Developing a Model for Accurate Prediction of Soil Fertility: A Comparative Study of Machine Learning Techniques
Author : Asha S and Sony PS
Abstract :
The prediction of soil fertility is a crucial aspect of agriculture as it helps farmers to make informed decisions about crop selection, fertilization, and irrigation. In this study, we propose use of three different machine learning techniques to predict the fertility of soil. Soil properties such as pH, nitrogen levels, and texture will be analyzed and used as input features for the models. A dataset of labeled soil samples will be collected and used to train the models. The three methods used in this study are: Random Forest, Naïve Bayes, and Support Vector Machine. The performance of the models will be evaluated using metrics such as accuracy, precision, and recall. The goal of this study is to compare the performance of these three machine learning methods and to identify the most efficient method for predicting soil fertility. This method will not only aid in agricultural decision-making but also help in improving crop yields. The results of this study will be useful for farmers, agronomists, and researchers in the field of agriculture. Additionally, this study will also explore the correlation between soil properties and fertility, which will help to understand the mechanism of soil fertility.
Keywords :
Soil fertility, machine learning, support vector machines
