Predicting the Soil Suitability using Machine Learning Techniques

Abstract

Agriculture is a detracting sector in the global providence, which is defined as the practice of cultivating crops. Precision agriculture using machine learning algorithms is one of the fast-growing methodologies. It explores the usage of modern technologies to increase the crop yield rate by decreasing the utilization of fertilizers. The main aim of this study is to predict the soil suitability by utilizing the sensors and machine learning techniques. The temperature, humidity, pH and soil moisture were the main sources for plant growth. The nature of the soil would be identified, by measuring the above said entities. This paper analyses the soil suitability using diversified machine learning techniques such as KNN, Support Vector Machine, Random Forest, Naive Bayes, and Extreme Learning Machine. ELM model predicts the soil suitability with 99% of accuracy.