Machine Learning in Sustainable Agriculture: Systematic Review and Research Perspectives

Agriculture, Journal Year: 2025, Volume and Issue: 15(4), P. 377 - 377

Published: Feb. 11, 2025

Machine learning (ML) has revolutionized resource management in agriculture by analyzing vast amounts of data and creating precise predictive models. Precision improves agricultural productivity profitability while reducing costs environmental impact. However, ML implementation faces challenges such as managing large volumes adequate infrastructure. Despite significant advances applications sustainable agriculture, there is still a lack deep systematic understanding several areas. Challenges include integrating sources adapting models to local conditions. This research aims identify trends key players associated with use agriculture. A review was conducted using the PRISMA methodology bibliometric analysis capture relevant studies from Scopus Web Science databases. The study analyzed literature between 2007 2025, identifying 124 articles that meet criteria for certainty assessment. findings show quadratic polynomial growth publication on notable increase up 91% per year. most productive years were 2024, 2022, 2023, demonstrating growing interest field. highlights importance multiple improved decision making, soil health monitoring, interaction climate, topography, properties land crop patterns. Furthermore, evolved weather advanced technologies like Internet Things, remote sensing, smart farming. Finally, agenda need deepening expansion predominant concepts, farming, develop more detailed specialized explore new maximize benefits sustainability.

Language: Английский

---

Transforming Agricultural Productivity with AI-Driven Forecasting: Innovations in Food Security and Supply Chain Optimization

Forecasting, Journal Year: 2024, Volume and Issue: 6(4), P. 925 - 951

Published: Oct. 19, 2024

Global food security is under significant threat from climate change, population growth, and resource scarcity. This review examines how advanced AI-driven forecasting models, including machine learning (ML), deep (DL), time-series models like SARIMA/ARIMA, are transforming regional agricultural practices supply chains. Through the integration of Internet Things (IoT), remote sensing, blockchain technologies, these facilitate real-time monitoring crop allocation, market dynamics, enhancing decision making sustainability. The study adopts a mixed-methods approach, systematic literature analysis case studies. Highlights include yield in European hydroponic systems optimization southeast Asian aquaponics, showcasing localized efficiency gains. Furthermore, AI applications processing, such as plasma, ozone Pulsed Electric Field (PEF) treatments, shown to improve preservation reduce spoilage. Key challenges—such data quality, model scalability, prediction accuracy—are discussed, particularly context data-poor environments, limiting broader applicability. paper concludes by outlining future directions, emphasizing context-specific implementations, need for public–private collaboration, policy interventions enhance scalability adoption contexts.

Language: Английский

---

Counterfactual Based Approaches for Feature Attributions of Stress Factors Affecting Rice Yield

Engineering Reports, Journal Year: 2025, Volume and Issue: 7(1)

Published: Jan. 1, 2025

ABSTRACT Agriculture is a crucial sector in many countries, particularly India, where it significantly influences the economy, food supply, and rural livelihoods. The increased integration of Deep Learning (DL) Machine (ML) into agriculture has enabled substantial advancements predicting crop yields analyzing factors affecting them. counterfactual reasoning framework DICE outperforms LIME offering finer insights feature importance relative impact different on yield prediction. provided clearest causal insights, demonstrating how adjustments to attributes like sandy alfisols surface texture could lead significant changes by water retention nutrient availability. SHAP ranked features phosphate potash based their average across dataset, global view influential but lacking in‐depth understanding. localized immediate influences, such as rainfall nitrogen content, although fell short revealing broader interactions essential for targeted agricultural interventions. findings highlight significance explanations ML models, they provide robust understanding relationships, going beyond correlation‐based attributions. study provides understandable practical allowing focused actions enhance productivity adaptability agriculture. By improving interpretability machine learning research ultimately supports creation predictive systems that strengthen sustainable practices economic development within industry.

Language: Английский

---

Review of Methods and Models for Potato Yield Prediction

Agriculture, Journal Year: 2025, Volume and Issue: 15(4), P. 367 - 367

Published: Feb. 9, 2025

This article provides a comprehensive overview of the development and application statistical methods, process-based models, machine learning, deep learning techniques in potato yield forecasting. It emphasizes importance integrating diverse data sources, including meteorological, phenotypic, remote sensing data. Advances computer technology have enabled creation more sophisticated such as mixed, geostatistical, Bayesian models. Special attention is given to techniques, particularly convolutional neural networks, which significantly enhance forecast accuracy by analyzing complex patterns. The also discusses effectiveness other algorithms, Random Forest Support Vector Machines, capturing nonlinear relationships affecting yields. According standards adopted agricultural research, Mean Absolute Percentage Error (MAPE) implementation prediction issues should generally not exceed 15%. Contemporary research indicates that, through use advanced accurate value this error can reach levels even less than 10 per cent, increasing efficiency Key challenges field include climatic variability difficulties obtaining on soil properties agronomic practices. Despite these challenges, technological advancements present new opportunities for Future focus leveraging Internet Things (IoT) real-time collection impact biological variables yield. An interdisciplinary approach, insights from ecology meteorology, recommended develop innovative predictive exploration methods has potential advance knowledge forecasting support sustainable

Language: Английский

---

Combining biomass production model with machine learning regression of critical nitrogen concentration for estimating grassland nitrogen requirements

Computers and Electronics in Agriculture, Journal Year: 2025, Volume and Issue: 233, P. 110159 - 110159

