Optimization of machine learning algorithms for remote alteration mapping.

• This study focused on applicability of optimized algorithms in alteration mapping. • The ML algorithms considered for optimization included QDA, CKNN, and BDT. • The optimization process employed various techniques, such as GS, RS, BO, and PCA. • Using the optimized algorithms holds promise for enhancing the precision of models. • The study highlights the implications of ML methods for mineral explorations. World-class large to sub-economic small porphyry copper deposits (PCDs) are primarily found in the Kerman Cenozoic Magmatic Arc (KCMA) which is a fascinating area for geological remote sensing investigations because of its well-exposed rocks and roughly vegetated surfaces. Remote hydrothermal alteration mapping is a critical component of mineral exploration and resource assessment, vital for identifying PCDs. This study explored the application of ML techniques, such as quadratic discriminant analysis (QDA), cosine K-nearest neighbor (CKNN), and bagging decision tree (BDT), in remote hydrothermal alteration mapping. Moreover, the study highlights the transformative impact of optimization methods such as grid search (GS), random search (RS), Bayesian optimization (BO), and principal component analysis (PCA) methods in fine-tuning these algorithms to achieve superior results. These algorithms were found accurate and helpful in identifying PCD-related argillic, phyllic, propylitic, and iron oxide/hydroxide alteration-type zones based on field observations, petrographic studies, and XRD analysis. This research revealed evidence for widespread phyllic and silicic alteration zones, as well as confined argillic and iron oxide/hydroxide zones surrounded by wider regions of propylitic alteration. As the field of ML continues to advance, the future holds promise for even more refined and innovative approaches to hydrothermal alteration mapping. This study underscores the pivotal role that optimized ML algorithms play in revolutionizing mineral exploration practices and paving the way for a more sustainable and responsible resource assessment industry. [ABSTRACT FROM AUTHOR]