Comparative Analysis of Machine Learning Approaches for Crop and Yield Prediction: A Survey

Agriculture is an integral part of the economy in most countries, and it provides the primary source of livelihood, income, food, and employment to most rural populations. The food and agriculture organization (FAO) reported that the agricultural population share in the total population is 67%. Agriculture in a country contributes to 39.4% of the GDP, and agricultural goods account for 43% of all the exports. Therefore, enhancing crop production is seen as an essential aspect of agriculture. Machine learning, data mining, and deep learning are the essential analytical technologies that support accurate decision-making in crop yield prediction, which includes some of the assisting conclusions on which crop to grow and the decisions regarding the crops in the growing season on the agricultural land. A mixture of machine learning, data mining, and deep learning algorithms are applied to support crop yield prediction research. The algorithms include several classifications, regression, and clustering techniques. Data in the agricultural field is enormous, considering various parameters and represented in structured/unstructured form. Hence, there is a need for an efficient technique to process these data and discover potential information. This paper mainly focuses on the algorithms that can be used to predict the most suitable crop and estimate the crop yield, which assists the farmers in selecting and growing the most profitable crop and thereby reducing the chances of loss and hence increasing the productivity and the value of his farming area.