002F\u002Fdoi.org\u002F10.7759\u002Fs44389-026-00295-5","An Automata-Driven Cognitive Explainable Artificial Intelligence Framework for Climate-Adaptive Precision Agriculture and Environmental Sustainability。Cureus Journal of Computer Science.","Cureus Journal of Computer Science.",25,false,39,{"impact":21,"substance":65,"depth":13,"authority":65,"freshness":77,"relevant":22,"comment":78},9,"主题契合智慧农业与农业AI，但摘要仅重复标题、无方法与数据细节，信息增量有限，暂不建议进入每日精选。",[80,81],{"name":72,"url":70},{"name":82,"url":83},"Cureus Journal of Computer Science 2026-09-25","https:\u002F\u002Fwww.cureusjournals.com\u002Farticles\u002F20543",[27,28,85,30,86],"可解释AI","气候适应",[88,89],"气候适应 精准农业 可解释AI","农业人工智能 智慧农业 气候适应 精准农业","气候适应精准农业可解释AI-3565","10.7759\u002Fs44389-026-00295-5",{"doi":91,"openalex_id":93,"authors":94,"venue":72,"cited_by_count":36,"oa_url":70,"card":9,"direction":61,"ingested_from":62},"W7214363342",[95,98,101,103,105,107],{"name":96,"orcid":97},"Mritunjay Kr. Ranjan","https:\u002F\u002Forcid.org\u002F0000-0003-0240-4909",{"name":99,"orcid":100},"Rohit Gupta","https:\u002F\u002Forcid.org\u002F0000-0002-4436-8275",{"name":102,"orcid":9},"Nitin  D Mali",{"name":104,"orcid":9},"Ansh  A Rajore",{"name":106,"orcid":9},"Gaurav Narendra Patil",{"name":108,"orcid":9},"Ankita  N Patil","2026-09-26T23:30:47.893666Z",{"id":111,"title":112,"url":113,"summary":114,"summary_zh":9,"content":115,"source_name":116,"source_url":9,"published_at":117,"category":118,"cover_url":9,"hotness":13,"is_selected":74,"score":119,"score_detail":120,"sources":125,"tags":127,"search_phrases":130,"slug":133,"view_count":36,"doi":9,"paper":9,"created_at":134},3387,"从会种粮到'慧'种粮——中联重科岳麓智慧农场、常州艾宝机器人、温氏华系种猪、牧原智能猪舍AI兽医集体亮相","https:\u002F\u002Fnews.cyol.com\u002Fgb\u002Fkeji\u002Farticles\u002F2026-09\u002F23\u002Fcontent_v6vKznCP2x.html","中联重科岳麓智慧农场'天空地人机'协同智慧作业模式搭载多光谱相机巡田无人机最高飞120米、20分钟完成整片地块巡检，长势图精度达3米×3米网格；常州艾宝机器人四臂采摘猕猴桃机器人经第三方检测适合山坡丘陵地形，一台机器能节约7至8个人工；温氏股份、德康农牧'华系种猪育种技术与核心种源创制及应用'项目获国家科学技术进步奖一等奖；牧原股份智能化猪舍AI兽医全天候值守。","◎记者 刘立 王乔琪\n\n又是一年丰收季。田野之间，沉甸甸的果实背后，农业生产逻辑正在发生深刻变化。\n\n今年5月，国务院印发的《加快农业农村现代化“十五五”规划》明确提出：“推进人工智能运用和智慧农业发展。加快农业人工智能技术创新、应用场景拓展和产业生态营造。”\n\n近期，上海证券报记者在对部分农业种养殖龙头企业和农业展会的调研中发现，从种子实验室到果园大田，智能化工具正在把农业从经验驱动推向数据驱动。AI不再只是实验室概念，它正从育种、田间种植延伸到畜牧水产全链条，改写着传统农业依靠经验、看天吃饭的旧模式。\n\n**以AI打磨农业“芯片”**\n\n种业被称作农业的“芯片”。记者在采访中发现，快速发展的AI技术，正在缩短育种探索的漫长周期。\n\n在近日举办的全国种子双交会上，中种集团展台上一台ELISA蛋白全自动分析检测工作站设备吸引不少参会者驻足。据工作人员介绍，传统育种蛋白检测多采用人工、半自动化操作，通量低、稳定性差，而这台设备针对生物育种性状鉴定、蛋白定量筛查场景，可实现实验全流程无人自动化运行，日检测通量比人工高5倍以上，成本降低50%，且完全国产化。\n\n记者在采访中发现，AI正在把复杂的农业科研数据转化成通俗直观的结论。育种不再完全依靠科研人员肉眼观察记录，而是由仪器采集数据，AI完成解析筛选。\n\n上海泽泉科技股份有限公司研发的花粉活力仪可以应用于玉米、小麦、果树、大豆杂交育种项目，帮助科研人员精准把握授粉窗口，稳定后代品质。公司副总经理张凯峰对记者说：“过去做大豆、玉米、小麦等杂交育种，收集花粉之后，不知道花粉活性。我们的设备可以对每一粒花粉活性做量化测定，活性不达标的花粉直接淘汰。随着科技的普及，我相信，未来这些智能设备都会逐步下沉到普通农户。”\n\n一位农业专家对记者表示，AI并非颠覆传统育种，而是对育种体系的升级赋能。智能育种的核心，是依靠高通量设备获取标准化的基因组、表型组数据，依托算法挖掘基因与性状之间的关联，推动育种从过去依赖经验的“试错筛选”，转向数据驱动的“精准设计”，助力我国种业向育种4.0阶段迈进。