“…ML models are developed to solve complex classification problems through recursive and iterative analysis of candidate solutions from given training samples and features without explicitly being programmed to do the task [233]. Various classification algorithms, such as artificial neural network (ANN) [50,52,141,146,163,199,207], SVM [145,180], decision tree (DT) [177], K-nearest neighbor [48,64], genetic algorithm [123,127,130], random forest (RF) [26], fuzzy logic [109,135,136,138], maximum likelihood [234], linear discriminant analysis [114,194], k-means [119], Mahalanobis distance [113], naïve Bayes [110], ensemble learning [46,115], Classification and Regression Trees (CART) [132], and others [38,51,140,142,235,236], have been used to classify oil spills and lookalikes. Widely used traditional ML classification model for oil detection from optical and SAR images are listed in Table 4.…”