Smart Farming in Rural Landscapes: Leveraging Machine Learning for Sustainable Agricultural Transformation

Smart Farming in Rural Landscapes: Leveraging Machine Learning for Sustainable Agricultural Transformation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Systematic Review Smart Farming in Rural Landscapes: Leveraging Machine Learning for Sustainable Agricultural Transformation Sukriadi Sukriadi, Andi Adawiah, Ismail Ismail This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7503723/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 13 You are reading this latest preprint version Abstract This study aims to systematically evaluate machine learning (ML) applications in rural agricultural contexts, a critical yet underrepresented area in the literature. Unlike prior reviews focusing on high-tech farming environments, this research uniquely centers on smallholder and resource-constrained systems to explore the intersection of ML, sustainable agriculture, and rural development. A Systematic Literature Review (SLR) was employed, following PRISMA guidelines, and drawing from peer-reviewed databases including Scopus, Web of Science, and IEEE Xplore. The analytical process encompassed five structured stages: data importation, descriptive analysis, interactive visualization, linkage analysis, and insight extraction, ensuring analytical rigor and replicability. The results reveal that although ML technologies such as CNNs, SVMs, and LSTM networks are increasingly used for crop monitoring, disease detection, and irrigation management, their deployment remains predominantly confined to well-resourced agricultural systems. Rural applications face persistent challenges, including limited digital infrastructure, data scarcity, and low digital literacy. Moreover, digital systems have improved rural education processes, showing potential for broader agricultural applications. This study contributes by identifying methodological trends and context-specific gaps, offering a roadmap for developing adaptable, low-cost ML solutions. Limitations include the exclusion of non-English and non-open-access literature and potential biases in database indexing. In conclusion, to realize the full potential of ML in transforming rural agriculture, future research should prioritize inclusive technology design, interdisciplinary collaboration, and policy support. Such efforts are vital to achieving equitable and sustainable food systems in alignment with global development goals. Machine Learning Smart Farming Rural Agriculture Sustainable Development Systematic Literature Review Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction The rapid advancement of digital technologies has significantly influenced the agricultural sector, particularly through the integration of smart farming systems that utilize data-driven tools to enhance productivity, efficiency, and sustainability. Among these tools, machine learning (ML) has emerged as a pivotal technology, enabling intelligent decision-making processes in areas such as crop yield prediction, disease detection, irrigation scheduling, and precision fertilization. However, while smart agriculture has made substantial strides in industrial and large-scale farming contexts, its application in rural and smallholder farming systems—which constitute the backbone of food production in many developing countries—remains limited and fragmented. Recent studies have demonstrated the potential of machine learning to address complex agricultural problems. For instance, Zhang et al. ( 2023 ) applied deep learning models to optimize irrigation in water-scarce regions, showing significant improvements in water-use efficiency. Similarly, Kumar and Bansal ( 2022 ) explored the use of convolutional neural networks (CNNs) for early detection of crop diseases, resulting in higher accuracy and faster diagnosis compared to traditional methods. Despite such advancements, most ML-based agricultural research has focused on well-resourced or commercial farming environments, with comparatively fewer studies explicitly targeting rural and low-resource settings (Alvarez et al., 2023 ; Ramesh & Singh, 2022 ). This gap suggests an urgent need to evaluate and synthesize the current landscape of ML applications in rural agricultural systems, especially in the context of the Sustainable Development Goals. This study addresses a crucial and underexplored intersection between machine learning, rural development, and sustainable agriculture. Unlike prior reviews that focus broadly on ML in agriculture or narrowly on precision farming in high-tech environments, this research uniquely centers on rural landscapes—areas that face distinct constraints such as limited digital infrastructure, low capital investment, and fragmented land holdings. By focusing specifically on rural contexts, the study highlights innovations that are both technologically sound and contextually appropriate, providing novel insights into how ML can transform agriculture in settings often overlooked in mainstream technological discourse. To this end, the study conducts a Systematic Literature Review (SLR) to address the following research questions: (1) What machine learning methods are being used in rural agricultural contexts? (2) What are the primary applications and challenges of ML in these settings? (3) What research gaps and future directions can be identified? Following PRISMA guidelines, this SLR synthesizes peer-reviewed literature from leading databases such as Scopus, Web of Science, and IEEE Xplore. The findings contribute to the academic field by offering a structured overview of existing research, identifying methodological trends and deficiencies, and providing a roadmap for future investigations. Practically, the review informs policymakers, agri-tech developers, and rural practitioners about the feasible integration of ML into smallholder and rural farming systems. 2. Literature Review 2.1. Overview of Smart Farming and Precision Agriculture Smart farming and precision agriculture are modern agricultural paradigms that leverage advanced technologies to optimize farming practices, increase productivity, and reduce environmental impact. Precision agriculture refers to the use of information and communication technologies (ICT) and data analytics to manage variations in the field and apply inputs (e.g., water, fertilizer, pesticides) with high accuracy and efficiency (Zhang et al., 2022). Smart farming expands upon this by integrating technologies such as Internet of Things (IoT) devices, remote sensing, drones, big data analytics, and machine learning (ML) to create intelligent and autonomous farming systems (Wolfert et al., 2017 ; Li et al., 2023 ). These technologies enable real-time monitoring of crop and soil conditions, predictive analytics for yield forecasting, and decision support systems that guide farmers in making informed, data-driven interventions. In addition to sensor-based systems, real-time image processing using YOLO has demonstrated high accuracy in rural health monitoring applications and offers significant potential for real-time agricultural monitoring in low-light and field conditions (Sukriadi et al., 2025 ). Current trends in smart farming are moving toward the development of fully automated agricultural systems that incorporate robotics, satellite imagery, and AI-powered platforms capable of handling large datasets for real-time decision-making. For instance, the integration of deep learning algorithms with drone imagery has shown promising results in detecting plant diseases at early stages (Rahman et al., 2023 ). Moreover, the rise of cloud-based farm management systems has allowed farmers to access remote diagnostics, input recommendations, and weather-based alerts, significantly improving operational efficiency. However, the implementation and effectiveness of smart farming technologies differ markedly between developed and rural or underdeveloped regions. In high-income countries such as the United States, Germany, and Australia, smart farming is supported by well-established digital infrastructure, high investment capacity, and access to technical expertise, making it easier to adopt and scale these innovations (Kamilaris & Prenafeta-Boldú, 2018 ). In contrast, rural agricultural systems in developing countries often face significant barriers, including limited internet connectivity, low literacy levels among farmers, high costs of technological adoption, and a lack of localized data for training ML models (Gandhi et al., 2022 ; Musa et al., 2023 ). These disparities result in a digital divide that risks excluding smallholder and subsistence farmers from the benefits of smart agriculture. Despite these challenges, there is growing interest in adapting smart farming technologies to rural contexts through low-cost, context-specific solutions. For example, mobile-based agricultural advisory apps, solar-powered IoT sensors, and lightweight ML models that operate offline are emerging as practical innovations tailored to rural needs. Furthermore, the development of open-source platforms and participatory research models has facilitated more inclusive technology design, allowing rural farmers to engage with and benefit from digital agriculture. Bridging the gap between high-tech smart farming and resource-constrained rural environments is essential for achieving equitable agricultural transformation and ensuring food security in the face of global climate and population pressures. 2.2. Machine Learning in Agriculture Machine learning (ML) has become a transformative technology in modern agriculture, enabling data-driven insights and automation across various farming processes. By learning patterns from historical and real-time data, ML algorithms can predict outcomes, detect anomalies, and support decision-making in complex environments, including textual classification tasks in education systems (Sukriadi et al., 2023 a). In complex agricultural environments. Among the most widely adopted ML techniques in agriculture are Support Vector Machines (SVMs), Random Forests (RFs), Convolutional Neural Networks (CNNs), and Long Short-Term Memory (LSTM) networks. Each of these models offers distinct advantages depending on the nature and complexity of the agricultural data. For example, SVMs are frequently used for classification tasks, such as distinguishing between healthy and diseased crops, due to their ability to handle high-dimensional data efficiently. Random Forests, with their ensemble structure, are well-suited for feature-rich datasets and have demonstrated robust performance in yield prediction and soil fertility analysis (Kussul et al., 2022 ). Convolutional Neural Networks (CNNs), a type of deep learning model, have shown exceptional results in image-based agricultural tasks such as leaf disease detection, crop type classification, and weed identification using drone or satellite imagery (Ferentinos, 2022 ). CNNs can automatically extract hierarchical features from complex visual inputs, making them particularly effective in environments where manual monitoring is time-consuming or impractical. On the other hand, LSTM networks, a variant of recurrent neural networks, excel at capturing temporal dependencies in sequential data and are increasingly applied in time-series forecasting tasks such as weather prediction, irrigation scheduling, and crop growth modeling (Gupta et al., 2023 ). These models help farmers anticipate conditions that could affect productivity and plan resource use more effectively. In terms of practical applications, ML has been employed across a wide range of agricultural functions. Crop monitoring systems use sensor data and remote imagery to assess plant health, detect nutrient deficiencies, and evaluate canopy development. In irrigation management, ML models analyze environmental variables like soil moisture, evapotranspiration, and weather forecasts to optimize water distribution, thereby conserving resources while maintaining yield (Pantazi et al., 2021 ). Similarly, ML is pivotal in disease and pest detection, where it enables early diagnosis and localized treatment, significantly reducing crop loss and pesticide use. Beyond production, ML is also being explored in post-harvest processes, such as grading and sorting of produce based on quality metrics. Overall, the integration of machine learning into agriculture marks a shift toward predictive and precision farming, where decisions are guided not by intuition but by data and algorithmic intelligence. While challenges remain—particularly regarding model generalization across diverse geographies and crop systems—ongoing advancements in data collection, cloud computing, and algorithmic efficiency are steadily expanding the applicability and accessibility of ML solutions in agriculture 2.3. Rural Agricultural Contexts Rural farming systems, particularly in developing and underdeveloped regions, are characterized by smallholder and subsistence-based agriculture, often operated by families with