\n\n**“空天地一体化”重构农业生产模式**\n\n在位于湖南长沙的中联重科岳麓智慧农场，中联智慧农业打造的“天空地人机”协同智慧作业模式，正在改写大田管护方式。\n\n以往人工巡查四五百亩稻田，要耗费一至两小时。中联智慧农业这套搭载多光谱相机的巡田无人机最高可飞120米，20分钟就能完成整片地块巡检，变焦可达100多倍，捕捉到虫害、长势异常等疑点便低空补拍核验。影像实时回传云端后，AI算法输出水稻长势图，精度可达3米乘以3米网格，明确标注地块补水、补肥位置与用量。\n\n天上完成巡查，地上设备随即落地执行。中联智慧农业部署的物联网灌排设备、微型气象站同步采集风、雨、光照数据，农户通过手机软件就能查看农事日历、操控田间设备。这种“按数据决策”的模式，改变了传统农业“凭经验跑田”的粗放管理方式。\n\n除了天上飞的无人机，地上走的机器人也开始应用到更多农业生产场景。\n\n在北京平谷的2026世界农业科技创新大会上，常州艾宝机器人有限公司研发的猕猴桃采摘机器人吸引了不少观众。公司工程师赵炳宇向记者介绍，这款四臂采摘机器人，经第三方机构检测，适合山坡丘陵地形，一台机器能够节约7至8个人工。现阶段设备尚处在定制阶段，已有3家企业完成定制采购。\n\n“我们主要从事AI智能水果采摘机器人研发，已突破多臂协同、视觉感知、自主导航等关键技术，已经推出苹果、葡萄、猕猴桃采摘等系列产品，目的就是推动采摘机器人从技术示范走向生产落地。”赵炳宇说。\n\n**农牧业智能化驶入快车道**\n\n中国是全球最大的猪肉消费国，年消费猪肉超过5000万吨。庞大的产业规模背后，是管理规模大、场景复杂的生猪养殖体系。如今，人工智能正加速渗透进这个最传统的行业。\n\n畜牧种业是畜牧业的“芯片”，在很长一段时间，国内要从海外进口种猪，还要高价采购海外生猪基因芯片，支付高昂的知识产权费。通常一头外引种猪的综合成本高达3万元至5万元。\n\n这一局面如今已被打破。日前，温氏股份、德康农牧等作为主要完成单位参与的“华系种猪育种技术与核心种源创制及应用”项目，荣获国家科学技术进步奖一等奖。该项目通过基因组选择、表型精准测定等技术创新，构建了华系种猪自主育种体系，创制了具有自主知识产权的核心种猪群体，标志着我国生猪核心种源实现高标准自主可控，彻底打破国外长期垄断格局。\n\n在河南，畜牧业龙头企业牧原股份的智能化猪舍里，AI兽医已投入全天候值守。\n\n“过去发现猪不吃食，饲养员找兽医、翻手册、凭经验判断，平均响应时间需要个把小时；现在从发现异常到推荐用药的闭环，仅需数秒。”牧原股份技术负责人介绍，名为“小牧助手”的AI应用，融合企业30余年养殖知识与全产业链330万套智能装备每日产生的20亿条数据，被封装为可嵌入日常流程的AI智能体。目前，牧原与阿里云合作共建的养猪大模型“猪小牧”已向全行业开放共享，助力行业高质量发展。\n\n此外，类似变革正在更多养殖场景中铺开。比如在水产养殖领域，广西防城港东兴市红树林农业有限公司的养殖基地引入AI“厂长”，集成AI算法、高清水下摄像头与传感器网络，24小时实时采集虾群数据，自动调控水阀与投料，每日精准投喂300余次。目前，该基地一年可养对虾5茬，年产量35万至40万公斤。\n\n【责任编辑：母建鑫】","中青在线 2026-09-23","2026-09-23T00:00:00Z","报道",85,{"impact":121,"substance":17,"depth":19,"authority":122,"freshness":123,"relevant":22,"comment":124},26,12,7,"紧扣十五五规划AI+农业主线，覆盖育种、种植、养殖全链条，案例与数据扎实，具行业风向标价值。",[126],{"name":116,"url":113},[27,28,29,128,129],"种业振兴","智慧养殖",[131,132],"中联重科 岳麓智慧农场","温氏 华系种猪","中联重科岳麓智慧农场-3387","2026-09-25T00:09:27.645047Z",{"id":136,"title":137,"url":138,"summary":139,"summary_zh":140,"content":9,"source_name":141,"source_url":138,"published_at":117,"category":12,"cover_url":9,"hotness":13,"is_selected":74,"score":142,"score_detail":143,"sources":148,"tags":150,"search_phrases":153,"slug":156,"view_count":36,"doi":157,"paper":158,"created_at":185},3358,"One Toolchain, Six Domains: A Multiple-Case, Document-Based Study of Rapid IoT Prototypes Built in a One-Week Immersive Course on a Master’s Program in Applied Artificial Intelligence","https:\u002F\u002Fdoi.org\u002F10.20944\u002Fpreprints202609.2011.v1","This paper reports a document-based, multiple-case study of six Internet-of-Things (IoT) prototypes designed and simulated during a one-week immersive course, “IoT for Data Intelligence,” delivered in July 2026 within the professional Master in Applied Artificial Intelligence (Maestría en Inteligencia Artificial Aplicada, MNA) at Tecnológico de Monterrey. Six teams followed the same five-day toolchain IoT theory; Oracle Application Express (APEX), SQL, and REST service design; MIT App Inventor; ESP32\u002FWokwi simulation; and generative-AI integration and produced Wokwi-simulated prototypes spanning