Crop Yield Prediction using Machine Learning and Deep Learning Techniques.

Agriculture is a significant contributor to India's economic growth. The rising population of country and constantly changing climatic conditions have an impact on crop production and food security. A variety of factors influence crop selection, including market price, production rate, soil type, rainfall, temperature, government policies, etc. Many changes are required in the agricultural sector in order to enhance the Indian economy. In this research work authors have implemented various machine learning techniques to estimate the crop yield in Rajasthan state of India on five identified crops. The results indicate that among all the applied algorithms; Random Forest, SVM, Gradient Descent, long short-term memory, and Lasso regression techniques; the random forest performed better than others with 0.963 R², 0.035 RMSE, and 0.0251 MAE. The results were validated using R², root mean squared error, and the mean absolute error to cross-validation techniques. This paper intends to put the crop selection method into practice to help farmers solve crop yield problems. [ABSTRACT FROM AUTHOR]