limited access to capital, technology, and infrastructure. These systems are typically labor-intensive, reliant on traditional knowledge, and vulnerable to environmental and market fluctuations (World Bank, 2022 ). Farms in rural areas usually operate on small plots of land with low input use, fragmented value chains, and limited market access. Despite these challenges, rural agriculture remains a critical pillar for food security, employment, and livelihoods, especially in regions such as Sub-Saharan Africa, South Asia, and parts of Southeast Asia (FAO, 2023 ). However, the integration of advanced technologies like machine learning in these contexts is still in its infancy due to several persistent barriers. One of the most pressing challenges in implementing machine learning solutions in rural agriculture is the lack of digital infrastructure. Many rural communities face poor internet connectivity, unreliable electricity, and limited access to smartphones or computing devices, all of which are essential for deploying and operating ML-based tools (Musa et al., 2023 ). Additionally, data scarcity poses a significant obstacle. High-quality, annotated datasets are the foundation of effective ML models, yet rural regions often lack the means to collect, store, and share relevant agricultural data at scale. Localized datasets that reflect the specific soil types, crop varieties, climate patterns, and farming practices of rural areas are rare, making it difficult to develop accurate and generalizable models (Gandhi et al., 2022 ). Moreover, limited digital literacy among rural farmers further impedes the adoption of machine learning technologies. Even when solutions are available, the absence of tailored training and user-friendly interfaces reduces their practical utility. There are also economic constraints—smallholder farmers may not have the financial capacity to invest in sensor-based equipment, drones, or proprietary ML platforms. In many cases, the perceived risks of adopting unfamiliar technology outweigh the potential benefits, especially when the returns are not immediate or clearly demonstrable. Social and cultural factors, such as resistance to change or a strong reliance on traditional practices, can also hinder technology adoption. Despite these challenges, there is growing recognition of the need to develop inclusive, low-cost, and context-sensitive ML solutions that are appropriate for rural environments. Recent initiatives have begun to focus on mobile-based advisory platforms, offline-compatible ML models, and participatory approaches that involve farmers in the design and deployment of smart farming technologies. Addressing the barriers to ML adoption in rural agriculture is essential not only for achieving technological equity but also for fostering resilient, sustainable, and productive food systems at a global scale. 2.4. Prior Reviews and Knowledge Gaps In recent years, several review articles have explored the role of machine learning in agriculture, reflecting growing academic and practical interest in data-driven agricultural innovations. For instance, Kamilaris and Prenafeta-Boldú ( 2018 ) conducted a broad review of deep learning applications across various agricultural domains, while Nalepa et al. ( 2021 ) provided a taxonomy of ML algorithms used in precision farming. Other studies, such as those by Liakos et al. ( 2018 ) and Zhai et al. ( 2022 ), have catalogued applications ranging from crop yield prediction to disease detection and soil monitoring. However, these reviews tend to focus predominantly on technologically advanced agricultural systems, often in high-income or industrialized settings, where digital infrastructure, data availability, and capital investment are significantly more robust. As a result, rural and smallholder farming contexts—despite representing a large share of the global agricultural landscape—remain underrepresented in the current literature. Furthermore, while prior reviews have shed light on machine learning techniques and their agricultural applications in general, they rarely address the intersection of ML, rural development, and sustainable agriculture in a structured or targeted way. Most reviews overlook the systemic challenges faced in rural environments, such as data scarcity, digital illiteracy, and limited infrastructure. In addition, few studies adopt a systematic approach that complies with established methodologies such as PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), which ensures transparency, reproducibility, and rigor in literature synthesis. These gaps highlight the need for a Systematic Literature Review (SLR) that not only surveys ML applications in agriculture but also critically examines their suitability, adaptation, and impact within rural farming systems. This SLR is therefore justified in its aim to bridge a critical knowledge gap by focusing explicitly on the use of machine learning in rural agricultural contexts. It contributes to the existing body of knowledge by consolidating recent research, identifying context-specific challenges, and outlining future research directions that align with the needs and constraints of rural communities. Such a targeted review is essential for advancing inclusive and equitable agricultural innovation, particularly as global efforts intensify to achieve the United Nations Sustainable Development Goals (SDGs), including zero hunger and climate resilience. 3. Methodology 3.1. Search Strategy Figure 1 . The document filtering process illustrated in the figure demonstrates a rigorous and systematic approach to literature selection, ensuring the relevance, quality, and accessibility of sources included in the study. Beginning with an initial pool of 5,371 articles, the filtering protocol employed sequential criteria including document type, language, keyword relevance, open access status, and subject area specificity. Notably, filtering by the exact keyword “Machine Learning” reduced the corpus to 3,468 documents, which was further refined to 3,404 English-language articles. The emphasis on open access narrowed the selection to 2,002 articles, ensuring transparency and reproducibility. Finally, the focus on engineering as a subject domain resulted in a highly curated set of 496 documents. This funnel-based strategy not only enhances the precision and reliability of the systematic review but also aligns with established academic standards for evidence-based research. The layered filtering enhances the thematic consistency of the dataset and affirms the robustness of the methodological framework used in this scholarly inquiry. 3.2. Data Analysis Stage The process of achieving data analysis insights, as illustrated in the provided visual framework, delineates a structured, multi-stage approach integral to rigorous quantitative and qualitative research—Figure 2 . Achieving Data Analysis Insight encapsulates this sequential process, beginning with data importation, where the foundation of any analytical endeavor is laid by systematically collecting and preparing raw data for processing. This is succeeded by descriptive analysis, which enables a foundational understanding of data distributions, central tendencies, and variability, serving as a precursor to more advanced inferential techniques. The third phase, interactive visualization, facilitates intuitive and dynamic engagement with data, allowing for the identification of trends, patterns, and anomalies through graphical exploration. Progressing to linkage analysis, researchers uncover complex interrelationships and cluster similar data points, enhancing the interpretative depth of the analysis. The final stage, insight extraction, synthesizes the findings into coherent narratives or conclusions, thereby translating raw data into actionable knowledge. This sequential model underscores a best-practice methodology, aligning with internationally recognized standards for analytical rigor and reproducibility in scientific inquiry. This Research uses RStudio for data analysis because it is a powerful and versatile integrated development environment (IDE) tailored for the R programming language, which excels in statistical computing and data visualization. RStudio provides an intuitive interface that simplifies the process of coding, manipulating datasets, and generating insightful visualizations, making it highly effective for both exploratory and advanced data analysis tasks. Its extensive library of packages supports a wide range of analytical techniques, including regression, clustering, machine learning, and time-series analysis. I particularly appreciate RStudio's ability to handle large datasets efficiently and to produce reproducible results through tools like R Markdown for reporting and Shiny for interactive dashboards. Additionally, as an open-source platform, RStudio aligns with the principles of transparency and scientific rigor, making it a valuable tool in academic and research settings. Overall, RStudio enables me to conduct comprehensive and reliable analyses with clarity and precision. 4. Results and Discussion 4.1. Result a. Trend Analysis Research The trend analysis of research document output from 2019 to 2024 reveals a clear and consistent upward trajectory, culminating in a substantial peak in 2024—Figure 3 . Trend Analysis Research illustrates this growth, showing that the number of documents has increased markedly from approximately 35 publications in 2019 to around 150 in 2024, indicating a more than fourfold growth over six years. This pattern suggests a growing scholarly interest and intensified research activity in the relevant field, likely driven by the emergence of novel technological applications and pressing global challenges that demand innovative solutions. The significant rise in 2024, in particular, may reflect both an acceleration of post-pandemic research productivity and the maturing of interdisciplinary domains such as smart agriculture, AI integration, and sustainable development technologies. This increasing trend not only demonstrates the vitality of the field but also highlights a robust expansion of the research community contributing to this domain, thereby offering promising prospects for future exploration, collaboration, and impactful knowledge dissemination. b. Most Relevant Source The source distribution visualization indicates a diversified yet uneven landscape of scholarly publication, with Agriengineering emerging as the most prominent outlet, contributing three articles, significantly more than other journals, which each contributed a single publication—figure 4 . Distribution of Research Publications by Source illustrates this trend, suggesting that Agriengineering has become a key platform for disseminating research at the intersection of agricultural technology and computational science. Its higher frequency of articles may reflect both the journal’s thematic alignment with current research trends and its receptivity to interdisciplinary submissions in precision agriculture and intelligent systems. The remaining journals—ACTA Technologica Agriculturae, Applied Sciences (Switzerland), Discover Applied Sciences, ENG, International Journal of Computational and Experimental Science and Engineering, IoT, and Sensors—each contribute equally to the scholarly discourse, indicating a broad yet distributed interest across various scientific domains. This dispersion reveals the interdisciplinary nature of the research, encompassing domains such as engineering, computing, environmental monitoring, and agricultural sciences. It highlights the growing recognition across multiple disciplines of the relevance of digital innovations in agriculture. The data overall underscores a strategic opportunity for future research dissemination: while Agriengineering appears central in the current publication landscape, the breadth of contributing journals signals a healthy cross-disciplinary interest that researchers can leverage to reach diverse academic audiences and foster wider impact through transdisciplinary collaboration. c. Most Global Cited Document The citation analysis presented in the table highlights significant variations in scholarly impact among recent publications in the field of machine learning applications in agriculture. Table 1 . Citations and Research Publication Performance in 2024 Based on Source and DOI summarizes these findings, revealing that papers by Prathibha I and Bouni M (2024) demonstrate the highest citation performance, each with a total citation (TC) count of 4 and a normalized total citation per year (TCpY) of 2.00, indicating strong early academic attention. Similarly, articles by Manju G and Praharsha CH each received two citations, reflecting moderate recognition within a short timeframe. In contrast, several papers, including those by Wang J, Boonrang A, and Da Silva ERO, have yet to receive any citations, underscoring the delayed uptake or limited reach of certain studies. Normalized total citation (NTC) values further contextualize the citation dynamics, with top-performing articles maintaining an NTC of 2.67, reinforcing their relative influence despite publication recency. Notably, the articles published in journals such as IOT and INT J COMPUT EXP SCI ENG appear to garner higher visibility, possibly due to their alignment with rapidly evolving subfields or broader dissemination channels. Conversely, some contributions in AGRIENG and ACTA TECHNOL AGRIC show minimal citation impact thus far, suggesting the need for broader outreach or more specialized research relevance. Overall, this citation pattern emphasizes the variability in early academic engagement across outlets and topics, highlighting the importance of both journal selection and topical novelty. It also suggests that the emerging research landscape within agricultural machine learning is still crystallizing, where early citation trends may offer predictive insight into future research trajectories and the positioning of influential work. Table 1 Citations and Research Publication Performance in 2024, Based on Source and DOI Paper DOI TC TCpY NTC PRATHIBHA I, 2024 , INT J COMPUT EXP SCI ENG 10.22399/ijcesen .. 