industrial energy monitoring, agricultural hazard response, residential automation, cardiovascular telemonitoring, industrial waste reduction, and precision agriculture. A fixed coding framework was applied across architecture, AI-integration pattern, platform-level failure modes, security debt, and Sustainable Development Goal alignment, distinguishing findings that the course structure itself prescribes from findings the teams introduced independently. The six cases converged on a shared five-layer architecture and, in a pattern only partly prescribed by the course, on keeping generative AI in an advisory or fail-safe-wrapped role. Deposited results were also compared, for illustrative purposes only, against the course’s internal competency rubric. An observed proposal from a Pontifical Catholic University of Chile’s collaboration is discussed as an informal reference point rather than as evidence for generalization. This paper discusses the implications and limits of this small, single-institution, single-cohort, simulation-only case set.","本文报告了一项基于文档的多案例研究，研究对象为六项物联网（Internet of Things, IoT）原型，这些原型是在2026年7月于蒙特雷理工学院（Tecnológico de Monterrey）应用人工智能专业硕士（Maestría en Inteligencia Artificial Aplicada, MNA）项目内开设的一周沉浸式课程“面向数据智能的物联网”（IoT for Data Intelligence）中设计与仿真的。六个团队遵循了相同的五日工具链——物联网理论；Oracle Application Express（APEX）、SQL与REST服务设计；MIT App Inventor；ESP32\u002FWokwi仿真；以及生成式AI集成——并产出了基于Wokwi仿真的原型，涵盖工业能源监测、农业灾害响应、住宅自动化、心血管远程监护、工业减废和精准农业。研究采用固定编码框架，从架构、AI集成模式、平台级失效模式、安全债务和可持续发展目标对齐五个维度进行分析，并区分了课程结构本身所规定的发现与各团队独立引入的发现。六个案例收敛于一个共享的五层架构，并在一种仅部分由课程规定的模式中，将生成式AI保持在顾问性或故障安全包裹的角色中。所提交的成果还仅出于示例目的与课程内部能力量规进行了比较。智利天主教大学一项合作中提出的方案作为非正式参照点加以讨论，而非作为可推广的证据。本文讨论了这一小型、单一机构、单一批次、仅仿真案例集的启示与局限。","Preprints.org",50,{"impact":65,"substance":144,"depth":145,"authority":146,"freshness":77,"relevant":22,"comment":147},16,15,4,"单校单期小样本的预印本教学案例研究，含农业物联网原型与生成式AI集成经验，但样本与仿真局限明显，公共价值有限。",[149],{"name":141,"url":138},[27,28,151,30,152],"农业物联网","农业教育",[154,155],"Tecnológico de Monterrey 物联网 课程","ESP32 Wokwi 农业物联网 原型","TecnológicodeMonterrey物联网课程-3358","10.20944\u002Fpreprints202609.2011.v1",{"doi":157,"openalex_id":159,"authors":160,"venue":141,"cited_by_count":36,"oa_url":138,"card":179,"direction":59,"ingested_from":62},"W7214071608",[161,164,167,170,173,176],{"name":162,"orcid":163},"Antonio Carlos Bento","https:\u002F\u002Forcid.org\u002F0000-0001-8264-4771",{"name":165,"orcid":166},"Alexandro Ortiz","https:\u002F\u002Forcid.org\u002F0000-0002-3945-6908",{"name":168,"orcid":169},"Grettel Barceló-Alonso","https:\u002F\u002Forcid.org\u002F0009-0004-3373-6441",{"name":171,"orcid":172},"Jose Reinaldo Silva","https:\u002F\u002Forcid.org\u002F0000-0003-2796-1613",{"name":174,"orcid":175},"Luis E. Falcón-Morales","https:\u002F\u002Forcid.org\u002F0000-0001-8760-5640",{"name":177,"orcid":178},"Sérgio