Machine Learning in Agriculture: A Comprehensive Updated Review.

Area of Science:

• Agricultural Science
• Computer Science
• Data Science

Background:

• Digital transformation is revolutionizing agriculture, integrating artificial intelligence (AI) for data-driven decision-making.
• Machine learning (ML), a subset of AI, presents a powerful approach to address challenges in establishing knowledge-based farming systems.
• The increasing volume and sources of agricultural data necessitate advanced analytical techniques.

Purpose of the Study:

• To systematically review and synthesize recent scholarly literature on machine learning applications in agriculture.
• To investigate the use of machine learning in crop, water, soil, and livestock management.
• To identify trends, popular algorithms, and data sources within agricultural machine learning research.

Main Methods:

• A comprehensive literature review adhering to PRISMA guidelines.
• Searches were conducted using keyword combinations: "machine learning" with "crop management", "water management", "soil management", and "livestock management".
• Eligible studies were limited to journal papers published between 2018 and 2020.

Main Results:

• The integration of machine learning in agriculture spans multiple disciplines, fostering international convergence research.
• Crop management emerged as the most researched area, with Artificial Neural Networks (ANNs) proving highly efficient among various ML algorithms.
• Maize, wheat, cattle, and sheep were the most frequently studied subjects, utilizing data from sensors on satellites and unmanned vehicles.

Conclusions:

• Machine learning holds substantial promise for advancing agricultural practices through intelligent data analysis.
• The study underscores the need for systematic research and highlights the benefits of ML for stakeholders in the agricultural sector.
• Continued research and adoption of ML technologies are crucial for enhancing agricultural efficiency and sustainability.