785 4 2.00 2.67 BOUNI M, 2024, IOT 10.3390/iot5040028 4 2.00 2.67 MANJU G, 2024 , ENG 10.3390/eng5040130 2 1.00 1.33 PRAHARSHA CH, 2024, SENSORS 10.3390/s24237858 2 1.00 1.33 MAHMOUD NTA, 2024 , AGRIENG 10.3390/agriengineering6040261 1 0.50 0.67 MANSOURIALAM A, 2024 , ACTA TECHNOL AGRIC 10.2478/ata-2024-0025 1 0.50 0.67 SOLTANI NEZHAD F, 2024 , DISCOV APPL SCI 10.1007/s42452-024-06388-x 1 0.50 0.67 WANG J, 2024 , APPL SCI 10.3390/app142411875 0 0.00 0.00 BOONRANG A, 2024, AGRIENG 10.3390/agriengineering6040250 0 0.00 0.00 DA SILVA ERO, 2024 , AGRIENG 10.3390/agriengineering6040275 0 0.00 0.00 d. Co-occurrence Network The keyword co-occurrence visualization clearly emphasizes "machine learning" as the dominant thematic core in the body of research, underscoring its central role in the context of agricultural innovation—Figure 5 . Visualization of Keyword Network Emphasizing 'Machine Learning' and Its Association with Precision Agriculture illustrates this dynamic, showing that surrounding terms such as "precision agriculture" and "smart agriculture" appear in close proximity, signifying their strong conceptual linkage and frequent joint occurrence in recent scholarly publications. This dense clustering suggests that machine learning is not being explored in isolation but rather as a foundational technology driving advanced agricultural practices. The visual prominence of "machine learning" highlights its interdisciplinary relevance, while the connected yet smaller keywords imply that research is increasingly focused on integrating intelligent technologies into practical agricultural systems. This thematic focus confirms the ongoing convergence of data science and agrotechnology as a critical frontier in addressing global food security and sustainability challenges. e. Collaboration Network The author collaboration network presented highlights the fragmented yet growing scholarly engagement in the intersection of machine learning and agriculture—Figure 6 . Author Collaboration Network in Agricultural Machine Learning Research illustrates this dynamic, revealing multiple small, isolated clusters of co-authorship, indicating limited cross-institutional or transnational collaboration within the field. The most prominent cluster—featuring authors such as Liivapuu, Zaman, Mahmud, and Nta—demonstrates a relatively cohesive research team contributing multiple works, suggesting localized leadership in this niche. However, the prevalence of small, disconnected author groups such as Zhang and Dong, or Rani and Rathibha, illustrates a siloed research environment. This fragmentation suggests that while interest in the topic is emerging globally, collaborative synergies remain underdeveloped. Strengthening inter-cluster collaborations could significantly enhance knowledge integration, innovation, and the overall impact of research on smart farming and machine learning applications in rural contexts. f. Tree Map The treemap visualization illustrates the thematic distribution of research keywords, reflecting the multidisciplinary convergence between artificial intelligence and agricultural science—Figure 7 . Keyword Frequency Treemap in Agricultural Machine Learning Research captures this distribution, where dominant terms such as deep learning (7%), machine learning (5%), and agriculture (3%) underscore the centrality of AI-driven methodologies in addressing agricultural challenges. The prominence of these terms suggests an intensifying scholarly focus on the application of advanced computational models to optimize productivity, decision-making, and environmental monitoring in agricultural contexts. Moreover, keywords such as fertilizers, adaptive boosting, and feature selection indicate growing interest in precision agriculture and data-driven cultivation strategies. Additionally, the presence of more specialized terms—such as image resolution, detection system, pest monitoring, and plant disease—highlights the increasing use of image-based diagnostics and real-time surveillance tools within smart farming practices. This reflects a broader trend of integrating machine vision and sensor data to improve crop health management and automate key agricultural processes. The inclusion of keywords like artificial neural network, benchmarking, and optical data further suggests an emphasis on model performance evaluation and diverse data modalities in system development. Overall, the thematic dispersion reveals a dynamic research landscape that is not only rooted in foundational AI concepts but also evolving toward highly specific, application-oriented innovations in agriculture. This underscores the critical role of interdisciplinary collaboration in fostering robust, scalable, and contextually relevant technological solutions that can meet both current and emerging demands in food systems and environmental sustainability. 4.2. Discussion This review underscores the growing body of literature on the integration of machine learning (ML) in agriculture, particularly within engineering disciplines. The increasing volume of publications between 2019 and 2024 reflects heightened global interest in using ML to address agricultural challenges such as crop prediction, pest detection, irrigation optimization, and soil monitoring. Popular algorithms like Support Vector Machines (SVM), Random Forests (RF), Convolutional Neural Networks (CNN), and Long Short-Term Memory (LSTM) networks dominate the field, evidencing the technological maturity of ML applications (Kamilaris & Prenafeta-Boldú, 2018 ). However, the majority of studies are centered in developed or urban-based agricultural systems with strong digital infrastructure and readily available datasets. Despite the technical advancements, the review reveals a notable lack of attention to rural agricultural contexts. Rural farming systems are often characterized by small-scale operations, traditional practices, and limited access to digital tools and infrastructure. These settings present significant barriers to the adoption of ML, including poor internet connectivity, data scarcity, low digital literacy, and limited institutional support (Zhang et al., 2022). Consequently, while ML holds great promise for transforming agriculture, its benefits have not been equitably distributed. Without targeted research and solutions tailored to rural needs, there is a risk that digital agriculture will exacerbate existing inequalities rather than bridge them. To address these gaps, future research should prioritize the development of context-adaptive ML models that function effectively in data-scarce and resource-limited environments. There is also a need for interdisciplinary approaches that combine technological innovation with rural development, policy, and education. Strengthening collaborations among researchers, local stakeholders, and policymakers is essential to ensuring that ML applications are socially inclusive, economically viable, and practically deployable in rural regions (Wolfert et al., 2017 ). By doing so, ML can truly contribute to sustainable agricultural transformation and improve the livelihoods of farmers in underserved communities. 5. Conclusion This systematic review identifies major trends in the application of artificial intelligence (AI) and deep learning technologies in information system security. There has been a marked rise in scholarly output focusing on machine learning, anomaly detection, and intelligent intrusion detection systems. Deep learning architectures such as convolutional and recurrent neural networks are increasingly favored over traditional rule-based methods due to their adaptability and precision. These approaches are being applied across diverse domains—from industrial IoT to smart agriculture—highlighting the interdisciplinary relevance of AI in enhancing detection accuracy, reducing false positives, and enabling autonomous response mechanisms. For practitioners, the findings emphasize the need to adopt AI-driven models to counter complex and evolving cyber threats effectively. AI systems offer the ability to automate threat detection, reduce response time, and analyze diverse data sources for more informed decision-making. Looking forward, future research should prioritize improving model interpretability and generalizability while exploring decentralized approaches like federated learning and edge AI. Emerging areas such as adversarial AI, quantum-resistant cryptography, and human-AI collaboration also offer promising directions for building next-generation, proactive cybersecurity frameworks. Declarations Funding This research received no external funding. Clinical trial number: not applicable. Corresponding author: Sukriadi ( [email protected] ) Consent to Publish declaration: not applicable. Consent to Participate declaration: not applicable. Ethics declaration: not applicable. Competing Interests: The authors declare no competing interests. Author Contribution S designed the research framework, conducted the systematic literature review, and wrote the main draft of the manuscript.AA contributed to data collection, descriptive analysis, and provided critical revisions to the manuscript.I performed the data visualization, linkage analysis, and supported the discussion of methodological trends.All authors contributed to the interpretation of findings, reviewed the manuscript thoroughly, and approved the final version for submission. Acknowledgement The authors would like to thank Universitas Lamappapoleonro, Soppeng, Indonesia, for providing institutional support throughout this research. We also extend our gratitude to colleagues in the Department of Informatics Engineering and Department of Management for their valuable feedback during the preparation of this manuscript.Special thanks are given to the library staff who facilitated access to academic databases, and to the anonymous reviewers whose constructive comments helped improve the clarity and rigor of the study. Data Availability not applicable. References Alvarez J, Gómez R, Fernández A. Machine learning adoption in smallholder farms: Challenges and perspectives. J Agricultural Inf. 2023;14(1):23–38. Da Silva ERO. Smart farming applications using IoT and ML. Agriengineering. 2024;6(4). https://doi.org/10.3390/agriengineering6040275 . Article 0275. Ferentinos KP. Deep learning models for plant disease detection and classification. Comput Electron Agric. 2022;145:311–8. FAO. The state of food and agriculture 2023: Leveraging automation for sustainable agriculture. Food and Agriculture Organization of the United Nations; 2023. Gandhi R, Patel A, Mehta S. Overcoming the digital divide in smart agriculture: A case of rural India. J Rural Technol. 2022;11(2):88–102. Gupta A, Rao S, Sharma N. Time-series forecasting of crop yields using LSTM networks. J Artif Intell Agric. 2023;6(2):57–70. Kamilaris A, Prenafeta-Boldú FX. Deep learning in agriculture: A survey. Comput Electron Agric. 2018;147:70–90. Kumar V, Bansal R. CNN-based early detection of crop diseases using drone imagery. Int J Precision Agric. 2022;8(3):102–15. Kussul N, Lavreniuk M, Skakun S, Shelestov A. Random Forest-based models for soil fertility analysis in Eastern Europe. Volume 98. Environmental Modelling & Software; 2022. pp. 87–96. Li Z, Yang M, Wu Q. IoT and AI in precision farming: Emerging technologies and applications. Smart Agric. 2023;2(1):15–30. Liakos KG, Busato P, Moshou D, Pearson S, Bochtis D. Machine learning in agriculture: A review. Sensors. 2018;18(8):2674. https://doi.org/10.3390/s18082674 . Manju G. ML-based crop yield analysis. ENG. 2024;5(4). https://doi.org/10.3390/eng5040130 . Article 0130. Mahmoud NTA. Smart irrigation scheduling using AI. Agriengineering. 2024;6(4). https://doi.org/10.3390/agriengineering6040261 . Article 0261. MansouriAlam A. Enhancing pest control with adaptive ML models. ACTA Technologica Agriculturae. 