Crop Recommendation System with Cloud Computing

Agriculture is the backbone of the developing countries and plays a primary role in the economy in these countries. For bringing in the most productivity, the decision of planting a suitable crop in a particular location is necessary. But, there is a general problem among farmers and other agricultural activists that they don't opt for better scientifically proven methods for crop recommendation. Thus, our proposed work would help farmers in selecting the right crop based on factors like cost of cultivation, cost of production, yield to increase productivity and get more profit out of this proposed technique. This paper discusses about the different machine learning algorithms to know about them, their metrics evaluation for a certain dataset, and finally, a proposed methodology that performs better than other learners. The paper proposes a methodology in which decision tree, k^th nearest neighbor, logistic regression, random forest and gradient boosting classifier are used to process the data set and then, these learners are passed through an ensemble model called voting classifier to get a more improved outcome. The comparison between these algorithms is also shown in terms of metrics – accuracy, f1 score and execution time on the certain dataset used. This paper also discusses cloud computing and the cloud server processing machine learning algorithms to give required output enquired by the end user.