[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3631":3,"related-3631":38},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":8,"published_at":11,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":24,"tags":26,"search_phrases":32,"slug":35,"view_count":36,"doi":8,"paper":8,"created_at":37},3631,"从自动驾驶农机到直播带川货——四川丰收节现场看农业'新变化'，资阳乐至40台丘区适用农机集中亮相","https:\u002F\u002Fwww.toutiao.com\u002Farticle\u002F7688731375461089792\u002F","9月23日，2026年中国农民丰收节四川省庆丰收主场活动在资阳市乐至县劳动镇庙山村举行。40台国内先进、丘区适用农机集中亮相，自动驾驶高速插秧机、自走式玉米收获机、植保无人机等装备直接参与作业。智慧农业展示区27个省内外智慧农业创新成果集中展示，从'天工开悟'农业智能体平台到空地协同巡田机器人系统，农业生产感知、决策、作业等环节被纳入智能化体系。'天府粮仓'金秋消费季和农产品电商直播展示区21个市州特色农产品集中亮相。",null,"## 从自动驾驶农机到直播带川货 四川丰收节现场看农业“新变化”\n\n2026-09-23 22:04·[川观新闻](https:\u002F\u002Fwww.toutiao.com\u002Fc\u002Fuser\u002Ftoken\u002FMS4wLjABAAAATWhymKV2Nylz8CurTDi-JElDP9oQyPA3W8UYZjI3sXA\u002F?source=tuwen_detail)\n\n[Video 3](https:\u002F\u002Fv6-web.toutiaovod.com\u002Fcd96ae7433f1fe200b7977133b4fda24\u002F6ab9bcf6\u002Fvideo\u002Ftos\u002Fcn\u002Ftos-cn-ve-31\u002FogIAUyqSQ39S2IfBa9Am5IE1zg9NsoZBfqRqLj\u002F?a=24&ch=0&cr=0&dr=0&er=0&cd=0%7C0%7C0%7C0&br=1608&bt=1608&cs=0&ds=3&ft=hGkwBy6LRR0s~hC32D12Nc.xBiGNbLPj8bRU_4TjILJJNv7TGW&mime_type=video_mp4&qs=0&rc=OGc0ZDM5OGU3aDRnNmk2PEBpajw7dDtvOjV3ZDMzNGkzM0BeMDZfXjVfNTQxYl8tNTQ0YSNjbTZzYWEvbW1hLS1kLmFzcw%3D%3D&btag=c0000e00018000&dy_q=1790553787&feature_id=5ecd8ea3ca9c3306454bb2c337ee3553&l=20260928080307AD2B8755F29209F90B2C)\n不支持的音频\u002F视频格式 请试试 刷新\n\n播放\n\n00:00\u002F00:00\n\n00:00\n\n进入全屏\n\n50\n\n川观新闻记者 昙昊\n\n9月23日，2026年中国农民丰收节四川省庆丰收主场活动在资阳市乐至县劳动镇庙山村举行。与其说这里是一场丰收的庆典，不如说是一场四川农业“新家底”的集中展示——农机下田作业、机器人巡田、直播间卖川货，传统乡土技艺与现代农业科技同场亮相。\n\n![Image 1](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-axegupay5k\u002F6e8f309b99ab4d0c9fae360413b13c7e~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=ilXo7%2BLYsOQY0u0%2FPERX0GxHMqo%3D)\n主舞台开场歌舞环节，人形机器人与歌手、舞蹈演员同台表演。\n\n![Image 2](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-tjoges91tu\u002F0e081707092d9a19654696a349994157~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=Tdjkmzo37bHzocBpjYmA9GEe5qI%3D)\n主会场观众席内，观众观演热情高涨。\n\n“怎么种田”，现场先给出了科技答案。在农机作业展示区，40台国内先进、丘区适用农机集中亮相，自动驾驶高速插秧机、自走式玉米收获机、植保无人机等装备直接参与作业。过去需要人工完成的耕种、收获等环节，如今越来越多交给机器完成。尤其针对四川丘陵地区地块分散、道路条件复杂等特点，一批“小而灵活”的农机装备也成为现场看点。