[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-3626":3,"related-3626":54},{"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":53},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。",null,"Journal of Global Innovations in Agricultural Sciences","2026-09-25T00:00:00Z","论文",10,false,71,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":21,"relevant":22,"comment":23},12,21,17,13,8,1,"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。",[25],{"name":10,"url":6},[27,28,29,30,31],"智慧农业","农业人工智能","机器学习","可解释AI","玉米产量预测",[33,34],"LSTM 玉米产量预测","SHAP LIME 农业模型","LSTM玉米产量预测-3626",0,"10.22194\u002Fjgias\u002F27.2061",{"doi":37,"openalex_id":39,"authors":40,"venue":10,"cited_by_count":36,"oa_url":6,"card":46,"direction":50,"ingested_from":52},"W7214446524",[41,43],{"name":42,"orcid":9},"Stephen Gitau Ndung’u",{"name":44,"orcid":45},"Consolata Gakii","https:\u002F\u002Forcid.org\u002F0000-0003-2783-9992",{"tldr":47,"method":48,"finding":49,"direction":50,"opportunity":51},"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",{"total":55,"page":22,"page_size":55,"items":56},6,[57,99,137,182,231,257],{"id":58,"title":59,"url":60,"summary":61,"summary_zh":9,"content":9,"source_name":62,"source_url":60,"published_at":11,"category":12,"cover_url":9,"hotness":63,"is_selected":14,"score":64,"score_detail":65,"sources":68,"tags":73,"search_phrases":76,"slug":79,"view_count":36,"doi":80,"paper":81,"created_at":98},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.",25,39,{"impact":21,"substance":55,"depth":13,"authority":55,"freshness":66,"relevant":22,"comment":67},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。",[69,70],{"name":62,"url":60},{"name":71,"url":72},"Cureus Journal of Computer Science 2026-09-25","https:\u002F\u002Fwww.cureusjournals.com\u002Farticles\u002F20543",[27,28,30,74,75],"精准农业","气候适应",[77,78],"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":80,"openalex_id":82,"authors":83,"venue":62,"cited_by_count":36,"oa_url":60,"card":9,"direction":50,"ingested_from":52},"W7214363342",[84,87,90,92,94,96],{"name":85,"orcid":86},"Mritunjay Kr. Ranjan","https:\u002F\u002Forcid.org\u002F0000-0003-0240-4909",{"name":88,"orcid":89},"Rohit Gupta","https:\u002F\u002Forcid.org\u002F0000-0002-4436-8275",{"name":91,"orcid":9},"Nitin  D Mali",{"name":93,"orcid":9},"Ansh  A Rajore",{"name":95,"orcid":9},"Gaurav Narendra Patil",{"name":97,"orcid":9},"Ankita  N Patil","2026-09-26T23:30:47.893666Z",{"id":100,"title":101,"url":102,"summary":103,"summary_zh":104,"content":9,"source_name":105,"source_url":102,"published_at":106,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":107,"score_detail":108,"sources":111,"tags":113,"search_phrases":114,"slug":117,"view_count":36,"doi":118,"paper":119,"created_at":136},3468,"A Study of Explainable AI In Smart Agriculture","https:\u002F\u002Fdoi.org\u002F10.64643\u002Fijirt.208800-459","A Study of Explainable AI In Smart Agriculture。International Journal of Innovative Research in Technology","可解释人工智能在智慧农业中的研究。《国际创新技术研究杂志》","International Journal of Innovative Research in Technology","2026-09-24T00:00:00Z",29,{"impact":55,"substance":109,"depth":55,"authority":109,"freshness":66,"relevant":22,"comment":110},4,"主题相关但仅标题与期刊名，无方法、数据与结论细节，信息增量不足，不宜进入每日精选。",[112],{"name":105,"url":102},[27,28,30],[115,116],"农业人工智能 智慧农业 可解释AI","农业人工智能 智慧农业","农业人工智能智慧农业可解释AI-3468","10.64643\u002Fijirt.208800-459",{"doi":118,"openalex_id":120,"authors":121,"venue":105,"cited_by_count":36,"oa_url":102,"card":130,"direction":135,"ingested_from":52},"W7214184415",[122,124,126,128],{"name":123,"orcid":9},"Lokesh Goregaonkar",{"name":125,"orcid":9},"Jay Kakade",{"name":127,"orcid":9},"Arnav Akhade",{"name":129,"orcid":9},"Shubhangi Gaikar",{"tldr":131,"method":132,"finding":133,"direction":50,"opportunity":134},"探讨可解释人工智能在智慧农业中的应用与挑战。","综述可解释AI方法及其在农业场景的适配。","可解释AI能提升农业模型透明度与农户信任。","可探索面向作物病害诊断的可解释模型与农户信任度实证研究。","智慧农业 \u002F 农业物联网","2026-09-25T23:30:09.588043Z",{"id":138,"title":139,"url":140,"summary":141,"summary_zh":142,"content":9,"source_name":143,"source_url":140,"published_at":144,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":145,"score_detail":146,"sources":152,"tags":154,"search_phrases":157,"slug":160,"view_count":36,"doi":161,"paper":162,"created_at":181},3368,"A PCA-based deep feature optimization framework for explainable orange fruit disease classification","https:\u002F\u002Fdoi.org\u002F10.1186\u002Fs12870-026-09984-8","Accurate classification of orange fruit diseases is important for precision agriculture and yield protection. This study develops and rigorously benchmarks a hybrid deep-feature framework for classifying Black Spot, Canker, Fresh, and Greening oranges (1,090 images), combining deep feature extraction, PCA-based dimensionality reduction, and classical machine-learning classification. Eight backbones (seven CNNs and a Vision Transformer, ViT) and four classifiers (32 configurations in total) were evaluated under 5 × 5 repeated stratified cross-validation, with PCA fitted exclusively on training-fold features in every iteration to eliminate data leakage. The proposed ViT + PCA+SVM configuration achieved the highest mean accuracy, 99.12%±0.71%, significantly outperforming every CNN-based backbone, including DenseNet201 + PCA + SVM (98.48%±0.81%, p \u003C 0.001). A dedicated variance-retention sensitivity analysis justifies the 98% threshold used throughout, and ablation experiments confirm that PCA substantially reduces feature dimensionality (by ~ 55.7% for ViT and ~ 76.6% for DenseNet201) without a significant loss of accuracy for either backbone. Explainability analysis — occlusion sensitivity and SHAP for the proposed ViT model, and Grad-CAM and SHAP for the DenseNet201 comparison model — shows that both configurations base predictions on biologically relevant, disease-affected regions of the fruit rather than spurious cues. These results identify ViT + PCA+SVM as the most accurate configuration evaluated, with DenseNet201 + PCA + SVM as a closely competitive, more compact convolutional alternative for intelligent orchard disease-monitoring systems.","橙类果实病害的准确分类对精准农业和产量保护具有重要意义。本研究开发并严格基准测试了一种混合深度特征框架，用于对黑斑病、溃疡病、新鲜和黄龙病橙类（1，090张图像）进行分类，该框架结合了深度特征提取、基于PCA的降维和经典机器学习分类。在5×5重复分层交叉验证下评估了八种骨干网络（七种CNN和一种视觉Transformer，ViT）和四种分类器（共32种配置），每次迭代中PCA仅在训练折特征上拟合以消除数据泄漏。所提出的ViT + PCA+SVM配置取得了最高平均准确率，为99.12%±0.71%，显著优于所有基于CNN的骨干网络，包括DenseNet201 + PCA + SVM（98.48%±0.81%，p \u003C 0.001）。专门的方差保留敏感性分析证明了全程使用的98%阈值是合理的，消融实验证实PCA大幅降低了特征维度（ViT约降低55.7%，DenseNet201约降低76.6%），且两种骨干网络均无显著准确率损失。可解释性分析——对所提出的ViT模型采用遮挡敏感性和SHAP，对DenseNet201对比模型采用Grad-CAM和SHAP——表明两种配置均基于果实中生物学相关的病害影响区域而非虚假线索进行预测。这些结果确定ViT + PCA+SVM为所评估的最准确配置，而DenseNet201 + PCA + SVM则是一种竞争力接近且更紧凑的卷积替代方案，可用于智能果园病害监测系统。","BMC Plant Biology","2026-09-23T00:00:00Z",79,{"impact":147,"substance":148,"depth":149,"authority":150,"freshness":66,"relevant":22,"comment":151},16,22,18,14,"方法严谨、数据与消融实验充分，对果园智能病害监测有实用参考价值，但属细分技术论文，产业影响有限。",[153],{"name":143,"url":140},[27,28,30,155,156],"病害识别","柑橘种植",[158,159],"柑橘病害 深度学习 分类","ViT PCA SVM 病害识别","柑橘病害深度学习分类-3368","10.1186\u002Fs12870-026-09984-8",{"doi":161,"openalex_id":163,"authors":164,"venue":143,"cited_by_count":36,"oa_url":140,"card":176,"direction":50,"ingested_from":52},"W7214068709",[165,167,169,171,174],{"name":166,"orcid":9},"Amruta Hingmire",{"name":168,"orcid":9},"Avinash Golande",{"name":170,"orcid":9},"Vinodkumar Bhutnal",{"name":172,"orcid":173},"Sagar Dhanraj Pande","https:\u002F\u002Forcid.org\u002F0000-0003-4506-6997",{"name":175,"orcid":9},"Tanuja