Crop Recommendation System Based on Soil Quality and Environmental Factors Using Machine Learning
Harsh Mavi, Santosh Kumar Upadhyay, Nittyansh Srivastava, Rakshita Sharma, Rashmi Bhargava
Abstract
This study is driven by the urgent need for advanced agricultural practices, aiming to transform crop management through a machine learning (ML) approach. Its primary goal is the recommendation of the best possible crop using environmental factors and soil quality based on a wide range of parameters. Employing cutting-edge machine learning techniques, this work seeks to equip farmers with a powerful tool that suggests the best crop to be planted. The proposed crop recommendation system is analyzed by considering important factors such as nitrogen, potassium, phosphorus, temperature, moisture, soil pH and rainfall using decision trees, Naive Bayes, logistic regression, support vector machines, random forests, and XG boost algorithms. The result analysis found that both Random Forest and the XG boost ML algorithms work well with an accuracy of 99.310/0. This technological advancement not only facilitates precision agriculture but also contributes to sustainable farming practices. Our research is an important step towards increasing agricultural productivity by bridging the gap between data-driven insights and on-the-ground decision-making.