Abstract

In agriculture, it is essential to make well-informed decisions regarding crop selection to guarantee the highest possible yield and usage of available resources. The parameters of the weather, particularly the rainfall patterns, play a considerable impact in defining the kind of crop being grown. As a result, the incorporation of machine learning strategies into creating a crop recommendation system founded on rainfall data becomes necessity. This importance comes form the fact that the world is confronted with challenges that motivate research to combine technologies to mitigate these issues in the agriculture sector such as global warming, draught and pollution. In this research, we released a review paper that examined the latest advancements in machine learning in the agricultural scientific domain. The primary objective of this research is to explore the benefits that can be derived from employing machine learning in recommender systems for agricultural purposes. This paper investigated various machine learning methodologies, such as support vector machines (SVM), kernel neural networks (KNN), and random forests. The research concluded that the current mechanisms used for creating the decisions in the recommender system are all based on the supervised machine learning techniques that require to train the knowledge based on the model. Thus, it is recommended that further research to be done in this domain for better accuracy and reduce the time needed for training the model.