\n\n![Image 3](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-tjoges91tu\u002F41927b9f7330d129be6aaa3bb75ca490~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=ysjMLr%2BK5zjVP%2FIe3wYrBy0CuXk%3D)\n智慧农业展示区内，大型农机正在演示水稻收割作业。\n\n![Image 4](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-tjoges91tu\u002F9ef4baf8665bb3ec03f04258a07c7a4c~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=XgtHJ6OIOeL%2FmFZcZ9V024c9QgQ%3D)\n小型农机展示区内陈列的特色农机——柑橘采摘机器人。\n\n“怎么管田”，则更多交给智慧农业。在智慧农业展示区，27个省内外智慧农业创新成果集中展示，从“天工开悟”农业智能体平台，到空地协同巡田机器人系统，农业生产中的感知、决策、作业等环节被纳入智能化体系。\n\n丰收之后，还要“卖得好”。活动现场设置“天府粮仓”金秋消费季和农产品电商直播展示区，21个市州特色农产品集中亮相，“一村六员一主播”代表走进直播间，为川字号农产品吆喝。\n\n![Image 5](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-tjoges91tu\u002Fe37c18c6052f722853560e22ee94ffdc~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=mTzpe6XiR3rLeJuVXu4w4EmNjbA%3D)\n农产品电商直播展示区内，带货主播正在售卖特色川货。\n\n而在另一侧，22项非遗技艺通过“画、织、装、剪、品”五艺绘丰收，把乡土记忆搬进庆丰收现场。\n\n![Image 6](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-tjoges91tu\u002Fff4b39cbde88516442a68cabe494e74e~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=CgkBaZvufxznQrcCB7RcCxEFgew%3D)\n主会场上演了丰富的庆丰收歌舞表演节目。\n\n![Image 7](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-tjoges91tu\u002Fe3428ed0baea69ba9f87bb1e9c3adb65~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=dCuqeQonUQ3xXEhu5hk7nXXEhnI%3D)\n“庆丰收”舞台上，舞龙表演团队正在入场。\n\n![Image 8](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-tjoges91tu\u002F4c433e6e1ca19871151cdc6d2a29a633~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=SOAF13O049PzGYhUx8de5%2FKHEZE%3D)\n非遗展示区内，参加丰收节的观众正在品尝特色农产品。\n\n![Image 9](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-tjoges91tu\u002F4cae36fdb4a0a4101389f5ef07f7854f~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=%2FlYgHzAKVfnzSoHMTtw%2BTS3hF2g%3D)\n参加丰收节的观众正在观看丰收题材摄影展。\n\n![Image 10](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-tjoges91tu\u002F391ed283270396ae18dbdb36036a3edd~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=J%2Fetz34jzLNRIDtwxz1O8XiiitQ%3D)\n非遗展示区内，参展业主正在制作安岳米卷。\n\n![Image 11](https:\u002F\u002Fp3-sign.toutiaoimg.com\u002Ftos-cn-i-tjoges91tu\u002Fb16a33718373e66d9ee6cd46773b0052~tplv-tt-origin-web:gif.jpeg?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1791158583&x-signature=WyE2MPLhE4e5u6h9ZfxtWN8qOm0%3D)\n主会场观众席座无虚席，场外还有大批村民围观表演。