An optimized machine learning approach for predicting various crop yields

Abstract

Agriculture being the most essential and crucial thing for the mankind as well as for the
economy for the countries like India, various crop patterns and their yearly production
statistics derives many conclusions for many places where the actual prediction for the
crop yield plays a major role with respect to certain factors concerned, in our work we
optimize the real time data and use machine learning approaches such as random forest,
multilinear regression, normalization and pearsons correlation coefficient for the prediction
of yield concerned to the state of Telangana considering the factors such as temperature,
humidity, underground water, canals, soil type, season etc. Our model is helpful for more
accurate prediction of the yield for different crops for a farmer friendly and profitable
cultivation. As the algorithms used are of more supervised and powerful it gives the best
results for the user.