A Bibliometric Analysis of Machine Learning Techniques for Predicting Crop Yield

Abstract

Accurate crop yield prediction is paramount in agriculture to improve decision-making and resource allocation. It is a key part of precision agriculture, a sustainable farming approach that uses data and technology to improve crop yields while reducing environmental impact. Machine learning techniques have been extensively researched to enhance the accuracy of yield prediction. Athorough comprehension of the available literature on this topic is imperative. This study aims to categorize research, map relevant literature, and evaluate the progress made in yield prediction with machine learning over the last 13 years using bibliometric techniques. It analyzed the Scopus database for articles using search keywords "Yield Prediction," "Yield Estimation," and "Yield Forecasting" with "Machine Learning." It located 1177 pertinent articles, confirming the significance of machine learning in yield prediction. The Biblioshiny package analyzes the aggregated Scopus database, identifying current trends, challenges, and future research directions. The insights gained from this study can advance agricultural practices and contribute to sustainability efforts. Outcome: The study highlights the crucial role of machine learning in predicting crop yield accurately by using the Scopus database for finding articles, and the biblioshiny package of R is used to identify the current trends, challenges, and future research directions, which has the potential to transform precision agriculture by increasing crop yield and minimizing environmental impact.