Machine learning approaches to predict the photocatalytic performance of bismuch ferrite-based materials in the removal of malachite green
Published in Journal of Hazardous Materials, 2022
Recommended citation: Jaffari, Z. H., Abbas, A., Lam, S-M., Sanghun, P., Chon, K., Kim, E-S., & Cho, K. H. (2022). Machine learning approaches to predict the photocatalytic performance of bismuch ferrite-based materials in the removal of malachite green. Journal of Hazardous Materials, https://doi.org/10.1016/j.jhazmat.2022.130031
This paper presents application of machine learning and deep learning models for efficiency prediction in removal of malachite green from waste water. We first compared the peformance of over 20 machine learning based algorithms . The hyperparametes of 8 best peforming algorithms were optimized and their performance was compared with two deep learning based architectures, ANN and LightLSTM. We found that the CatBoost algorithm performs best in terms of prediction. The performance of CatBoost model was further analyzed using partial dependence plots, feature interaction plotsnd permutation importance plots.