Impact of Machine Learning Techniques in Precision Agriculture

Agriculture and its accessories contribute to approximately 17% of India's GDP and it is still the most popular occupation amongst 70% of India's population. The agriculture sector provides different outputs used by diverse segments which include, but not limited to, use as raw materials by different industries, sources of nutrition and businesses, etc. However, the different methods used for growing crops are still mostly traditional and even borderline outdated. Indian farmer still struggles when it comes to picking up the right crop for right biological and non-biological factors. Thus to accelerate the yield of crops, different AI techniques been proposed worldwide and in this paper, we present a summarization of these different approaches. These techniques are a part of the paradigm, Precision Agriculture, more specifically `crop recommender systems'. The diverse procedures presented in this paper include KNN, Similarity-based Models, Ensemble-based Models, Neural Networks, etc. These algorithms take into account various different factors that are external in nature like meteorological data, temperature and others like soil profile and texture to give best recommendations which not only lead to better yields but also minimum use of resources and capital.