Mathematical Research at FSU

Data Science

Research in data science at the FSU Mathematics Department covers a wide spectrum of topics in mathematical modeling, foundations, optimization and computation. Methods based on Riemannian geometry, in both finite and infinite dimensions, are employed in shape analysis and studies of manifold-valued data arising in multiple scientific domains. Several data analysis projects employ topology and geometry in the construction of informative data summaries used in such problems as uncovering and analyzing genotype-to-phenotype associations and mapping phenotypic plasticity. Research on random graph theory and stochastic optimization target applications such as development of social network models. Machine learning is being applied to forecasting high frequency financial price data, studying market behavior, and exploiting market inefficiencies. In biological and medical imaging, machine learning has been applied to such problems as segmentation and classification of lung nodules in CT-scans as malignant or benign, and analysis of the morphology of pollen grains, the richest fossil record on planet Earth.

Selected publications: