Automatic Crop Recommendation System Using LightGBM and Decision Tree Machine Learning Models

Abstract

An Automatic Crop Recommendation System is a system that makes use of data analysis and algorithms to recommend crops that are suitable and proper about soil quality, climate, and local factors. Such a system eases the decision-making process for farmers. The necessity for efficient agricultural techniques is growing rapidly, and it is impossible without the application of modern technology that would promote the quality of the ideal crop selection list and production. This paper introduces a new concept of the Automatic Crop Recommendation System, integrating the LightGBM and the Decision Tree algorithms. The research uses the strengths of LightGBM, a type of gradient boosting framework, and Decision Tree, a conventional machine learning model, to form a powerful mixed ensemble approach. These approaches are combined to exploit their complementary strengths, leading to a more accurate and dependable agricultural advisory system. The effectiveness of the proposed algorithm’s approach is verified through experimental results it has the following accuracies, recalls, and F-1 scores. The process has proven very successful; an accuracy of 98.64% makes it possible to recommend appropriate and accurate crops.