Big Data in agriculture: leveraging large datasets to analyse and improve rice production for better decision-making and operational efficiency

This expanded systematic literature review examines big data technologies in rice production systems through comprehensive analysis of 111 peer-reviewed publications from multiple databases (Google Scholar, PubMed, SciSpace, ArXiv) spanning 2014-2025. The review synthesizes evidence on data integration approaches, machine learning methodologies, operational deployment patterns, and scaling challenges across diverse rice production contexts. Key findings demonstrate significant advances in yield prediction accuracy (85-95%), nutrient management efficiency (10-20% input reductions), and water use optimization (15-30% improvements) through multi-source data fusion and advanced analytics (Cao et al., 2021; Claude et al., 2024; Jeong et al., 2024; Akhter & Sofi, 2024). However, persistent gaps remain in large-scale operational validation, smallholder inclusion, data governance frameworks, and economic impact assessment. The expanded database coverage reveals emerging trends in digital twins, causal machine learning, and microservice architectures for agricultural IoT systems (Patel & Dusi, 2024; Guzman-Lopez et al., 2024; Ahmad et al., 2024), while highlighting insufficient evidence for federated learning implementations and reinforcement learning applications in operational farm settings.