2024;24(2):134–45. https://doi.org/10.2478/ata-2024-0025 . Musa A, Khan T, Abdullahi M. Infrastructure challenges in rural smart farming. Rural Inform Syst. 2023;5(1):12–27. Nalepa J, Marciniak T, Kawulok M. Machine learning applications in precision agriculture: A taxonomy and review. Comput Electron Agric. 2021;176:105638. Pantazi XE, Moshou D, Bochtis D. Intelligent irrigation using ML: Review and case studies. Agric Water Manage. 2021;243:106416. Prathibha I. Agricultural yield prediction using hybrid models. Int J Comput Experimental Sci Eng. 2024;9(1). https://doi.org/10.22399/ijcesen.785 . Article 0785. Rahman A, Islam T, Hasan R. Deep learning for plant disease detection using UAVs. AI Agric. 2023;4(2):48–59. Ramesh S, Singh V. Low-cost ML solutions for smallholder farmers. Technol Dev. 2022;14(3):176–89. Sukriadi S, Ismail I, Andzar AM. (2023). Penerapan text mining dalam klasifikasi judul skripsi yang diusulkan mahasiswa menggunakan metode Naïve Bayes. *Jurnal Ilmiah Sistem Informasi dan Teknik Informatika (JISTI)*, 6(2), 184–90. https://doi.org/10.57093/jisti.v6i2.174 Sukriadi S, Irma I, Ansar H. Sistem informasi pendaftaran peserta didik baru berbasis web di SMP Satap Negeri Tengapadange menggunakan pemodelan waterfall. *Jurnal Ilmiah Sistem Informasi dan Teknik Informatika (JISTI)*. 2023;6(1):68–72. https://doi.org/10.57093/jisti.v6i1.150 . Sukriadi S, Gani H, Yuyun. (2025). Deteksi pengguna masker berbasis pengolahan citra menggunakan algoritma YOLO. *Jurnal Ilmiah Sistem Informasi dan Teknik Informatika (JISTI)*, 8(1), 76–80. https://doi.org/10.57093/jisti.v8i1.274 Soltani Nezhad F. Enhancing crop monitoring using feature selection techniques. Discover Appl Sci. 2024;4(1). https://doi.org/10.1007/s42452-024-06388-x . Article 06388. Wang J. Drone data and ML integration for pest management. Appl Sci. 2024;14(24). https://doi.org/10.3390/app142411875 . Article 11875. Wolfert S, Ge L, Verdouw C, Bogaardt MJ. Big data in smart farming: A review. Agric Syst. 2017;153:69–80. World Bank. World development report 2022: Digital development and the rural economy. World Bank Group; 2022. Zhai Z, Cheng Z, Zhang Y. Machine learning in precision agriculture: Current applications and future directions. Agricultural Inf. 2022;12(4):201–19. Zhang Y, Liu H, Chen L. Smart irrigation optimization with deep learning. J Environ Modelling. 2023;17(3):243–56. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 31 Oct, 2025 Reviews received at journal 23 Oct, 2025 Reviews received at journal 07 Oct, 2025 Reviews received at journal 02 Oct, 2025 Reviewers agreed at journal 02 Oct, 2025 Reviewers agreed at journal 29 Sep, 2025 Reviewers agreed at journal 29 Sep, 2025 Reviewers agreed at journal 29 Sep, 2025 Reviewers agreed at journal 24 Sep, 2025 Reviewers invited by journal 16 Sep, 2025 Editor assigned by journal 09 Sep, 2025 Submission checks completed at journal 09 Sep, 2025 First submitted to journal 31 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7503723","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Systematic Review","associatedPublications":[],"authors":[{"id":517891952,"identity":"6b812fae-1912-4f6a-81db-b2b49e3b22b0","order_by":0,"name":"Sukriadi Sukriadi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5ElEQVRIie3RMQrCMBiG4a8UnCJdI4i9QqQggoJXSRHaRcXRwaEgtJtzBw8T+UGXHKCjk5ODIIiTWKuLCLGjQ94pCTzwJwFstr+sAUi4zCuXCuhXZ4rVIa2kIrwmAVwIVe1eBCbi51F0PiwH7aCYbmm+4PAy5aiTgYgiolzuYtYrZpJyzcG1xHZjIjxeQTaoJBNBzbQcrADIPNiT3IkF+Zv4vwjKwRCmxAR/E/GLCH0cI1zHjOuTeN6FdXWYGO/iZ1Hg3K6DkZdNgst8Mex09kRn04t95KL6ESepC17EZrPZbF89ACO/SuPEp1qmAAAAAElFTkSuQmCC","orcid":"","institution":"Universitas Lamappapoleonro","correspondingAuthor":true,"prefix":"","firstName":"Sukriadi","middleName":"","lastName":"Sukriadi","suffix":""},{"id":517891953,"identity":"d13d3be1-cfc2-4ada-911b-e4138fa109c9","order_by":1,"name":"Andi Adawiah","email":"","orcid":"","institution":"Universitas Lamappapoleonro","correspondingAuthor":false,"prefix":"","firstName":"Andi","middleName":"","lastName":"Adawiah","suffix":""},{"id":517891954,"identity":"7a123cca-c597-4370-bc02-2b2a5eb2e8ac","order_by":2,"name":"Ismail Ismail","email":"","orcid":"","institution":"Universitas Lamappapoleonro","correspondingAuthor":false,"prefix":"","firstName":"Ismail","middleName":"","lastName":"Ismail","suffix":""}],"badges":[],"createdAt":"2025-09-01 03:08:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7503723/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7503723/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92249081,"identity":"14b14e77-682d-4295-a531-cbecefe051fd","added_by":"auto","created_at":"2025-09-26 10:12:38","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":766406,"visible":true,"origin":"","legend":"","description":"","filename":"JurnalSukriadiSmartFarminginRuralLandscapesarticelFIX.docx","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/b70d6b5edd857ba8ca8376f4.docx"},{"id":92246586,"identity":"44554042-e16c-496a-a781-ca833b131082","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6138,"visible":true,"origin":"","legend":"","description":"","filename":"a0706e9d4f1a43af9164aae464583da2.json","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/6aa1ae94daf9eb3b29c6995f.json"},{"id":92247955,"identity":"8a1e7c72-75ad-4322-aab1-78b5a050325f","added_by":"auto","created_at":"2025-09-26 10:04:38","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79720,"visible":true,"origin":"","legend":"","description":"","filename":"TitlePageSmartFarminginRuralLandscapes.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/02757b68814369a9f8e2efbb.pdf"},{"id":92247649,"identity":"7a47b5df-0f4e-4437-9267-4df972f8e3e6","added_by":"auto","created_at":"2025-09-26 09:56:38","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":87336,"visible":true,"origin":"","legend":"","description":"","filename":"a0706e9d4f1a43af9164aae464583da21enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/515b206823d99090155a2e08.xml"},{"id":92247652,"identity":"c9a9d327-a47a-4fa5-89f3-c8977119773e","added_by":"auto","created_at":"2025-09-26 09:56:38","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79253,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/6e2ab2cd1893aa620838c18e.png"},{"id":92246598,"identity":"47dedabb-02c2-47f8-81cc-4d4c5778291c","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":198254,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/e0bd3359576f825574e9c918.png"},{"id":92247655,"identity":"d46ba592-32e0-4ddc-8594-00b1a00702b3","added_by":"auto","created_at":"2025-09-26 09:56:38","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":47798,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/95dfe854408088e9646d6cec.png"},{"id":92246595,"identity":"7121b906-7971-42b8-bb9d-d11222f2f851","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":47499,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/5828ac4423018b0016d29118.png"},{"id":92246594,"identity":"b9ec8ed7-243f-4491-9c6b-01e85ce2daa9","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":208354,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/7495526973083449079430d7.png"},{"id":92246602,"identity":"5c3bfd97-5bb8-4fdf-9ca3-ac1d54ad1a75","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":137724,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/2ea96fe3a7441f5407ea33a7.png"},{"id":92246607,"identity":"5f43ea31-1548-49bf-a28e-18a7f21ef983","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10730,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/8fee7e2d862e1e4a1bb9a5d1.png"},{"id":92246608,"identity":"35f9c9d2-0c25-4f78-8209-92ae59a72a56","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29904,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/141fff562e63a79a9790b279.png"},{"id":92246601,"identity":"7b918e31-5f5d-4717-a250-bee5b9df8d4a","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":12533,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/a586bce76d701eeaeb64d53b.png"},{"id":92246605,"identity":"97cc9722-4174-47ff-acbb-cd11d3aca178","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":10488,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/01ed70a705d3d3134f52ceb8.png"},{"id":92246604,"identity":"1c87b903-c663-4ab7-a98e-b09e9c7e5d9e","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35264,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/bf9eb4b16c1ac3013cb3c0fc.png"},{"id":92246606,"identity":"76e70247-ae76-4555-8521-598202ebc588","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":42934,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/026ae9b02322a512d2aabd99.png"},{"id":92246609,"identity":"d38075a4-7eae-445d-999e-80aed9eceec1","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"xml","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":84752,"visible":true,"origin":"","legend":"","description":"","filename":"a0706e9d4f1a43af9164aae464583da21structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/aaa80dfe7a08714b44391613.xml"},{"id":92246610,"identity":"519a70a4-8608-47b5-9ec7-e85a26f0f04f","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"html","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":93142,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/9085fc0607d65437134d77eb.html"},{"id":92246585,"identity":"408381d3-0ec7-4865-aee5-9da115666169","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":137345,"visible":true,"origin":"","legend":"\u003cp\u003eDocument Filtering Process\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/e386352420a1854d285ceaf1.png"},{"id":92246587,"identity":"acde6168-a623-45ce-bf7b-c938d817d703","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":295240,"visible":true,"origin":"","legend":"\u003cp\u003eAchieving Data Analysis Insight\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/8dcc54f0bdd65c6cf7c4e670.png"},{"id":92247648,"identity":"b437a9f4-35ca-4e52-8bc1-dd7fea068a47","added_by":"auto","created_at":"2025-09-26 09:56:38","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":96274,"visible":true,"origin":"","legend":"\u003cp\u003eTrend Analysis Research\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/eb3a3e5fb22b58218bb107ed.jpg"},{"id":92246592,"identity":"aa8849f8-18a5-4c05-950c-2ac7a50b0d3f","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":249111,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of Research Publications by Source\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/0a5c9dc89a2d20cccc2d0b4a.png"},{"id":92246590,"identity":"bfd72a50-32cf-4abd-9131-bbae4f34eed1","added_by":"auto","created_at":"2025-09-26 09:48:38","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":102566,"visible":true,"origin":"","legend":"\u003cp\u003eVisualization of Keyword Network Emphasizing 'Machine Learning' and Its Association with Precision Agriculture\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/7c50bff312e812353dd551e6.png"},{"id":92247957,"identity":"aefb9e69-9f8e-4745-a59d-f68b4e674611","added_by":"auto","created_at":"2025-09-26 10:04:38","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":254463,"visible":true,"origin":"","legend":"\u003cp\u003eAuthor Collaboration Network in Agricultural Machine Learning Research\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/acdb44fdd788c7eceb6492b8.png"},{"id":92247653,"identity":"7ef556b6-170c-4a2b-aa84-68879315e0f1","added_by":"auto","created_at":"2025-09-26 09:56:38","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":516724,"visible":true,"origin":"","legend":"\u003cp\u003eKeyword Frequency Treemap in Agricultural Machine Learning Research\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/a5f82c19b6e17ff15ea62327.png"},{"id":92249262,"identity":"703e026a-c4e5-4698-b5fe-37fa415c0542","added_by":"auto","created_at":"2025-09-26 10:20:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2303357,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7503723/v1/7a54735b-bd3f-4471-a20b-1f67597d2384.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Smart Farming in Rural Landscapes: Leveraging Machine Learning for Sustainable Agricultural Transformation","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eThe rapid advancement of digital technologies has significantly influenced the agricultural sector, particularly through the integration of smart farming systems that utilize data-driven tools to enhance productivity, efficiency, and sustainability. Among these tools, machine learning (ML) has emerged as a pivotal technology, enabling intelligent decision-making processes in areas such as crop yield prediction, disease detection, irrigation scheduling, and precision fertilization. However, while smart agriculture has made substantial strides in industrial and large-scale farming contexts, its application in rural and smallholder farming systems\u0026mdash;which constitute the backbone of food production in many developing countries\u0026mdash;remains limited and fragmented.