Camacho-León","https:\u002F\u002Forcid.org\u002F0000-0002-5996-9997",{"tldr":180,"method":181,"finding":182,"direction":183,"opportunity":184},"基于六组一周IoT课程原型文档，分析其架构、AI集成与安全模式。","文档多案例研究，固定编码框架，Wokwi仿真与生成式AI集成。","六案例收敛于五层架构，生成式AI多限于建议或故障保护角色。","其他","可探究仿真原型向真实农田部署时，安全债务与AI角色如何演变。","2026-09-24T23:30:13.353443Z",{"id":187,"title":188,"url":189,"summary":190,"summary_zh":9,"content":9,"source_name":191,"source_url":9,"published_at":192,"category":118,"cover_url":9,"hotness":13,"is_selected":74,"score":193,"score_detail":194,"sources":197,"tags":199,"search_phrases":201,"slug":203,"view_count":36,"doi":9,"paper":9,"created_at":204},3288,"会种粮到慧种粮、会养殖到慧养殖——AI驱动农业转型升级 中联重科岳麓智慧农场天空地人机协同模式","https:\u002F\u002Ffinance.china.com.cn\u002Fnews\u002F20260923\u002F6324918.shtml","9月23日中国网财经报道：在湖南长沙中联重科岳麓智慧农场，中联智慧农业打造的天空地人机协同智慧作业模式正在改写大田管护方式：以往人工巡查四五百亩稻田要耗费1-2小时；巡田无人机最高可飞120米，20分钟就能完成整片地块巡检，变焦可达100多倍；影像实时回传云端后AI算法输出水稻长势图，精度可达3米乘3米网格。常州艾宝机器人公司研发的猕猴桃采摘机器人已突破多臂协同、视觉感知、自主导航等关键技术；日前温氏股份、德康农牧作为主要完成单位参与的华系种猪育种技术与核心种源创制及应用项目荣获国家科学技术进步奖一等奖。","中国网财经 2026年09月23日","2026-09-23T03:00:00Z",71,{"impact":17,"substance":144,"depth":195,"authority":77,"freshness":13,"relevant":22,"comment":196},14,"报道呈现AI巡田、采摘机器人与种猪育种获奖等实质进展，但属企业宣传性综合报道，深度与权威度中等。",[198],{"name":191,"url":189},[27,28,29,128,200],"天空地一体化",[131,202],"猕猴桃采摘机器人 常州艾宝","中联重科岳麓智慧农场-3288","2026-09-24T00:03:56.850138Z",{"id":206,"title":207,"url":208,"summary":209,"summary_zh":210,"content":9,"source_name":211,"source_url":208,"published_at":212,"category":12,"cover_url":9,"hotness":73,"is_selected":74,"score":193,"score_detail":213,"sources":216,"tags":220,"search_phrases":223,"slug":226,"view_count":36,"doi":227,"paper":228,"created_at":238},3282,"Design and Implementation of an Ensemble Learning Based Decision Support Model for Crop Selection in Precision Agriculture","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22895811","Soil composition governs which crop can be grown profitably in a given field, and the relationship between soil variables and crop suitability is nonlinear, interacting and therefore poorly served by heuristic rules. This paper presents SCR-XGB, a five-layer framework that couples a disciplined data-conditioning stage with a regularised gradient-boosted tree ensemble for crop recommendation from soil and climatic parameters. The acquisition layer collects nitrogen, phosphorus and potassium concentration together with temperature, humidity, soil pH and rainfall; the conditioning layer imputes missing values and removes outliers by an interquartile filter; the feature engineering layer derives nutrient ratios, normalises and standardises the numeric fields and encodes the crop label; the ensemble layer fits an additive sequence of regression trees under a regularised objective with shrinkage and column subsampling; and the recommendation layer issues a ranked crop list with per-crop confidence. Four algorithms are specified in full, covering conditioning, feature construction, boosted training and