Machine Learning Based Soil Fertility Prediction
DOI:
https://doi.org/10.59890/ijsss.v4i1.179Keywords:
Machine Learning, Soil Fertility, Prediction, Random Forest, Sustainable AgricultureAbstract
Soil fertility prediction is crucial for optimizing crop yields and reducing environmental impacts. This study proposes a machine learning (ML)-based approach to predict soil fertility using key parameters like pH, nitrogen, phosphorus, and potassium levels. We trained and tested ML models (Random Forest, SVM, Neural Networks) on a dataset of soil samples from Nigerian agricultural lands. Results show that Random Forest achieved 92% accuracy in predicting fertility levels. The model enables farmers to make data-driven decisions on fertilizer application, improving crop productivity and sustainability
References
Akinyemi et al. (2021). Soil classification using ML in Nigeria. _Journal of Soil Science_, 25(3), 123-135.
Bhargava et al. (2020). ML for soil health prediction. _Computers and Electronics in Agriculture_, 178, 105-115.
McGill et al. (2018). Precision agriculture technologies. _Agronomy Journal_, 110(4), 1235-1245.
Nabi et al. (2022). Soil fertility prediction using ML. _Soil Science_, 187(1), 15-28.
Vasques et al. (2018). Digital soil mapping with ML. _Geoderma_, 325, 1-12.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Madumere Smart Onyemaechi, John Peter Uzoma, Ugo Chima, Bob Chile- Agada, Ihim Kingsley, Odoemene .O Ijeoma
This work is licensed under a Creative Commons Attribution 4.0 International License.
