Citation

Janani, U. and Manuel, M. L. Alphin Ezhil and Devi, B. Durga and Thirumalaisamy, Manikandan and Shunmugam, Ramesh and Yogarayan, Sumendra (2026) Machine Learning for Phyto-Remediation of Contaminated Agricultural Soil. In: 9th International Conference on Micro-electronics and Telecommunication Engineering, ICMETE 2025, 4 December 2025 - 5 December 2025, Ghaziabad.

Text Machine Learning for Phyto-Remediation of Contaminated Agricultural Soil _ Springer Nature Link.pdf - Published Version Restricted to Repository staff only Download (738kB)

Abstract

Contaminating lands with metals and harmful chemicals poses a serious risk, to the safety of our food supply and the environment as well as human health concerns. With Phyto remediation becoming popular using plants to absorb and remove soil toxins has proven to be an affordable method. However, picking the right plant species for cleaning up types of soils is tricky due, to the varied soil characteristics and types of pollutants involved. In our study we investigate how machine learning (ML) techniques can be used to forecast the plant species, for cleansing by considering soil traits and contaminant levels. We build a Random Forest model by examining a range of soil features like pH balance, moisture content and organic matter well as the presence of pollutants such as lead, cadmium and arsenic. This model helps us determine the plant species for purifying heavy metals, in polluted soils. The accuracy of the model’s predictions exceeds 85%, which greatly enhances decision making for farmers and environmental scientists alike. Through hyperparameter tuning and analyzing feature importance while comparing models, with Support Vector Machines (SVM) efforts are made to strengthen the model’s reliability and robustness. This study showcases the promise of incorporating machine learning techniques into farming methods by providing a data driven strategy, for soil restoration while aiding in endeavors to preserve soil health amidst industrialization and environmental decline.

Item Type:	Conference or Workshop Item (Paper)
Uncontrolled Keywords:	Machine learning, Pesticides, Phyto-remediation
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Rosnani Abd Wahab
Date Deposited:	29 Oct 2026 07:23
Last Modified:	29 Oct 2026 07:23
URII:	http://shdl.mmu.edu.my/id/eprint/17001

Downloads

Edit (login required)