Pande",{"tldr":177,"method":178,"finding":179,"direction":50,"opportunity":180},"提出PCA深度特征优化框架，用ViT+SVM分类橙子病害，准确率达99.12%。","8种骨干网络提取特征，PCA降维，4种分类器，5×5交叉验证。","ViT+PCA+SVM最优，PCA降维超55%且精度不降，可解释性验证有效。","可探索轻量化模型在移动端或边缘设备的实时病害检测与多作物泛化。","2026-09-24T23:30:34.116938Z",{"id":183,"title":184,"url":185,"summary":186,"summary_zh":187,"content":9,"source_name":188,"source_url":185,"published_at":189,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":190,"score_detail":191,"sources":194,"tags":196,"search_phrases":199,"slug":202,"view_count":36,"doi":203,"paper":204,"created_at":230},3280,"Machine Learning for Surface Water Quality Management in Malaysia: Integrating Physicochemical Parameters, Nutrient Loads, Land Use, Hydrological Conditions, and Anthropogenic Pressure","https:\u002F\u002Fdoi.org\u002F10.64440\u002Fijes\u002Fengineeringx0021","Surface water quality in Malaysia is increasingly influenced by rapid urbanization, industrial expansion, agricultural activities, wastewater discharge, climatic variability, and changes in hydrological regimes. These interacting pressures create complex and nonlinear relationships among physicochemical water-quality parameters, watershed characteristics, and pollutant dynamics, making conventional monitoring approaches insufficient for comprehensive and timely assessment. This study develops a data-driven machine learning framework for surface water quality assessment, prediction, pollution-event detection, and management in Malaysian river systems. The proposed framework integrates physicochemical, hydrological, spatial, land-use, and anthropogenic variables to characterize the temporal and spatial variability of water-quality conditions. The investigated parameters include dissolved oxygen, biochemical oxygen demand, chemical oxygen demand, pH, temperature, electrical conductivity, turbidity, total suspended solids, ammoniacal nitrogen, nitrate, phosphate, total nitrogen, and total phosphorus, together with rainfall, streamflow, water level, antecedent rainfall, urbanization, agricultural land use, industrial land use, population density, wastewater pressure, and road density. Seven machine learning approaches are incorporated, including Decision Trees (DT), Random Forest (RF), Support Vector Machines (SVM), Support Vector Regression (SVR), Artificial Neural Networks (ANN), Long Short-Term Memory (LSTM) networks, and hybrid ensemble models. The Malaysian National Water Quality Standards and Water Quality Index (WQI) are incorporated to establish the regulatory and quantitative basis for water-quality classification and prediction. The framework further integrates explainable artificial intelligence techniques, including SHAP and LIME, to quantify predictor contributions and identify the dominant environmental and anthropogenic factors governing model outputs. The proposed methodology enables simultaneous water-quality classification, continuous prediction, anomaly and pollution-event detection, source-pressure identification, and management-oriented decision support. Particular emphasis is placed on temporal variability, nonlinear interactions, hydrological influence, land-use effects, and the integration of multiple environmental data sources. The study provides a comprehensive Malaysia-oriented machine learning framework that can enhance conventional surface-water monitoring by combining predictive modeling, explainability, and integrated watershed information. The findings establish a methodological basis for developing real-time and transferable water-quality prediction systems capable of supporting pollution mitigation, watershed management, and sustainable water-resource planning across Malaysian river basins.","马来西亚地表水水质日益受到快速城市化、工业扩张、农业活动、废水排放、气候变率以及水文情势变化的影响。这些相互作用压力在物理化学水质参数、流域特征与污染物动态之间形成了复杂的非线性关系，使传统监测方法难以实现全面且及时的评估。本研究开发了一个数据驱动的机器学习框架，用于马来西亚河流系统的地表水水质评估、预测、污染事件检测与管理。所提出的框架整合了物理化学、水文、空间、土地利用和人为变量，以刻画水质条件的时间与空间变异性。所研究的参数包括溶解氧、生化需氧量、化学需氧量、pH、温度、电导率、浊度、总悬浮固体、氨氮、硝酸盐、磷酸盐、总氮和总磷，以及降雨量、河流流量、水位、前期降雨、城市化、农业用地、工业用地、人口密度、废水压力和道路密度。该框架纳入了七种机器学习方法，包括决策树（DT）、随机森林（RF）、支持向量机（SVM）、支持向量回归（SVR）、人工神经网络（ANN）、长短期记忆（LSTM）网络以及混合集成模型。