Modeling the effects of essential heavy metals on environmental pollution: A linear and nonlinear prediction model via cascade forward‐neural network

• Published in: Mathematical methods in the applied sciences Vol. 47; no. 6; pp. 4306 - 4318
• Main Authors: Yolcu, Ufuk, Yalcin, Ibrahim Ertugrul, Uras, Mehmet Emin, Ozyigit, Ibrahim Ilker
• Format: Journal Article
• Language: English
• Published: Freiburg Wiley Subscription Services, Inc 01.04.2024
• Subjects: ANN; Artificial intelligence; Artificial neural networks; Biomonitoring; Cost analysis; environmental pollution; Heavy metals; Iron; Manganese; model design; Neural networks; prediction model; Prediction models; Soil properties
• ISSN: 0170-4214; 1099-1476
• DOI: 10.1002/mma.9815

The use of artificial neural networks (ANN) in studies with different content is increasing day by day, and its impact and acceptability are also increasing. The importance of using ANN in popular and large‐scale fields such as environmental pollution studies is known. Studies are carried out considering the need to develop ANNs, which can be an alternative to the data obtained as a result of the analyses made with instrumental devices that require high costs and trained human labor. This study, which was carried out using real data and samples showing biomonitoring properties such as soil and plants, is an example to fill the gaps in the literature. In this study, tools that can model both linear and nonlinear relationships were used. The simulation of the dynamic ecological system containing Fe, Mn, and Ni heavy metals was carried out from the point of view of artificial intelligence. It is revealed that ANN systems are supportive methods in studies of determining environmental pollution, especially in biomonitoring studies.