\n\n从机器下田，到智慧管田，再到直播卖货，今年四川丰收节展示的不只是“收成”，也是一幅正在发生变化的天府农业新图景。","川观新闻（今日头条转载）2026-09-23","2026-09-23T00:00:00Z","报道",10,false,56,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},16,15,12,9,4,1,"省级丰收节现场报道，集中展示40台丘区适用农机与27项智慧农业成果，有实质信息但时效偏旧、深度一般。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","丰收节","农产品直播","丘陵农机",[33,34],"四川丰收节 乐至 丘区农机","天工开悟 农业智能体","四川丰收节乐至丘区农机-3631",0,"2026-09-28T00:02:58.061609Z",{"total":39,"page":22,"page_size":39,"items":40},6,[41,77,108,135,158,179],{"id":42,"title":43,"url":44,"summary":45,"summary_zh":8,"content":8,"source_name":46,"source_url":8,"published_at":47,"category":48,"cover_url":8,"hotness":13,"is_selected":49,"score":50,"score_detail":51,"sources":57,"tags":59,"search_phrases":63,"slug":66,"view_count":36,"doi":8,"paper":67,"created_at":76},3653,"面向智慧农业的自主农业机械——观测、异质性与智能基础设施的集成框架","https:\u002F\u002Fwww.jstage.jst.go.jp\u002Fbrowse\u002Frdj\u002F5\u002F0\u002F_contents\u002F-char\u002Fja","由Hongjin Li、Chunjiang Gao等在Resources Data Journal 2026年第5卷发表的综述文章。面向Agriculture 4.0过渡，针对自主农业机械这一核心议题提出'观测-异质性-基础设施'三位一体的集成框架。系统综述了自主农业机械的技术基础、应用领域、性能优势与采纳约束，并围绕多模态感知、AI驱动决策、自主导航与控制、精准执行、多机协同的耦合展开分析，特别关注环境不确定性、实时决策、互操作性与系统级可扩展性的挑战。研究识别出从孤立任务自动化向数据驱动、适应性、网络化农业自主性的渐进转型，并强调单一组件的改进若无配套计算、通信、机械与制度基础设施支撑，未必带来系统级性能提升。","Resources Data Journal 2026 vol.5 p.576-596 (Hongjin Li, Chunjiang Gao)","2026-09-18T00:00:00Z","论文",true,74,{"impact":52,"substance":53,"depth":54,"authority":55,"freshness":39,"relevant":22,"comment":56},18,20,17,13,"核心期刊综述提出观测-异质性-基础设施三位一体框架，对自主农机系统级落地有参考价值，但属学术综述、时效一般，适合主题聚合而非每日精选头条。",[58],{"name":46,"url":44},[27,28,60,61,62],"多模态感知","自主农机","Agriculture 4.0",[64,65],"自主农业机械 集成框架","Resources Data Journal 智慧农业","自主农业机械集成框架-3653",{"doi":8,"openalex_id":8,"authors":68,"venue":8,"cited_by_count":36,"oa_url":8,"card":69,"direction":73,"ingested_from":75},[],{"tldr":70,"method":71,"finding":72,"direction":73,"opportunity":74},"综述自主农业机械，提出观测-异质性-基础设施三位一体集成框架。","文献综述，围绕多模态感知、AI决策、自主导航与多机协同分析。","单组件改进若无计算、通信与制度基础设施配套，难带来系统级性能提升。","智慧农业 \u002F 农业物联网","可研究异构农机互操作协议与边缘计算协同，量化基础设施配套对系统级性能的增益。","agent","2026-09-28T00:03:02.555619Z",{"id":78,"title":79,"url":80,"summary":81,"summary_zh":8,"content":8,"source_name":82,"source_url":8,"published_at":83,"category":48,"cover_url":8,"hotness":13,"is_selected":49,"score":84,"score_detail":85,"sources":89,"tags":91,"search_phrases":95,"slug":98,"view_count":36,"doi":8,"paper":99,"created_at":107},3652,"[预印本] 智慧农业平台AI模块的边际绿色贡献模拟——两项蒙特卡洛实验证据","https:\u002F\u002Farxiv.org\u002Fhtml\u002F2609.06740v1","在前期平台级评估基础上，本研究将组件显性化并设计两项受控模拟实验。