A New Approach of Machine Learning And Deep Learning Algorithms Based Crop Yield Prediction

The science and art of raising plants and wild animals is called agriculture. India’s agriculture ranks second in the world and its agricultural areas make up 60.45% of the country’s land. The Indian economy mainly supports the agriculture and agro industrial sectors. Crop rotation, soil consistency, weather and temperature, precipitation, and other factors all affect crop growth. The most important thing is the composition of the soil, which includes nitrogen, phosphate and potassium. Ongoing work in this area includes product selection models using machine learning techniques (random forests, tree pruning, artificial neural networks). In this paper, a recommendation model has been developed using deep learning techniques to obtain accurate information about the amount of suitable land and their added value, as well as crop estimation. It provides better accuracy than existing models. Existing data are reviewed to help farmers estimate crop yields. Take soil and climate differences into account to estimate yields. This project introduces a Python-based system that uses cunning techniques to predict profits while using limited resources. This work combines SVM machine learning algorithm with LSTM and RNN deep learning algorithms.