Advancing crop recommendation system with supervised machine learning and explainable artificial intelligence

Abstract Agriculture is one of the most important sectors, as many countries’ economies depend on it. At the same time, meeting the food requirements of the increasing population is also challenging, as the land for agriculture is curtailed everywhere. Thus, there is a need to increase crop producti...

Full description

Saved in:
Bibliographic Details
Main Authors: Sourabh Shastri, Sachin Kumar, Vibhakar Mansotra, Rohit Salgotra
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-07003-8
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Agriculture is one of the most important sectors, as many countries’ economies depend on it. At the same time, meeting the food requirements of the increasing population is also challenging, as the land for agriculture is curtailed everywhere. Thus, there is a need to increase crop productivity, and machine learning (ML) techniques are very helpful in recommending suitable crops based on soil, weather, and other parameters. This work presents a crop recommendation model based on Gradient Boosting trained on a crop recommendation dataset. The model can accurately recommend crops based on nutrients and environmental parameters. The proposed method shows promising results in agricultural crop recommendation, with a 99.27% accuracy rate, 99.32% precision, 99.36% recall, and 99.32% F1 score. This proposed model is pertinent because, further, the amalgamation of Explainable Artificial Intelligence (XAI) helps to provide detailed explanations to give agronomists a reliable and steady tool for fast and accurate recommendation of crops.
ISSN:2045-2322