Evaluation of machine learning techniques to select marine oil spill response methods under small-sized dataset conditions

• Published in: Journal of hazardous materials 2022-08, Vol.436, p.129282-129282, Article 129282
• Main Authors: Mohammadiun, Saeed, Hu, Guangji, Gharahbagh, Abdorreza Alavi, Li, Jianbing, Hewage, Kasun, Sadiq, Rehan
• Format: Article
• Language: English
• Subjects: Artificial intelligence; Decision support; Knowledge discovery; Oil spill incidents; Spill response operations
• ISSN: 0304-3894; 1873-3336
• DOI: 10.1016/j.jhazmat.2022.129282

Oil spill incidents can significantly impact marine ecosystems in Arctic/subarctic areas. Low biodegradation rate, harsh environments, remoteness, and lack of sufficient response infrastructure make those cold waters more susceptible to the impacts of oil spills. A major challenge in Arctic/subarctic areas is to timely select suitable oil spill response methods (OSRMs), concerning the process complexity and insufficient data for decision analysis. In this study, we used various regression-based machine learning techniques, including artificial neural networks (ANNs), Gaussian process regression (GPR), and support vector regression, to develop decision-support models for OSRM selection. Using a small hypothetical oil spill dataset, the modelling performance was thoroughly compared to find techniques working well under data constraints. The regression-based machine learning models were also compared with integrated and optimized fuzzy decision trees models (OFDTs) previously developed by the authors. OFDTs and GPR outperformed other techniques considering prediction power (> 30 % accuracy enhancement). Also, the use of the Bayesian regularization algorithm enhanced the performance of ANNs by reducing their sensitivity to the size of the training dataset (e.g., 29 % accuracy enhancement compared to an unregularized ANN). [Display omitted] •Machine learning-based models are developed for effective response selection.•Models’ performance is compared on a hypothetical oil spill dataset in the Arctic.•The best-performing models suited for small training datasets are selected.•Optimized fuzzy decision trees and Gaussian process regression models outperformed.•Recommendations are provided to prevent models’ overtraining on small datasets.