Radiolarian Classification Decision Support Using Supervised and Unsupervised Learning Approaches

Radiolarians are species belonging to the kingdom Protista characterized by their perfect geometric form and symmetrical glassy skeletons. Most of them are planktonic, that is, they get around by coasting along with ocean currents. Radiolarian assemblages have played a significant role as a biostratigraphic and paleoenvironmental tool used in age-dating, correlation, and studying deep-sea sedimentary rocks that lacks calcareous fossils. The rapid classification of Radiolarian species would allow micropaleontologists to proceed further into studying the their structure as well as their way of living. A recent study on radiolarians included the development of RaDSS, which is a decision support system that could help researchers in classifying microphotographs of Radiolarian species through image processing and machine learning algorithms. This study extends RaDSS by applying approaches in both classical statistics and machine learning in clustering and classifying radiolarians.