Comparison of Classification Algorithm for Crop Decision based on Environmental Factors using Machine Learning

Abstract

Crop decision is a very complex process. In Agriculture it plays a vital role. Various biotic and abiotic factors affect this decision. Some crucial Environmental factors are Nitrogen Phosphorus, Potassium, pH, Temperature, Humidity, Rainfall. Machine Learning Algorithm can perfectly predict the crop necessary for this environmental condition. Various algorithms and model are used for this process such as feature selection, data cleaning, Training, and testing split etc. Algorithms such as Logistic regression, Decision Tree, Support vector machine, K- Nearest Neighbour, Navies Bayes, Random Forest. A comparison based on the accuracy parameter is presented in this paper along with various training and testing split for optimal choice of best algorithm. This comparison is done on two tools i.e., on Google collab using python and its libraries for implementation of Machine Learning Algorithm and WEKA which is a pre-processing tool to compare various algorithm of machine learning.