inference, and a complexity analysis is given for each stage. Evaluated on a public corpus of soil and climate records against five baseline learners trained over the identical feature matrix, the proposed framework attains 99.31% accuracy, 100% precision, 99% recall and an F1-score of 99%, ahead of naive Bayes and random forest at 99.09%, support vector machine at 97.95%, logistic regression at 95.22% and a single decision tree at 90.00%. The 9.31 percentage point margin over the single tree, set against the 0.22 point margin over the strongest baseline, quantifies the benefit of boosting and shows where the remaining headroom on this task actually lies.","土壤组成决定了特定田块适宜种植何种作物才能获得经济效益，而土壤变量与作物适宜性之间的关系是非线性的、相互作用的，因此启发式规则难以有效处理这一问题。本文提出SCR-XGB，一个五层框架，将规范化的数据调理阶段与正则化梯度提升树集成相结合，用于基于土壤和气候参数的作物推荐。采集层收集氮、磷、钾浓度以及温度、湿度、土壤pH值和降雨量；调理层通过四分位距滤波器插补缺失值并剔除异常值；特征工程层推导养分比率，对数值字段进行归一化和标准化，并对作物标签进行编码；集成层在带有收缩和列子采样的正则化目标函数下拟合加性回归树序列；推荐层输出带有每种作物置信度的排序作物列表。本文完整给出了四种算法，涵盖调理、特征构建、提升训练和推理，并对每个阶段进行了复杂度分析。在公开的土壤和气候记录语料库上，与在相同特征矩阵上训练的五个基线学习器进行对比评估，所提框架达到了99.31%的准确率、100%的精确率、99%的召回率和99%的F1分数，优于朴素贝叶斯和随机森林的99.09%、支持向量机的97.95%、逻辑回归的95.22%以及单棵决策树的90.00%。相较于单棵决策树9.31个百分点的优势，与相较于最强基线0.22个百分点的优势相比，量化了提升方法的收益，并揭示了该任务上剩余提升空间的实际所在。","Zenodo (CERN European Organization for Nuclear Research)","2026-09-22T00:00:00Z",{"impact":122,"substance":18,"depth":214,"authority":122,"freshness":77,"relevant":22,"comment":215},17,"方法完整、对比基线充分，但属常规机器学习应用论文，公共价值有限，可作主题聚合素材而非每日精选。",[217,218],{"name":211,"url":208},{"name":211,"url":219},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22895812",[27,28,30,221,222],"作物推荐","土壤数据",[224,225],"SCR-XGB 作物推荐 土壤","集成学习 精准农业 选种","SCR-XGB作物推荐土壤-3282","10.5281\u002Fzenodo.22895811",{"doi":227,"openalex_id":229,"authors":230,"venue":211,"cited_by_count":36,"oa_url":208,"card":233,"direction":61,"ingested_from":62},"W7214043002",[231],{"name":232,"orcid":9},"Prof. Nagendra Patel Sahil Verma",{"tldr":234,"method":235,"finding":236,"direction":61,"opportunity":237},"提出SCR-XGB五层框架，用梯度提升树集成从土壤和气候参数推荐作物。","基于土壤气候数据，采用正则化梯度提升树集成，含缺失值插补、异常值过滤和特征工程。","模型准确率达99.31%，优于朴素贝叶斯、随机森林等基线，比单决策树提升9.31个百分点。","可探索将模型部署到田间实时决策，并融合遥感与物联网数据提升泛化能力。","2026-09-23T23:30:35.342497Z",{"id":240,"title":241,"url":242,"summary":243,"summary_zh":244,"content":9,"source_name":211,"source_url":242,"published_at":212,"category":12,"cover_url":9,"hotness":73,"is_selected":74,"score":245,"score_detail":246,"sources":248,"tags":252,"search_phrases":254,"slug":257,"view_count":36,"doi":258,"paper":259,"created_at":268},3271,"A Systematic Study of Supervised and Ensemble Learning Approaches for Crop Selection in Smart Agriculture","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22896343","Selecting the crop best matched to a field’s soil and climate is one of the highest-leverage decisions in agriculture, and one that farmers have traditionally made by intuition or inherited practice. Soil pH together with