Published: Feb. 27, 2025

Language: Английский

---

Breeding of Solanaceous Crops Using AI: Machine Learning and Deep Learning Approaches—A Critical Review

Agronomy, Journal Year: 2025, Volume and Issue: 15(3), P. 757 - 757

Published: March 20, 2025

This review discusses the potential of artificial intelligence (AI), particularly machine learning (ML) and its subset, deep (DL), in advancing genetic improvement Solanaceous crops. AI has emerged as a powerful solution to overcome limitations traditional breeding techniques, which often involve time-consuming, resource-intensive processes with limited predictive accuracy. Through advanced algorithms models, ML DL facilitate identification optimization key traits, including higher yield, improved quality, pest resistance, tolerance extreme climatic conditions. By integrating big data analytics omics, these methods enhance genomic selection (GS), support gene-editing technologies like CRISPR-Cas9, accelerate crop breeding, thus enabling development resilient adaptable highlights role improving Solanaceae crops, such tomato, potato, eggplant, pepper, aim developing novel varieties superior agronomic quality traits. Additionally, this study examines advantages AI-driven compared Solanaceae, emphasizing contribution agricultural resilience, food security, environmental sustainability.

Language: Английский

---

Durum Wheat (Triticum durum Desf.) Grain Yield and Protein Estimation by Multispectral UAV Monitoring and Machine Learning Under Mediterranean Conditions

AgriEngineering, Journal Year: 2025, Volume and Issue: 7(4), P. 99 - 99

Published: April 1, 2025

Durum wheat (Triticum durum Desf.), among the herbaceous crops, is one of most extensively grown in Mediterranean area due to its fundamental role supporting typical food productions like bread, pasta, and couscous. Among environmental technical aspects, nitrogen (N) fertilization crucial shaping plant development that kernels by also affecting their protein concentration. Today, new techniques for monitoring fields using uncrewed aerial vehicles (UAVs) can detect crop multispectral (MS) responses, while advanced machine learning (ML) models enable accurate predictions. However, date, there still little research related prediction N nutritional status effects on productivity environment through application these techniques. The present aimed monitor MS responses two different varieties, ancient (Timilia) modern (Ciclope), under three regimens (0, 60, 120 kg ha−1), estimate quantitative qualitative production (i.e., grain yield concentration) Pearson’s correlations five ML approaches. results showed difficulty obtaining good predictive with correlation both varieties data merged together Timilia variety. In contrast, Ciclope, several vegetation indices (VIs) CVI, GNDRE, SRRE) performed well (r-value > 0.7) estimating productive parameters. implementation approaches, particularly random forest (RF) regression, neural network (NN), support vector (SVM), overcame limitations (R2 0.6, RMSE = 0.56 t ha−1, MAE 0.43 ha−1) 0.7, 1.2%, 0.47%) Timilia, whereas RF approach outperformed other methods 0.79, 0.44 ha−1).

Language: Английский

---

South Indian agricultural crop yield prediction using deep learning and transfer learning models

Theoretical and Applied Climatology, Journal Year: 2025, Volume and Issue: 156(5)

Published: May 1, 2025

Language: Английский

---

Predicting the Impact of Extreme Weather on Agricultural Losses in the Delmarva Peninsula using Multi-Step Machine Learning and Financial Crop Loss Data

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 29, 2024

The increasing occurrence of extreme weather events due to climate change presents significant challenges for agricultural production. Existing research on climatic impacts agriculture has predominantly focused changes in yield major crops, providing limited insights into overall losses and diverse regional systems. This study addresses this gap by using financial crop loss data insurance payouts gain a more comprehensive understanding regions. To address the irregular structure data, we developed multi-step machine learning models quantify relationship between weather-related contributing factors. Delmarva Peninsula Eastern United States is used as case location demonstrate methodology over period from 1980 2018. Multi-step configurations linear regression, random forest, support vector approaches are compared terms their classification estimation accuracy repeated hold-out cross-validation analysis. Results indicate that methods, particularly outperform both statistical our null baseline model, demonstrating superior generalizability damage estimation. Multistep distributions shown have influence models' capacity detect estimate occurrence. reveals preference simpler modeling minimize variance handling unseen well importance accounting seasonal patterns, spatial groupings, persistent phenomena accurately estimating losses.

Language: Английский

---

Predictive Modeling of Crop Yield Using Deep Learning Based Transformer with Climate Change Effects

International Research Journal of Multidisciplinary Technovation, Journal Year: 2024, Volume and Issue: unknown, P. 223 - 240

Published: Nov. 30, 2024

Climate change is a significant global challenge concerning agriculture and food security. The understanding of climate effects on crop production necessary for developing an effective adaptation strategies predicting yield accurately. This paper suggests the combined Clustering Long Short Term Memory Transformer (CLSTMT) model prediction. CLSTMT hybrid that integrates clustering, deep learning based LSTM techniques. outliers from historical data are removed using k-means clustering. Followed by, predicted Transformer-based neural network with layers feed-forward (FNN) components. design effectively captures climate-influenced patterns, enhances precision comprehensiveness experiment conducted dataset yield, climate, pesticide details over 101 countries collected 1990 to 2013. comparative analysis reveals outperforms other regression models such as SGDRegressor (SGDR), Lasso Regression (LR), Support Vector (SVR), ElasticNet (EN) Ridge (RR). proposed enhancing predictions. findings indicate provides accurate prediction high R2 0.951 lesser Mean Absolute Percentage Error (MAPE) 0.195. value minimal average percentage deviation between actual yields. more compared others.

Language: Английский

---