Evaluation of geochemical characteristics of eastern pontide granitoids with supervised and unsupervised algorithms

Geochemical studies conducted in earth sciences in recent years have begun to produce larger data than before. This increase in geochemical data has paved the way for the use of supervised and unsupervised learning methods in earth sciences. Supervised and unsupervised learning methods are interdisciplinary and include probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It is an effective method for clustering, classification, dimensionality reduction and regression of multivariate systems containing only a few or thousands of variables. In today's modern earth science studies, supervised and unsupervised learning methods are used as alternatives for multivariate analyses. For example, discrimination diagrams, classifying rock samples, mineral identification, etc. With this proposed thesis project, it is aimed to analyze the characteristics of geochemical data of granitoids in the Eastern Black Sea Mountain Belt using supervised and unsupervised algorithms. Using the geochemical data in the existing literature, the geochemical characteristics of the region will be revealed and an attempt will be made to contribute to the geodynamic setting of the region.