研究纳入马来西亚国家水质标准和水质指数（WQI），以建立水质分类与预测的监管和定量基础。该框架进一步整合了可解释人工智能技术，包括SHAP和LIME，以量化预测因子的贡献并识别主导模型输出的主要环境与人为因素。所提出的方法能够同时实现水质分类、连续预测、异常与污染事件检测、源压力识别以及面向管理的决策支持。研究特别强调时间变异性、非线性相互作用、水文影响、土地利用效应以及多环境数据源的整合。该研究提供了一个面向马来西亚的综合机器学习框架，可增强传统","The International Journal of Engineering Sciences","2026-09-22T00:00:00Z",68,{"impact":17,"substance":192,"depth":19,"authority":13,"freshness":66,"relevant":22,"comment":193},20,"马来西亚河流水质机器学习框架，方法体系完整但属区域性研究，对国内农业信息化参考价值有限。",[195],{"name":188,"url":185},[28,29,30,197,198],"水质监测","流域管理",[200,201],"马来西亚 地表水 机器学习","水质指数 机器学习 预测","马来西亚地表水机器学习-3280","10.64440\u002Fijes\u002Fengineeringx0021",{"doi":203,"openalex_id":205,"authors":206,"venue":188,"cited_by_count":36,"oa_url":185,"card":225,"direction":50,"ingested_from":52},"W7214031821",[207,209,211,213,215,217,219,221,223],{"name":208,"orcid":9},"Iskandar M. Wahyu",{"name":210,"orcid":9},"Gusti E. Rosyadi",{"name":212,"orcid":9},"May Abdul Hafed Abdul kader",{"name":214,"orcid":9},"Nuryani U. Humaira",{"name":216,"orcid":9},"Safaruddin P. Nasyita",{"name":218,"orcid":9},"Sudarijati Sudarijati;",{"name":220,"orcid":9},"Ujang Asmil Zuwariah",{"name":222,"orcid":9},"Anisa Ricardi",{"name":224,"orcid":9},"Achuo Azmaine",{"tldr":226,"method":227,"finding":228,"direction":50,"opportunity":229},"构建马来西亚地表水质机器学习框架，融合多源环境数据实现分类、预测与污染事件检测。","用DT、RF、SVM、SVR、ANN、LSTM及混合集成模型，结合SHAP\u002FLI","多源变量与可解释AI能有效刻画水质非线性时空变化并识别主要人为压力源。","可迁移至中国流域，探索农业面源污染与水文气候耦合下的可解释水质预警模型。","2026-09-23T23:30:32.639861Z",{"id":232,"title":233,"url":234,"summary":235,"summary_zh":9,"content":9,"source_name":236,"source_url":234,"published_at":189,"category":12,"cover_url":9,"hotness":13,"is_selected":14,"score":190,"score_detail":237,"sources":239,"tags":241,"search_phrases":244,"slug":247,"view_count":36,"doi":248,"paper":249,"created_at":256},3274,"Farmer-Friendly Decision Support System Using Explainable Orangutan Artificial Intelligence for Soil Fertility","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs42979-026-05329-2","Farmer-Friendly Decision Support System Using Explainable Orangutan Artificial Intelligence for Soil Fertility。SN Computer Science","SN Computer Science",{"impact":17,"substance":149,"depth":147,"authority":20,"freshness":66,"relevant":22,"comment":238},"论文提出可解释猩猩优化算法驱动的土壤肥力决策支持系统，方法新颖且面向农户，但尚属学术探索阶段，产业影响有限。",[240],{"name":236,"url":234},[27,28,30,242,243],"决策支持系统","土壤肥力",[245,246],"农业人工智能 决策支持系统 土壤肥力 智慧农业","农业人工智能 决策支持系统","农业人工智能决策支持系统土壤肥力智慧农业-3274","10.1007\u002Fs42979-026-05329-2",{"doi":248,"openalex_id":250,"authors":251,"venue":236,"cited_by_count":36,"oa_url":9,"card":9,"direction":135,"ingested_from":52},"W7214018467",[252,254],{"name":253,"orcid":9},"K. Komala Devi",{"name":255,"orcid":9},"Josephine Prem Kumar","2026-09-23T23:30:12.314548Z",{"id":258,"title":259,"url":260,"summary":261,"summary_zh":262,"content":9,"source_name":263,"source_url":260,"published_at":189,"category":12,"cover_url":9,"hotness":63,"is_selected":14,"score":264,"score_detail":265,"sources":267,"tags":271,"search_phrases":273,"slug":276,"view_count":36,"doi":277,"paper":278,"created_at":288},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。","Zenodo (CERN European Organization for Nuclear Research)",78,{"impact":147,"substance":148,"depth":149,"authority":20,"freshness":66,"relevant":22,"comment":266},"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。",[268,269],{"name":263,"url":260},{"name":263,"url":270},"https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.22896344",[27,28,29,74,272],"作物推荐",[274,275],"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":277,"openalex_id":279,"authors":280,"venue":263,"cited_by_count":36,"oa_url":260,"card":283,"direction":135,"ingested_from":52},"W7214002748",[281],{"name":282,"orcid":9},"Prof. Nagendra Patel Sahil Verma",{"tldr":284,"method":285,"finding":286,"direction":50,"opportunity":287},"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"]