实验1沿'AI能力→农户行为→农化投入减少'链路建模，将农药\u002F化肥减量建模为可避免盲施份额、处方有效性、决策触达覆盖率与采纳率的乘积；对比经验推广模式与AI模式：AI模式下达成农药减量20%的概率从推广模式下的接近0上升至基线20.7%，并在诊断精度0.95、采纳率0.85下最高达49%；化肥减量15%的概率从接近0升至52.0%。实验2对比现行做法（P0）、IoT工程改造（P1）及P1+AI灌溉调度（P2）：灌溉水节水由7.8%（P0）升至11.0%（P1）与16.0%（P2），AI在工程之外再增加5.0个百分点；稻田CH4在AI调度下减幅达30.5%，碳强度降低27.9%。","arXiv 2609.06740v1 (2026-09)","2026-09-15T00:00:00Z",75,{"impact":52,"substance":86,"depth":52,"authority":20,"freshness":87,"relevant":22,"comment":88},22,8,"预印本以两项蒙特卡洛模拟量化AI模块在农化减量与稻田减排上的边际绿色贡献，方法新颖、数据具体，但尚未经同行评审，属细分领域前沿进展。",[90],{"name":82,"url":80},[27,28,92,93,94],"精准灌溉","稻田甲烷","化肥农药减量",[96,97],"AI灌溉调度 稻田甲烷 节水","智慧农业 农药化肥减量 蒙特卡洛","AI灌溉调度稻田甲烷节水-3652",{"doi":8,"openalex_id":8,"authors":100,"venue":8,"cited_by_count":36,"oa_url":8,"card":101,"direction":105,"ingested_from":75},[],{"tldr":102,"method":103,"finding":104,"direction":105,"opportunity":106},"通过两项蒙特卡洛实验模拟智慧农业平台AI模块对农药化肥减量和灌溉节水减碳的边际绿色贡献。","蒙特卡洛模拟，构建AI能力到农户行为再到农化投入减少的链路模型，对比经验推广与A","AI模式使农药减量20%概率从近0升至20.7%-49%，化肥减量15%概率升至52%，灌溉节水额外","农业绿色发展与碳","可进一步实证检验AI诊断精度与农户采纳率对减量减碳效果的交互影响及区域异质性。","2026-09-28T00:03:02.391014Z",{"id":109,"title":110,"url":111,"summary":112,"summary_zh":8,"content":8,"source_name":113,"source_url":8,"published_at":83,"category":48,"cover_url":8,"hotness":13,"is_selected":49,"score":114,"score_detail":115,"sources":117,"tags":119,"search_phrases":123,"slug":126,"view_count":36,"doi":8,"paper":127,"created_at":134},3651,"[预印本] 蒙特卡洛事前评估AI驱动智慧农业平台绿色效益——以海南热带农业为例","https:\u002F\u002Farxiv.org\u002Fhtml\u002F2609.06737v1","以整合大语言模型问答、多模态病虫害诊断、物联网土壤湿度传感、NASA GIBS卫星遥感与闭环田间记录系统的热带农业AI决策平台为对象，构建覆盖农药化肥生产、田间N2O、灌溉电力与稻田CH4的'摇篮到农场大门'农业碳核算模型。在三个海南代表性场景（芒果园、冬季蔬菜、稻\u002F南繁育种田，按面积40%：30%：30%加权）下，通过蒙特卡洛模拟参数不确定性传播：在完全采纳试点情景下，化学农药使用减少23.5%（90%置信区间15.0%–33.2%），化肥施用减少21.0%（13.8%–28.9%），灌溉水减少16.5%（10.9%–23.5%），碳强度降低21.5%（16.1%–27.2%）。达成化肥减量≥15%与碳强度明确下降的概率分别为90.6%与98.1%，而总节水≥20%的概率仅约20%，表明应采用场景化表述。","arXiv 2609.06737v1 (2026-09)",76,{"impact":52,"substance":86,"depth":52,"authority":19,"freshness":39,"relevant":22,"comment":116},"预印本以蒙特卡洛量化AI平台在海南三类场景的减药减肥与碳减排概率，方法新颖、数据具体，但尚未经同行评审，属细分领域进展。",[118],{"name":113,"url":111},[27,28,120,121,122],"农药减量","农业碳核算","海南热带农业",[124,125],"海南 热带农业 