\u003c/p\u003e\u003cp\u003eRecent studies have demonstrated the potential of machine learning to address complex agricultural problems. For instance, Zhang et al. (\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2023\u003c/span\u003e) applied deep learning models to optimize irrigation in water-scarce regions, showing significant improvements in water-use efficiency. Similarly, Kumar and Bansal (\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) explored the use of convolutional neural networks (CNNs) for early detection of crop diseases, resulting in higher accuracy and faster diagnosis compared to traditional methods. Despite such advancements, most ML-based agricultural research has focused on well-resourced or commercial farming environments, with comparatively fewer studies explicitly targeting rural and low-resource settings (Alvarez et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Ramesh \u0026amp; Singh, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). This gap suggests an urgent need to evaluate and synthesize the current landscape of ML applications in rural agricultural systems, especially in the context of the Sustainable Development Goals.\u003c/p\u003e\u003cp\u003eThis study addresses a crucial and underexplored intersection between machine learning, rural development, and sustainable agriculture. Unlike prior reviews that focus broadly on ML in agriculture or narrowly on precision farming in high-tech environments, this research uniquely centers on rural landscapes\u0026mdash;areas that face distinct constraints such as limited digital infrastructure, low capital investment, and fragmented land holdings. By focusing specifically on rural contexts, the study highlights innovations that are both technologically sound and contextually appropriate, providing novel insights into how ML can transform agriculture in settings often overlooked in mainstream technological discourse.\u003c/p\u003e\u003cp\u003eTo this end, the study conducts a Systematic Literature Review (SLR) to address the following research questions: (1) What machine learning methods are being used in rural agricultural contexts? (2) What are the primary applications and challenges of ML in these settings? (3) What research gaps and future directions can be identified? Following PRISMA guidelines, this SLR synthesizes peer-reviewed literature from leading databases such as Scopus, Web of Science, and IEEE Xplore. The findings contribute to the academic field by offering a structured overview of existing research, identifying methodological trends and deficiencies, and providing a roadmap for future investigations. Practically, the review informs policymakers, agri-tech developers, and rural practitioners about the feasible integration of ML into smallholder and rural farming systems.\u003c/p\u003e"},{"header":"2. Literature Review","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003e2.1. Overview of Smart Farming and Precision Agriculture\u003c/h2\u003e\u003cp\u003eSmart farming and precision agriculture are modern agricultural paradigms that leverage advanced technologies to optimize farming practices, increase productivity, and reduce environmental impact. Precision agriculture refers to the use of information and communication technologies (ICT) and data analytics to manage variations in the field and apply inputs (e.g., water, fertilizer, pesticides) with high accuracy and efficiency (Zhang et al., 2022). Smart farming expands upon this by integrating technologies such as Internet of Things (IoT) devices, remote sensing, drones, big data analytics, and machine learning (ML) to create intelligent and autonomous farming systems (Wolfert et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). These technologies enable real-time monitoring of crop and soil conditions, predictive analytics for yield forecasting, and decision support systems that guide farmers in making informed, data-driven interventions. In addition to sensor-based systems, real-time image processing using YOLO has demonstrated high accuracy in rural health monitoring applications and offers significant potential for real-time agricultural monitoring in low-light and field conditions (Sukriadi et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eCurrent trends in smart farming are moving toward the development of fully automated agricultural systems that incorporate robotics, satellite imagery, and AI-powered platforms capable of handling large datasets for real-time decision-making. For instance, the integration of deep learning algorithms with drone imagery has shown promising results in detecting plant diseases at early stages (Rahman et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Moreover, the rise of cloud-based farm management systems has allowed farmers to access remote diagnostics, input recommendations, and weather-based alerts, significantly improving operational efficiency.\u003c/p\u003e\u003cp\u003eHowever, the implementation and effectiveness of smart farming technologies differ markedly between developed and rural or underdeveloped regions. In high-income countries such as the United States, Germany, and Australia, smart farming is supported by well-established digital infrastructure, high investment capacity, and access to technical expertise, making it easier to adopt and scale these innovations (Kamilaris \u0026amp; Prenafeta-Bold\u0026uacute;, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). In contrast, rural agricultural systems in developing countries often face significant barriers, including limited internet connectivity, low literacy levels among farmers, high costs of technological adoption, and a lack of localized data for training ML models (Gandhi et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Musa et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). These disparities result in a digital divide that risks excluding smallholder and subsistence farmers from the benefits of smart agriculture.\u003c/p\u003e\u003cp\u003eDespite these challenges, there is growing interest in adapting smart farming technologies to rural contexts through low-cost, context-specific solutions. For example, mobile-based agricultural advisory apps, solar-powered IoT sensors, and lightweight ML models that operate offline are emerging as practical innovations tailored to rural needs. Furthermore, the development of open-source platforms and participatory research models has facilitated more inclusive technology design, allowing rural farmers to engage with and benefit from digital agriculture. Bridging the gap between high-tech smart farming and resource-constrained rural environments is essential for achieving equitable agricultural transformation and ensuring food security in the face of global climate and population pressures.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e\u003ch2\u003e2.2. Machine Learning in Agriculture\u003c/h2\u003e\u003cp\u003eMachine learning (ML) has become a transformative technology in modern agriculture, enabling data-driven insights and automation across various farming processes. By learning patterns from historical and real-time data, ML algorithms can predict outcomes, detect anomalies, and support decision-making in complex environments, including textual classification tasks in education systems (Sukriadi et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2023\u003c/span\u003ea). In complex agricultural environments. Among the most widely adopted ML techniques in agriculture are Support Vector Machines (SVMs), Random Forests (RFs), Convolutional Neural Networks (CNNs), and Long Short-Term Memory (LSTM) networks. Each of these models offers distinct advantages depending on the nature and complexity of the agricultural data. For example, SVMs are frequently used for classification tasks, such as distinguishing between healthy and diseased crops, due to their ability to handle high-dimensional data efficiently. Random Forests, with their ensemble structure, are well-suited for feature-rich datasets and have demonstrated robust performance in yield prediction and soil fertility analysis (Kussul et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eConvolutional Neural Networks (CNNs), a type of deep learning model, have shown exceptional results in image-based agricultural tasks such as leaf disease detection, crop type classification, and weed identification using drone or satellite imagery (Ferentinos, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). CNNs can automatically extract hierarchical features from complex visual inputs, making them particularly effective in environments where manual monitoring is time-consuming or impractical. On the other hand, LSTM networks, a variant of recurrent neural networks, excel at capturing temporal dependencies in sequential data and are increasingly applied in time-series forecasting tasks such as weather prediction, irrigation scheduling, and crop growth modeling (Gupta et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). These models help farmers anticipate conditions that could affect productivity and plan resource use more effectively.\u003c/p\u003e\u003cp\u003eIn terms of practical applications, ML has been employed across a wide range of agricultural functions. Crop monitoring systems use sensor data and remote imagery to assess plant health, detect nutrient deficiencies, and evaluate canopy development. In irrigation management, ML models analyze environmental variables like soil moisture, evapotranspiration, and weather forecasts to optimize water distribution, thereby conserving resources while maintaining yield (Pantazi et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Similarly, ML is pivotal in disease and pest detection, where it enables early diagnosis and localized treatment, significantly reducing crop loss and pesticide use. Beyond production, ML is also being explored in post-harvest processes, such as grading and sorting of produce based on quality metrics.\u003c/p\u003e\u003cp\u003eOverall, the integration of machine learning into agriculture marks a shift toward predictive and precision farming, where decisions are guided not by intuition but by data and algorithmic intelligence. While challenges remain\u0026mdash;particularly regarding model generalization across diverse geographies and crop systems\u0026mdash;ongoing advancements in data collection, cloud computing, and algorithmic efficiency are steadily expanding the applicability and accessibility of ML solutions in agriculture\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003e2.3. Rural Agricultural Contexts\u003c/h2\u003e\u003cp\u003eRural farming systems, particularly in developing and underdeveloped regions, are characterized by smallholder and subsistence-based agriculture, often operated by families with limited access to capital, technology, and infrastructure. These systems are typically labor-intensive, reliant on traditional knowledge, and vulnerable to environmental and market fluctuations (World Bank, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Farms in rural areas usually operate on small plots of land with low input use, fragmented value chains, and limited market access. Despite these challenges, rural agriculture remains a critical pillar for food security, employment, and livelihoods, especially in regions such as Sub-Saharan Africa, South Asia, and parts of Southeast Asia (FAO, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). However, the integration of advanced technologies like machine learning in these contexts is still in its infancy due to several persistent barriers.\u003c/p\u003e\u003cp\u003eOne of the most pressing challenges in implementing machine learning solutions in rural agriculture is the lack of digital infrastructure. Many rural communities face poor internet connectivity, unreliable electricity, and limited access to smartphones or computing devices, all of which are essential for deploying and operating ML-based tools (Musa et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Additionally, data scarcity poses a significant obstacle. High-quality, annotated datasets are the foundation of effective ML models, yet rural regions often lack the means to collect, store, and share relevant agricultural data at scale. Localized datasets that reflect the specific soil types, crop varieties, climate patterns, and farming practices of rural areas are rare, making it difficult to develop accurate and generalizable models (Gandhi et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eMoreover, limited digital literacy among rural farmers further impedes the adoption of machine learning technologies. Even when solutions are available, the absence of tailored training and user-friendly interfaces reduces their practical utility. There are also economic constraints\u0026mdash;smallholder farmers may not have the financial capacity to invest in sensor-based equipment, drones, or proprietary ML platforms. In many cases, the perceived risks of adopting unfamiliar technology outweigh the potential benefits, especially when the returns are not immediate or clearly demonstrable. Social and cultural factors, such as resistance to change or a strong reliance on traditional practices, can also hinder technology adoption.