Survey on Progress in Agricultural Yield Prediction-from Machine Learning to Deep Learning

Neetu Agarwal; Susmita Ray; K. C. Tripathi

Abstract

Machine learning has offered new prospects for data-intensive study in the multidisciplinary field of Agri-technology, especially with the rise of big data systems and advanced computers. Machine learning is a powerful decision-making tool for predicting crop yields, as well as determining which crops to plant and what activities to engage in during the crop's growing season. To facilitate the study of the prediction of the agricultural yield many techniques of machine learning have been used. We have seen a lot of progress in agriculture during the previous few decades. During this time, there was a significant shift from simpler algorithms of machine learning to the application of algorithms of deep learning. This study extracts and synthesises the techniques and features which focused on the prediction of crop yield research, specifically connected to machine learning, neural networks, and deep learning, from papers published in last decade. This study examines the work of several researchers to provide a quick summary of the present state of automation in agriculture. In today’s scenario the systems related to Farm management are growing into real-time artificial intelligence-enabled programmes that provides the awareness for taking decisions and actions to be taken by farmer by applying machine learning.