AI平台","蒙特卡洛 农业碳核算","海南热带农业AI平台-3651",{"doi":8,"openalex_id":8,"authors":128,"venue":8,"cited_by_count":36,"oa_url":8,"card":129,"direction":105,"ingested_from":75},[],{"tldr":130,"method":131,"finding":132,"direction":105,"opportunity":133},"用蒙特卡洛模拟评估AI智慧农业平台在海南热带农业中的绿色减排效益。","构建摇篮到农场大门碳核算模型，结合蒙特卡洛模拟参数不确定性。","完全采纳下碳强度降21.5%，化肥减量达标概率90.6%，但总节水≥20%概率仅20%。","可延伸研究AI平台在不同作物场景下的节水短板及多目标优化策略。","2026-09-28T00:03:02.301756Z",{"id":136,"title":137,"url":138,"summary":139,"summary_zh":8,"content":140,"source_name":141,"source_url":8,"published_at":142,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":143,"score_detail":144,"sources":147,"tags":149,"search_phrases":153,"slug":156,"view_count":36,"doi":8,"paper":8,"created_at":157},3639,"聚焦人工智能前沿 助力农业科学发现——中国农科院基础科学研究中心第六期科学会议在京召开","https:\u002F\u002Fcaas.cn\u002Fxwzx\u002Fnkyw\u002F46880aef2ea74d298a2e3cdc53802154.htm","9月22日，中国农业科学院基础科学研究中心第六期科学会议在北京召开。会议以'人工智能助力农业科学发现'重大科学问题为主题开展专题研讨，特邀北京科学智能研究院张林峰研究员作主旨报告。会议集聚北京大学、清华大学、哈尔滨工业大学、中国科学院等高水平院校的10位专家学者，聚焦人工智能驱动的科研范式变革，系统交流人工智能赋能科学发现领域前沿研究进展，共同探讨AI4S基础理论与实践路径。会议达成共识，推进人工智能与农业科学发现深度融合需要从数据、基础设施、基础模型、协同创新等多维度着手。","9月22日，基础科学研究中心第六期科学会议在北京召开。会议以“人工智能助力农业科学发现”重大科学问题为主题开展专题研讨，特邀北京科学智能研究院张林峰研究员作主旨报告。\n\n会议集聚北京大学、清华大学、哈尔滨工业大学、中国科学院等高水平院校的10位专家学者，聚焦人工智能驱动的科研范式变革，系统交流人工智能赋能科学发现领域前沿研究进展，共同探讨AI4S（AI for Science）基础理论与实践路径。\n\n会议达成共识，推进人工智能与农业科学发现深度融合，需要从数据、基础设施、基础模型、协同创新等多维度着手，重点围绕建设AI-Ready农业科学数据与高质量知识底座、构建面向科学发现的农业专业大模型、孵化农业领域科研智能体、建设农业智能实验室等发力，助力攻克农业科学难题。\n\n院科技局、信息所、农机化所负责同志，装备信息基础科学研究中心科研骨干代表等参加会议。（通讯员 张月遥）","中国农业科学院 2026-09-22","2026-09-22T00:00:00Z",69,{"impact":52,"substance":17,"depth":145,"authority":18,"freshness":39,"relevant":22,"comment":146},14,"国家级科研机构主办的AI4S专题研讨会，提出农业专业大模型、科研智能体等方向性共识，对智慧农业领域有参考价值，但属会议报道、无具体成果落地。",[148],{"name":141,"url":138},[27,28,150,151,152],"农业大模型","AI4S","科研智能体",[154,155],"中国农科院 基础科学研究中心 人工智能","AI4S 农业科学发现","中国农科院基础科学研究中心人工智能-3639","2026-09-28T00:03:00.214073Z",{"id":159,"title":160,"url":161,"summary":162,"summary_zh":8,"content":163,"source_name":164,"source_url":8,"published_at":142,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":165,"score_detail":166,"sources":168,"tags":170,"search_phrases":174,"slug":177,"view_count":36,"doi":8,"paper":8,"created_at":178},3633,"农业科技促丰收 应用场景展未来——天河区举办庆祝2026年中国农民丰收节暨现代农业领域应用场景发布会","https:\u002F\u002Fwww.nfncb.cn\u002Findex.php\u002Fyaowen\u002F59850.html","9月22日下午，天河区庆祝2026年中国农民丰收节暨现代农业领域应用场景发布会在广州（国际）科技成果转化天河基地举行。