\u003c/p\u003e\u003cp\u003eDespite these challenges, there is growing recognition of the need to develop inclusive, low-cost, and context-sensitive ML solutions that are appropriate for rural environments. Recent initiatives have begun to focus on mobile-based advisory platforms, offline-compatible ML models, and participatory approaches that involve farmers in the design and deployment of smart farming technologies. Addressing the barriers to ML adoption in rural agriculture is essential not only for achieving technological equity but also for fostering resilient, sustainable, and productive food systems at a global scale.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003e2.4. Prior Reviews and Knowledge Gaps\u003c/h2\u003e\u003cp\u003eIn recent years, several review articles have explored the role of machine learning in agriculture, reflecting growing academic and practical interest in data-driven agricultural innovations. For instance, Kamilaris and Prenafeta-Bold\u0026uacute; (\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) conducted a broad review of deep learning applications across various agricultural domains, while Nalepa et al. (\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) provided a taxonomy of ML algorithms used in precision farming. Other studies, such as those by Liakos et al. (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) and Zhai et al. (\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), have catalogued applications ranging from crop yield prediction to disease detection and soil monitoring. However, these reviews tend to focus predominantly on technologically advanced agricultural systems, often in high-income or industrialized settings, where digital infrastructure, data availability, and capital investment are significantly more robust. As a result, rural and smallholder farming contexts\u0026mdash;despite representing a large share of the global agricultural landscape\u0026mdash;remain underrepresented in the current literature.\u003c/p\u003e\u003cp\u003eFurthermore, while prior reviews have shed light on machine learning techniques and their agricultural applications in general, they rarely address the intersection of ML, rural development, and sustainable agriculture in a structured or targeted way. Most reviews overlook the systemic challenges faced in rural environments, such as data scarcity, digital illiteracy, and limited infrastructure. In addition, few studies adopt a systematic approach that complies with established methodologies such as PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), which ensures transparency, reproducibility, and rigor in literature synthesis. These gaps highlight the need for a Systematic Literature Review (SLR) that not only surveys ML applications in agriculture but also critically examines their suitability, adaptation, and impact within rural farming systems.\u003c/p\u003e\u003cp\u003eThis SLR is therefore justified in its aim to bridge a critical knowledge gap by focusing explicitly on the use of machine learning in rural agricultural contexts. It contributes to the existing body of knowledge by consolidating recent research, identifying context-specific challenges, and outlining future research directions that align with the needs and constraints of rural communities. Such a targeted review is essential for advancing inclusive and equitable agricultural innovation, particularly as global efforts intensify to achieve the United Nations Sustainable Development Goals (SDGs), including zero hunger and climate resilience.\u003c/p\u003e\u003c/div\u003e"},{"header":"3. Methodology","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003e3.1. Search Strategy\u003c/h2\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The document filtering process illustrated in the figure demonstrates a rigorous and systematic approach to literature selection, ensuring the relevance, quality, and accessibility of sources included in the study. Beginning with an initial pool of 5,371 articles, the filtering protocol employed sequential criteria including document type, language, keyword relevance, open access status, and subject area specificity. Notably, filtering by the exact keyword \u0026ldquo;Machine Learning\u0026rdquo; reduced the corpus to 3,468 documents, which was further refined to 3,404 English-language articles. The emphasis on open access narrowed the selection to 2,002 articles, ensuring transparency and reproducibility. Finally, the focus on engineering as a subject domain resulted in a highly curated set of 496 documents. This funnel-based strategy not only enhances the precision and reliability of the systematic review but also aligns with established academic standards for evidence-based research. The layered filtering enhances the thematic consistency of the dataset and affirms the robustness of the methodological framework used in this scholarly inquiry.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\u003ch2\u003e3.2. Data Analysis Stage\u003c/h2\u003e\u003cp\u003eThe process of achieving data analysis insights, as illustrated in the provided visual framework, delineates a structured, multi-stage approach integral to rigorous quantitative and qualitative research\u0026mdash;Figure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. \u003cem\u003eAchieving Data Analysis Insight\u003c/em\u003e encapsulates this sequential process, beginning with data importation, where the foundation of any analytical endeavor is laid by systematically collecting and preparing raw data for processing. This is succeeded by descriptive analysis, which enables a foundational understanding of data distributions, central tendencies, and variability, serving as a precursor to more advanced inferential techniques. The third phase, interactive visualization, facilitates intuitive and dynamic engagement with data, allowing for the identification of trends, patterns, and anomalies through graphical exploration. Progressing to linkage analysis, researchers uncover complex interrelationships and cluster similar data points, enhancing the interpretative depth of the analysis. The final stage, insight extraction, synthesizes the findings into coherent narratives or conclusions, thereby translating raw data into actionable knowledge. This sequential model underscores a best-practice methodology, aligning with internationally recognized standards for analytical rigor and reproducibility in scientific inquiry.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThis Research uses RStudio for data analysis because it is a powerful and versatile integrated development environment (IDE) tailored for the R programming language, which excels in statistical computing and data visualization. RStudio provides an intuitive interface that simplifies the process of coding, manipulating datasets, and generating insightful visualizations, making it highly effective for both exploratory and advanced data analysis tasks. Its extensive library of packages supports a wide range of analytical techniques, including regression, clustering, machine learning, and time-series analysis. I particularly appreciate RStudio's ability to handle large datasets efficiently and to produce reproducible results through tools like R Markdown for reporting and Shiny for interactive dashboards. Additionally, as an open-source platform, RStudio aligns with the principles of transparency and scientific rigor, making it a valuable tool in academic and research settings. Overall, RStudio enables me to conduct comprehensive and reliable analyses with clarity and precision.\u003c/p\u003e\u003c/div\u003e"},{"header":"4. Results and Discussion","content":"\u003cp\u003e\u003cstrong\u003e4.1. Result\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ea. Trend Analysis Research\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cdiv class=\"BlockQuote\"\u003e\n \u003cp\u003eThe trend analysis of research document output from 2019 to 2024 reveals a clear and consistent upward trajectory, culminating in a substantial peak in 2024\u0026mdash;Figure \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e. \u003cem\u003eTrend Analysis Research\u003c/em\u003e illustrates this growth, showing that the number of documents has increased markedly from approximately 35 publications in 2019 to around 150 in 2024, indicating a more than fourfold growth over six years. This pattern suggests a growing scholarly interest and intensified research activity in the relevant field, likely driven by the emergence of novel technological applications and pressing global challenges that demand innovative solutions. The significant rise in 2024, in particular, may reflect both an acceleration of post-pandemic research productivity and the maturing of interdisciplinary domains such as smart agriculture, AI integration, and sustainable development technologies. This increasing trend not only demonstrates the vitality of the field but also highlights a robust expansion of the research community contributing to this domain, thereby offering promising prospects for future exploration, collaboration, and impactful knowledge dissemination.\u003c/p\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003eb. Most Relevant Source\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv class=\"BlockQuote\"\u003e\n \u003cp\u003eThe source distribution visualization indicates a diversified yet uneven landscape of scholarly publication, with Agriengineering emerging as the most prominent outlet, contributing three articles, significantly more than other journals, which each contributed a single publication\u0026mdash;figure \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e. Distribution of Research Publications by Source illustrates this trend, suggesting that Agriengineering has become a key platform for disseminating research at the intersection of agricultural technology and computational science. Its higher frequency of articles may reflect both the journal\u0026rsquo;s thematic alignment with current research trends and its receptivity to interdisciplinary submissions in precision agriculture and intelligent systems.\u003c/p\u003e\n \u003cp\u003eThe remaining journals\u0026mdash;ACTA Technologica Agriculturae, Applied Sciences (Switzerland), Discover Applied Sciences, ENG, International Journal of Computational and Experimental Science and Engineering, IoT, and Sensors\u0026mdash;each contribute equally to the scholarly discourse, indicating a broad yet distributed interest across various scientific domains. This dispersion reveals the interdisciplinary nature of the research, encompassing domains such as engineering, computing, environmental monitoring, and agricultural sciences. It highlights the growing recognition across multiple disciplines of the relevance of digital innovations in agriculture. The data overall underscores a strategic opportunity for future research dissemination: while Agriengineering appears central in the current publication landscape, the breadth of contributing journals signals a healthy cross-disciplinary interest that researchers can leverage to reach diverse academic audiences and foster wider impact through transdisciplinary collaboration.\u003c/p\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cstrong\u003ec. Most Global Cited Document\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv class=\"BlockQuote\"\u003e\n \u003cp\u003eThe citation analysis presented in the table highlights significant variations in scholarly impact among recent publications in the field of machine learning applications in agriculture. Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e. Citations and Research Publication Performance in 2024 Based on Source and DOI summarizes these findings, revealing that papers by Prathibha I and Bouni M (2024) demonstrate the highest citation performance, each with a total citation (TC) count of 4 and a normalized total citation per year (TCpY) of 2.00, indicating strong early academic attention. Similarly, articles by Manju G and Praharsha CH each received two citations, reflecting moderate recognition within a short timeframe. In contrast, several papers, including those by Wang J, Boonrang A, and Da Silva ERO, have yet to receive any citations, underscoring the delayed uptake or limited reach of certain studies.\u003c/p\u003e\n \u003cp\u003eNormalized total citation (NTC) values further contextualize the citation dynamics, with top-performing articles maintaining an NTC of 2.67, reinforcing their relative influence despite publication recency. Notably, the articles published in journals such as IOT and INT J COMPUT EXP SCI ENG appear to garner higher visibility, possibly due to their alignment with rapidly evolving subfields or broader dissemination channels. Conversely, some contributions in AGRIENG and ACTA TECHNOL AGRIC show minimal citation impact thus far, suggesting the need for broader outreach or more specialized research relevance.