nitrogen, phosphorus and potassium concentration, and the local temperature, humidity and rainfall regime, jointly determine which crop will flourish and which will fail, and the relationship between those variables and crop performance is neither linear nor independent. Machine learning has therefore become the dominant approach to automated crop recommendation. This paper reviews the field across four technique families — classical supervised learning, ensemble and boosting methods, deep learning and metaheuristic hybrids, and IoT and deployment-oriented systems — and compares twenty-four representative studies published between 2016 and 2026 in terms of method, data source, reported accuracy, advantage and limitation. A generic seven-stage recommendation pipeline is presented and each family is situated within it. The comparison shows that reported accuracy on the standard nutrient-and-climate benchmark has converged in a narrow band between roughly 98 and 99.5 per cent, that boosting and ensemble methods occupy the upper part of that band, and that further gains on the benchmark are no longer the binding constraint on the field. The gaps that remain open are instead the absence of socio-economic and market variables from the decision, the lack of region-specific and long-horizon environmental validation, dataset narrowness and geographic bias, limited interpretability, and the accessibility of these systems to small and resource-poor farmers. These are consolidated into a set of research directions for future work.","选择与田块土壤和气候最匹配的作物是农业中杠杆效应最高的决策之一，而农民传统上依靠直觉或世代相传的经验来做出这一决策。土壤pH值以及氮、磷、钾浓度，加上当地的气温、湿度和降雨状况，共同决定了哪种作物能够茁壮成长、哪种会歉收，而这些变量与作物表现之间的关系既非线性也非相互独立。因此，机器学习已成为自动化作物推荐的主流方法。本文从四个技术族系——经典监督学习、集成与提升方法、深度学习与元启发式混合方法，以及物联网与面向部署的系统——对该领域进行了综述，并从方法、数据来源、报告精度、优势和局限性方面比较了2016年至2026年间发表的二十四项代表性研究。本文提出了一个通用的七阶段推荐流程，并将每个技术族系置于该流程中加以定位。比较结果表明，在标准养分与气候基准上的报告精度已收敛于约98%至99.5%的狭窄区间内，提升与集成方法占据该区间的上端，而在该基准上进一步提升已不再是该领域的约束瓶颈。真正尚未填补的空白在于：决策中缺乏社会经济和市场变量，缺少针对特定区域和长期环境验证，数据集狭窄且存在地理偏差，可解释性有限，以及这些系统对小型和资源匮乏农户的可及性不足。这些空白被归纳为未来工作的一系列研究方向。",78,{"impact":144,"substance":17,"depth":19,"authority":20,"freshness":77,"relevant":22,"comment":247},"系统综述24项研究并指出基准精度已趋饱和，真正瓶颈转向社会经济变量与可解释性，对智慧农业选种方向有参考价值。",[249,250],{"name":211,"url":242},{"name":211,"url":251},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22896344",[27,28,253,30,221],"机器学习",[255,256],"农业人工智能 作物推荐 智慧农业 机器学习","农业人工智能 作物推荐","农业人工智能作物推荐智慧农业机器学习-3271","10.5281\u002Fzenodo.22896343",{"doi":258,"openalex_id":260,"authors":261,"venue":211,"cited_by_count":36,"oa_url":242,"card":263,"direction":59,"ingested_from":62},"W7214002748",[262],{"name":232,"orcid":9},{"tldr":264,"method":265,"finding":266,"direction":61,"opportunity":267},"系统综述2016-2026年24项作物推荐研究，比较四类机器学习方法并指出基准精度已趋饱和。","综述监督学习、集成提升、深度学习与元启发式、物联网部署四类方法及七阶段流程。","标准基准精度收敛于98%-99.5%，提升集成法最优，但精度已非领域瓶颈。","将社会经济与市场变量、区域长期环境验证及可解释性纳入作物推荐，服务小农户。","2026-09-23T23:30:09.369987Z"]