A Comparative Analysis of Supervised Machine Learning Algorithm for Agriculture Crop Prediction

Abstract

Due to scarcity of natural resources and degradation in land quality, it has become necessary for every country to take maximum advantage of available natural resources to enhance agricultural activities. Use of machine learning technology with available data, information and knowledge can help us to achieve these goals. Machine learning has emerged with big data to perform large computation. Application of machine learning in agriculture to predict crop yield, recommendation of fertilizers and pesticides can help to improve the quality and quantity of crop production which enhances country's economic condition. This research discuss the importance of machine learning in agriculture with comparative analysis of popular supervised machine learning algorithms like Support Vector Machine, K-Nearest Neighbor, Random Forest and Artificial Neural Network; that predicts suitable crop for a given land based on seasons and soil parameters which can help farmers to make the right decisions about crop selection. We have analyzed & evaluated impact of these algorithms on crop prediction using standard deviation and absolute error method in order to conclude that K-Nearest Neighbor Algorithm out performs other approaches. © 2021 IEEE.