Revolutionizing Agriculture Through Smart Farming by Employing Advanced Machine Learning Techniques for Optimal Crop Selection

Rajesh Natarajan; Sujatha Krishna; Pradeepa Ganesan; Amna  Al Kaabi; Rajesh  Kala; Anthony Mendoza Madlambayan

doi:10.47852/bonviewJCCE52026049

Authors

Rajesh Natarajan Department of Computing and Information Sciences, University of Technology and Applied Sciences, Oman https://orcid.org/0000-0003-1255-9621
Sujatha Krishna Department of Computing and Information Sciences, University of Technology and Applied Sciences, Oman https://orcid.org/0000-0002-9686-6508
Pradeepa Ganesan Department of Computing and Information Sciences, University of Technology and Applied Sciences, Oman https://orcid.org/0000-0002-5920-066X
Amna Al Kaabi Department of Computing and Information Sciences, University of Technology and Applied Sciences, Oman https://orcid.org/0009-0006-1397-1115
Rajesh Kala Department of Computing and Information Sciences, University of Technology and Applied Sciences, Oman https://orcid.org/0009-0001-2141-3301
Anthony Mendoza Madlambayan Department of Computing and Information Sciences, University of Technology and Applied Sciences, Oman

DOI:

https://doi.org/10.47852/bonviewJCCE52026049

Keywords:

crop selection, machine learning (ML) models, prediction, random forest (RF) classifier, precision agriculture

Abstract

Precision agriculture relies heavily on crop selection to improve sustainability and increase productivity. The increasing problems of soil erosion, climate change, and water shortages have made it crucial to optimize crop selection through cutting-edge methods to increase farm yields and enhance resource efficiency. This research attempts to develop machine learning (ML) models, such as logistic regression (LR), Gaussian naive Bayes (GNB), support vector classifier (SVC), K-nearest neighbors (KNN) classifier, decision tree (DT) classifier, extra tree (ET) classifier, random forest (RF) classifier, and bagging classifier, to optimize crop selection for precision agricultural systems. A large dataset comprising information on crop recommendations, weather patterns, and soil characteristics was used in this research. The data is preprocessed using the interquartile range (IQR) method to remove outliers and ensure that all features contribute equally to the model. Linear discriminant analysis (LDA) is used to extract the important features for feature extraction. The RF classifier was determined to be the most effective method for raising the precision of crop selection forecasts. This framework is designed to provide actionable insights for selecting optimal crops based on environmental conditions and resource availability. The suggested model is a valuable tool for crop selection optimization in precision agriculture, as preliminary results show that it outperforms traditional ML models in terms of F1-score, precision, accuracy, and recall at 99.88% (all measures). According to the research, modern ML algorithms can revolutionize agricultural practices and provide a sustainable solution to increase crop output while reducing resource wastage.

Received: 29 April 2025 | Revised: 13 August 2025 | Accepted: 19 August 2025

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

The Crop Recommendation datasets that support the findings of this study are openly available in Kaggle at https://www.kaggle.com/datasets/atharvaingle/crop-recommendation-dataset. The data that support the findings of this study are openly available in Kaggle at https://www.kaggle.com/datasets/shekharyada/crop-soilcsv.

Author Contribution Statement

Rajesh Natarajan: Conceptualization, Methodology, Validation, Writing – original draft, Writing – review & editing, Funding acquisition. Sujatha Krishna: Validation, Formal analysis, Writing – original draft, Writing – review & editing. Pradeepa Ganesan: Software, Formal analysis, Visualization. Amna Al Kaabi: Writing – review & editing, Supervision. Rajesh Kala: Resources, Data curation. Anthony Mendoza Madlambayan: Resources, Data curation, Visualization.