多所科研院所和农业科技企业围绕智慧农业、精准种养、农业智能装备、农业大数据、农业金融等领域集中发布创新成果和应用场景。天河区集聚华南农业大学、广东省农业科学院、中国科学院华南植物园、广东省现代农业技术推广中心等科研院校，在作物分子育种、智能农业装备、种质资源创新利用等前沿领域达到国际先进水平。","活动现场设置了优质特色农产品展区。参会嘉宾在签到之余，参观品鉴了来自天河区涉农企业生产的绿色、优质、特色农产品，现场气氛热烈。\n\n区领导在致辞中代表天河区向全区广大农民朋友和“三农”工作者致以节日问候。他指出，天河区作为广州创新强区、城市客厅和经济大区，拥有得天独厚的农业科创资源，集聚了华南农业大学、广东省农业科学院、中国科学院华南植物园、广东省现代农业技术推广中心、广东省现代农业装备研究院、广州国家现代农业产业科技创新中心等科研院校，在作物分子育种、智能农业装备、种质资源创新利用等前沿领域达到国际先进水平。当前，人工智能、大数据等新技术正深刻重塑农业生产方式，天河作为广州科技创新策源地，要把握机遇，推动农业科技与产业深度融合，让更多科技成果从实验室走向田间地头，从科研机构走入乡村基层。他希望通过“科技展示+产品品鉴”，让大家看到天河农业科技的硬实力，也品尝到丰收的好味道，助力农民增收、农业增效。","南方农村报 2026-09-22",48,{"impact":19,"substance":55,"depth":20,"authority":87,"freshness":39,"relevant":22,"comment":167},"地市级丰收节配套应用场景发布会，有科研机构与AI赋能农业的实质表述，但整体偏地方宣传通稿，时效已过一周。",[169],{"name":164,"url":161},[27,28,171,172,173],"农业科技成果转化","农民丰收节","广州天河",[175,176],"天河区 现代农业 应用场景发布会","华南农业大学 农业科技 成果转化","天河区现代农业应用场景发布会-3633","2026-09-28T00:02:58.341602Z",{"id":180,"title":181,"url":182,"summary":183,"summary_zh":184,"content":8,"source_name":185,"source_url":182,"published_at":186,"category":48,"cover_url":8,"hotness":13,"is_selected":14,"score":187,"score_detail":188,"sources":191,"tags":193,"search_phrases":197,"slug":200,"view_count":36,"doi":201,"paper":202,"created_at":217},3626,"A Hybrid Model for Crop Yield Prediction Using Recurrent Neural Networks and Explainable Artificial Intelligence","https:\u002F\u002Fdoi.org\u002F10.22194\u002Fjgias\u002F27.2061","Accurate maize yield prediction is essential for ensuring food security and supporting agricultural planning in Kenya. However, the changes in climate and severe weather are posing more challenges to the stability of yield and food security. While advanced machine learning models, such as Long Short–Term Memory (LSTM) networks, offer strong predictive performance, their inherent \"black–box\" nature limits their practical adoption by farmers and policymakers who require interpretable and trustworthy insights. This study developed a transparent predictive model by integrating LSTM with Explainable Artificial Intelligence (XAI). Using county–level maize yield data (2012–2023), the hybrid model was compared against Random Forest and Gradient Boosting baselines. The LSTM model achieved superior performance (R² = 0.8551, RMSE = 0.3715, MAE = 0.2705), compared to Random Forest (R2 = 0.8441, RMSE= 0.3851 and MAE= 0.2856) and Gradient Boosting (R2 = 0.7925, RMSE = 0.4443 and MAE = 0.3234) baselines. XAI analysis, using SHapley Additive explanations (SHAP) and Local Interpretable Model–agnostic Explanations (LIME), identified longitude, latitude, and annual rainfall as key predictors. The model maintained high accuracy while improving interpretability, increasing transparency and trust. This provides actionable insights for farmers and policymakers, supporting evidence–based planning and enhancing resilience in smallholder systems. Keywords: Crop yield prediction, long short–term memory (LSTM), explainable artificial intelligence, precision agriculture, machine learning.","准确的玉米产量预测对于保障肯尼亚粮食安全和支持农业规划至关重要。然而，气候变化和极端天气对产量稳定性和粮食安全构成了更多挑战。尽管长短期记忆（LSTM）网络等先进机器学习模型具有强大的预测性能，但其固有的“黑箱”特性限制了农民和政策制定者的实际采用，因为他们需要可解释且可信的洞见。本研究通过将LSTM与可解释人工智能（XAI）相结合，开发了一个透明的预测模型。利用县级玉米产量数据（2012—2023年），将该混合模型与随机森林和梯度提升基线模型进行了比较。LSTM模型取得了更优的性能（R² = 0.8551，RMSE = 0.3715，MAE = 0.2705），优于随机森林（R2 = 0.8441，RMSE= 0.3851，MAE= 0.2856）和梯度提升（R2 = 0.7925，RMSE = 0.4443，MAE = 0.3234）基线模型。使用SHapley加法解释（SHAP）和局部可解释模型无关解释（LIME）进行的XAI分析，确定了经度、纬度和年降雨量是关键预测因子。该模型在保持高精度的同时提高了可解释性，增强了透明度和信任。这为农民和政策制定者提供了可操作的洞见，支持基于证据的规划，并增强小农系统的韧性。关键词：作物产量预测，长短期记忆（LSTM），可解释人工智能，精准农业，机器学习。","Journal of Global Innovations in Agricultural Sciences","2026-09-25T00:00:00Z",71,{"impact":19,"substance":189,"depth":54,"authority":55,"freshness":87,"relevant":22,"comment":190},21,"将LSTM与可解释AI结合用于肯尼亚县级玉米产量预测，方法新颖、指标详实，对智慧农业有参考价值，但属境外区域研究，公共影响有限。",[192],{"name":185,"url":182},[27,28,194,195,196],"机器学习","可解释AI","玉米产量预测",[198,199],"LSTM 玉米产量预测","SHAP LIME 农业模型","LSTM玉米产量预测-3626","10.22194\u002Fjgias\u002F27.2061",{"doi":201,"openalex_id":203,"authors":204,"venue":185,"cited_by_count":36,"oa_url":182,"card":210,"direction":214,"ingested_from":216},"W7214446524",[205,207],{"name":206,"orcid":8},"Stephen Gitau Ndung’u",{"name":208,"orcid":209},"Consolata Gakii","https:\u002F\u002Forcid.org\u002F0000-0003-2783-9992",{"tldr":211,"method":212,"finding":213,"direction":214,"opportunity":215},"用LSTM结合可解释AI预测肯尼亚玉米产量，兼顾精度与透明度。","基于2012-2023县级玉米产量数据，LSTM融合SHAP与LIME，对比随机","LSTM精度最高（R²=0.8551），经度、纬度和年降雨量是主要预测因子。","农业人工智能与决策模型","可探索将可解释AI与遥感、气象多源数据融合，提升小农户区域产量预测的可信度与推广性。","openalex","2026-09-27T23:31:22.771605Z"]