\u003c/p\u003e\n \u003cp\u003eOverall, this citation pattern emphasizes the variability in early academic engagement across outlets and topics, highlighting the importance of both journal selection and topical novelty. It also suggests that the emerging research landscape within agricultural machine learning is still crystallizing, where early citation trends may offer predictive insight into future research trajectories and the positioning of influential work.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eCitations and Research Publication Performance in 2024, Based on Source and DOI\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003ePaper\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eDOI\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTC\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eTCpY\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eNTC\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePRATHIBHA I, \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e, INT J COMPUT EXP SCI ENG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.22399/ijcesen\u003c/span\u003e\u003c/span\u003e.. 785\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2.67\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBOUNI M, 2024, IOT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/iot5040028\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2.67\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMANJU G, \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e, ENG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/eng5040130\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePRAHARSHA CH, 2024, SENSORS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/s24237858\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMAHMOUD NTA, \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e, AGRIENG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/agriengineering6040261\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMANSOURIALAM A, \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e, ACTA TECHNOL AGRIC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2478/ata-2024-0025\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSOLTANI NEZHAD F, \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e, DISCOV APPL SCI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s42452-024-06388-x\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eWANG J, \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e, APPL SCI\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/app142411875\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBOONRANG A, 2024, AGRIENG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/agriengineering6040250\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eDA SILVA ERO, \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e, AGRIENG\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/agriengineering6040275\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.00\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003ed. \u003cstrong\u003eCo-occurrence Network\u003c/strong\u003e\u003c/p\u003e\n\u003cdiv class=\"BlockQuote\"\u003e\n \u003cp\u003eThe keyword co-occurrence visualization clearly emphasizes \u0026quot;machine learning\u0026quot; as the dominant thematic core in the body of research, underscoring its central role in the context of agricultural innovation\u0026mdash;Figure \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e. \u003cem\u003eVisualization of Keyword Network Emphasizing \u0026apos;Machine Learning\u0026apos; and Its Association with Precision Agriculture\u003c/em\u003e illustrates this dynamic, showing that surrounding terms such as \u0026quot;precision agriculture\u0026quot; and \u0026quot;smart agriculture\u0026quot; appear in close proximity, signifying their strong conceptual linkage and frequent joint occurrence in recent scholarly publications. This dense clustering suggests that machine learning is not being explored in isolation but rather as a foundational technology driving advanced agricultural practices. The visual prominence of \u0026quot;machine learning\u0026quot; highlights its interdisciplinary relevance, while the connected yet smaller keywords imply that research is increasingly focused on integrating intelligent technologies into practical agricultural systems. This thematic focus confirms the ongoing convergence of data science and agrotechnology as a critical frontier in addressing global food security and sustainability challenges.\u003c/p\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cspan\u003e\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ee. Collaboration Network\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cdiv class=\"BlockQuote\"\u003e\n \u003cp\u003eThe author collaboration network presented highlights the fragmented yet growing scholarly engagement in the intersection of machine learning and agriculture\u0026mdash;Figure \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e. \u003cem\u003eAuthor Collaboration Network in Agricultural Machine Learning Research\u003c/em\u003e illustrates this dynamic, revealing multiple small, isolated clusters of co-authorship, indicating limited cross-institutional or transnational collaboration within the field. The most prominent cluster\u0026mdash;featuring authors such as Liivapuu, Zaman, Mahmud, and Nta\u0026mdash;demonstrates a relatively cohesive research team contributing multiple works, suggesting localized leadership in this niche. However, the prevalence of small, disconnected author groups such as Zhang and Dong, or Rani and Rathibha, illustrates a siloed research environment. This fragmentation suggests that while interest in the topic is emerging globally, collaborative synergies remain underdeveloped. Strengthening inter-cluster collaborations could significantly enhance knowledge integration, innovation, and the overall impact of research on smart farming and machine learning applications in rural contexts.\u003c/p\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cspan\u003e\u003c/span\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ef. Tree Map\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cdiv class=\"BlockQuote\"\u003e\n \u003cp\u003eThe treemap visualization illustrates the thematic distribution of research keywords, reflecting the multidisciplinary convergence between artificial intelligence and agricultural science\u0026mdash;Figure \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e. Keyword Frequency Treemap in Agricultural Machine Learning Research captures this distribution, where dominant terms such as deep learning (7%), machine learning (5%), and agriculture (3%) underscore the centrality of AI-driven methodologies in addressing agricultural challenges. The prominence of these terms suggests an intensifying scholarly focus on the application of advanced computational models to optimize productivity, decision-making, and environmental monitoring in agricultural contexts. Moreover, keywords such as fertilizers, adaptive boosting, and feature selection indicate growing interest in precision agriculture and data-driven cultivation strategies.\u003c/p\u003e\n \u003cp\u003eAdditionally, the presence of more specialized terms\u0026mdash;such as image resolution, detection system, pest monitoring, and plant disease\u0026mdash;highlights the increasing use of image-based diagnostics and real-time surveillance tools within smart farming practices. This reflects a broader trend of integrating machine vision and sensor data to improve crop health management and automate key agricultural processes. The inclusion of keywords like artificial neural network, benchmarking, and optical data further suggests an emphasis on model performance evaluation and diverse data modalities in system development.\u003c/p\u003e\n \u003cp\u003eOverall, the thematic dispersion reveals a dynamic research landscape that is not only rooted in foundational AI concepts but also evolving toward highly specific, application-oriented innovations in agriculture. This underscores the critical role of interdisciplinary collaboration in fostering robust, scalable, and contextually relevant technological solutions that can meet both current and emerging demands in food systems and environmental sustainability.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003e4.2. Discussion\u003c/h2\u003e\n \u003cp\u003eThis review underscores the growing body of literature on the integration of machine learning (ML) in agriculture, particularly within engineering disciplines. The increasing volume of publications between 2019 and 2024 reflects heightened global interest in using ML to address agricultural challenges such as crop prediction, pest detection, irrigation optimization, and soil monitoring. Popular algorithms like Support Vector Machines (SVM), Random Forests (RF), Convolutional Neural Networks (CNN), and Long Short-Term Memory (LSTM) networks dominate the field, evidencing the technological maturity of ML applications (Kamilaris \u0026amp; Prenafeta-Bold\u0026uacute;, \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e). However, the majority of studies are centered in developed or urban-based agricultural systems with strong digital infrastructure and readily available datasets.\u003c/p\u003e\n \u003cp\u003eDespite the technical advancements, the review reveals a notable lack of attention to rural agricultural contexts. Rural farming systems are often characterized by small-scale operations, traditional practices, and limited access to digital tools and infrastructure. These settings present significant barriers to the adoption of ML, including poor internet connectivity, data scarcity, low digital literacy, and limited institutional support (Zhang et al., 2022). Consequently, while ML holds great promise for transforming agriculture, its benefits have not been equitably distributed. Without targeted research and solutions tailored to rural needs, there is a risk that digital agriculture will exacerbate existing inequalities rather than bridge them.\u003c/p\u003e\n \u003cp\u003eTo address these gaps, future research should prioritize the development of context-adaptive ML models that function effectively in data-scarce and resource-limited environments. There is also a need for interdisciplinary approaches that combine technological innovation with rural development, policy, and education. Strengthening collaborations among researchers, local stakeholders, and policymakers is essential to ensuring that ML applications are socially inclusive, economically viable, and practically deployable in rural regions (Wolfert et al., \u003cspan class=\"CitationRef\"\u003e2017\u003c/span\u003e). By doing so, ML can truly contribute to sustainable agricultural transformation and improve the livelihoods of farmers in underserved communities.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThis systematic review identifies major trends in the application of artificial intelligence (AI) and deep learning technologies in information system security. There has been a marked rise in scholarly output focusing on machine learning, anomaly detection, and intelligent intrusion detection systems. Deep learning architectures such as convolutional and recurrent neural networks are increasingly favored over traditional rule-based methods due to their adaptability and precision. These approaches are being applied across diverse domains\u0026mdash;from industrial IoT to smart agriculture\u0026mdash;highlighting the interdisciplinary relevance of AI in enhancing detection accuracy, reducing false positives, and enabling autonomous response mechanisms.\u003c/p\u003e\u003cp\u003eFor practitioners, the findings emphasize the need to adopt AI-driven models to counter complex and evolving cyber threats effectively. AI systems offer the ability to automate threat detection, reduce response time, and analyze diverse data sources for more informed decision-making. Looking forward, future research should prioritize improving model interpretability and generalizability while exploring decentralized approaches like federated learning and edge AI. Emerging areas such as adversarial AI, quantum-resistant cryptography, and human-AI collaboration also offer promising directions for building next-generation, proactive cybersecurity frameworks.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003cbr\u003e\u0026nbsp;This research received no external funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial number:\u003c/strong\u003e not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorresponding author:\u003c/strong\u003e Sukriadi ([email protected])\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Publish declaration:\u003c/strong\u003e not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Participate declaration:\u003c/strong\u003e not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics declaration:\u003c/strong\u003e not applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests:\u003c/strong\u003e The authors declare no competing interests.