Machine Learning for Sustainable Agriculture : Enhancing Resource Efficiency and Environmental Conservation

Abstract

Rainfall forecasting is crucial as far as our nation's civilization is concerned, and it occupies a significant part of the daily lives of people. It is the meteorology wing's obligation to anticipate the pattern of downpours considering any kind of uncertainties. Considering the ever-shifting weather patterns, correctly predicting downpours is difficult. This prediction routine becomes even more difficult whenever the season changes take place. Researchers from all around the world have created a variety of methods for forecasting rainfall, most of which use random values and generally are comparable with data on the climate of our nation. As a result of artificial climatic changes, food production and anticipating have declined which will harm the contribution of farming people to the economy by resulting in yields that are low and cause those farming people to become less comfortable with anticipating forthcoming crops. Therefore, in this research work, we are employing five distinct algorithms, namely, DT- Decision Tree, XG Boost, AdaBoost, RF- Random Forest, and SVM- Support Vector Machine for the purpose of predicting the rainfall effects, in turn, predict the yield of the crops for the betterment of the agricultural activities. Out of utilized 5 Machine learning approaches, the Decision Tree approach outperforms the others in terms of accuracy.