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eS designed the research framework, conducted the systematic literature review, and wrote the main draft of the manuscript.AA contributed to data collection, descriptive analysis, and provided critical revisions to the manuscript.I performed the data visualization, linkage analysis, and supported the discussion of methodological trends.All authors contributed to the interpretation of findings, reviewed the manuscript thoroughly, and approved the final version for submission.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe authors would like to thank Universitas Lamappapoleonro, Soppeng, Indonesia, for providing institutional support throughout this research. We also extend our gratitude to colleagues in the Department of Informatics Engineering and Department of Management for their valuable feedback during the preparation of this manuscript.Special thanks are given to the library staff who facilitated access to academic databases, and to the anonymous reviewers whose constructive comments helped improve the clarity and rigor of the study.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003enot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAlvarez J, G\u0026oacute;mez R, Fern\u0026aacute;ndez A. Machine learning adoption in smallholder farms: Challenges and perspectives. J Agricultural Inf. 2023;14(1):23\u0026ndash;38.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDa Silva ERO. Smart farming applications using IoT and ML. Agriengineering. 2024;6(4). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/agriengineering6040275\u003c/span\u003e\u003cspan address=\"10.3390/agriengineering6040275\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Article 0275.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFerentinos KP. Deep learning models for plant disease detection and classification. Comput Electron Agric. 2022;145:311\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFAO. The state of food and agriculture 2023: Leveraging automation for sustainable agriculture. Food and Agriculture Organization of the United Nations; 2023.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGandhi R, Patel A, Mehta S. Overcoming the digital divide in smart agriculture: A case of rural India. J Rural Technol. 2022;11(2):88\u0026ndash;102.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGupta A, Rao S, Sharma N. Time-series forecasting of crop yields using LSTM networks. J Artif Intell Agric. 2023;6(2):57\u0026ndash;70.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKamilaris A, Prenafeta-Bold\u0026uacute; FX. Deep learning in agriculture: A survey. Comput Electron Agric. 2018;147:70\u0026ndash;90.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKumar V, Bansal R. CNN-based early detection of crop diseases using drone imagery. Int J Precision Agric. 2022;8(3):102\u0026ndash;15.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKussul N, Lavreniuk M, Skakun S, Shelestov A. Random Forest-based models for soil fertility analysis in Eastern Europe. Volume 98. Environmental Modelling \u0026amp; Software; 2022. pp. 87\u0026ndash;96.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi Z, Yang M, Wu Q. IoT and AI in precision farming: Emerging technologies and applications. Smart Agric. 2023;2(1):15\u0026ndash;30.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiakos KG, Busato P, Moshou D, Pearson S, Bochtis D. Machine learning in agriculture: A review. Sensors. 2018;18(8):2674. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/s18082674\u003c/span\u003e\u003cspan address=\"10.3390/s18082674\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eManju G. ML-based crop yield analysis. ENG. 2024;5(4). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/eng5040130\u003c/span\u003e\u003cspan address=\"10.3390/eng5040130\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Article 0130.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMahmoud NTA. Smart irrigation scheduling using AI. Agriengineering. 2024;6(4). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/agriengineering6040261\u003c/span\u003e\u003cspan address=\"10.3390/agriengineering6040261\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Article 0261.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMansouriAlam A. Enhancing pest control with adaptive ML models. ACTA Technologica Agriculturae. 2024;24(2):134\u0026ndash;45. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2478/ata-2024-0025\u003c/span\u003e\u003cspan address=\"10.2478/ata-2024-0025\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMusa A, Khan T, Abdullahi M. Infrastructure challenges in rural smart farming. Rural Inform Syst. 2023;5(1):12\u0026ndash;27.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNalepa J, Marciniak T, Kawulok M. Machine learning applications in precision agriculture: A taxonomy and review. Comput Electron Agric. 2021;176:105638.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePantazi XE, Moshou D, Bochtis D. Intelligent irrigation using ML: Review and case studies. Agric Water Manage. 2021;243:106416.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePrathibha I. Agricultural yield prediction using hybrid models. Int J Comput Experimental Sci Eng. 2024;9(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.22399/ijcesen.785\u003c/span\u003e\u003cspan address=\"10.22399/ijcesen.785\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Article 0785.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRahman A, Islam T, Hasan R. Deep learning for plant disease detection using UAVs. AI Agric. 2023;4(2):48\u0026ndash;59.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRamesh S, Singh V. Low-cost ML solutions for smallholder farmers. Technol Dev. 2022;14(3):176\u0026ndash;89.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSukriadi S, Ismail I, Andzar AM. (2023). Penerapan text mining dalam klasifikasi judul skripsi yang diusulkan mahasiswa menggunakan metode Na\u0026iuml;ve Bayes. *Jurnal Ilmiah Sistem Informasi dan Teknik Informatika (JISTI)*, 6(2), 184\u0026ndash;90. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.57093/jisti.v6i2.174\u003c/span\u003e\u003cspan address=\"10.57093/jisti.v6i2.174\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSukriadi S, Irma I, Ansar H. Sistem informasi pendaftaran peserta didik baru berbasis web di SMP Satap Negeri Tengapadange menggunakan pemodelan waterfall. *Jurnal Ilmiah Sistem Informasi dan Teknik Informatika (JISTI)*. 2023;6(1):68\u0026ndash;72. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.57093/jisti.v6i1.150\u003c/span\u003e\u003cspan address=\"10.57093/jisti.v6i1.150\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSukriadi S, Gani H, Yuyun. (2025). Deteksi pengguna masker berbasis pengolahan citra menggunakan algoritma YOLO. *Jurnal Ilmiah Sistem Informasi dan Teknik Informatika (JISTI)*, 8(1), 76\u0026ndash;80. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.57093/jisti.v8i1.274\u003c/span\u003e\u003cspan address=\"10.57093/jisti.v8i1.274\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSoltani Nezhad F. Enhancing crop monitoring using feature selection techniques. Discover Appl Sci. 2024;4(1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s42452-024-06388-x\u003c/span\u003e\u003cspan address=\"10.1007/s42452-024-06388-x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Article 06388.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang J. Drone data and ML integration for pest management. Appl Sci. 2024;14(24). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/app142411875\u003c/span\u003e\u003cspan address=\"10.3390/app142411875\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Article 11875.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWolfert S, Ge L, Verdouw C, Bogaardt MJ. Big data in smart farming: A review. Agric Syst. 2017;153:69\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWorld Bank. World development report 2022: Digital development and the rural economy. World Bank Group; 2022.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhai Z, Cheng Z, Zhang Y. Machine learning in precision agriculture: Current applications and future directions. Agricultural Inf. 2022;12(4):201\u0026ndash;19.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhang Y, Liu H, Chen L. Smart irrigation optimization with deep learning. J Environ Modelling. 2023;17(3):243\u0026ndash;56.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"discover-food","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"discoverfood","sideBox":"Learn more about [Discover Food](https://www.springer.com/44187)","snPcode":"","submissionUrl":"","title":"Discover Food","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Machine Learning, Smart Farming, Rural Agriculture, Sustainable Development, Systematic Literature Review","lastPublishedDoi":"10.21203/rs.3.rs-7503723/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7503723/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study aims to systematically evaluate machine learning (ML) applications in rural agricultural contexts, a critical yet underrepresented area in the literature. Unlike prior reviews focusing on high-tech farming environments, this research uniquely centers on smallholder and resource-constrained systems to explore the intersection of ML, sustainable agriculture, and rural development. A Systematic Literature Review (SLR) was employed, following PRISMA guidelines, and drawing from peer-reviewed databases including Scopus, Web of Science, and IEEE Xplore. The analytical process encompassed five structured stages: data importation, descriptive analysis, interactive visualization, linkage analysis, and insight extraction, ensuring analytical rigor and replicability. The results reveal that although ML technologies such as CNNs, SVMs, and LSTM networks are increasingly used for crop monitoring, disease detection, and irrigation management, their deployment remains predominantly confined to well-resourced agricultural systems. Rural applications face persistent challenges, including limited digital infrastructure, data scarcity, and low digital literacy. Moreover, digital systems have improved rural education processes, showing potential for broader agricultural applications. This study contributes by identifying methodological trends and context-specific gaps, offering a roadmap for developing adaptable, low-cost ML solutions. Limitations include the exclusion of non-English and non-open-access literature and potential biases in database indexing. In conclusion, to realize the full potential of ML in transforming rural agriculture, future research should prioritize inclusive technology design, interdisciplinary collaboration, and policy support. Such efforts are vital to achieving equitable and sustainable food systems in alignment with global development goals.\u003c/p\u003e","manuscriptTitle":"Smart Farming in Rural Landscapes: Leveraging Machine Learning for Sustainable Agricultural Transformation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-26 09:48:33","doi":"10.21203/rs.3.rs-7503723/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-10-31T17:15:30+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-23T21:54:54+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-07T14:16:21+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-10-02T20:43:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"327366821154824937439467439149233355598","date":"2025-10-02T12:50:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"156815074291954397636283103667598165892","date":"2025-09-29T13:42:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"80071015829158678912162314129915294070","date":"2025-09-29T12:47:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"148094119776081465525518365127072457941","date":"2025-09-29T12:42:00+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"267376517634243510598692904441348539382","date":"2025-09-25T03:56:50+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-16T13:27:56+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-09T15:15:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-09T15:14:09+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Food","date":"2025-09-01T02:53:10+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-food","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"discoverfood","sideBox":"Learn more about [Discover Food](https://www.springer.com/44187)","snPcode":"","submissionUrl":"","title":"Discover Food","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e3533fb2-d9ec-4d1d-80db-9b895624f6aa","owner":[],"postedDate":"September 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-02-13T14:26:06+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-26 09:48:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7503723","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7503723","identity":"rs-7503723","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}