 |
Prof. Adil Bagirov Centre for Smart Analytics, Institute of Innovation, Science and Sustainability, Federation University Australia, Australia In this talk, we present global optimization strategies to some popular problems in data science, including cluster analysis, semi-supervised and clusterwise linear regression. We demonstrate that optimization problems in data science have special structures which can be used to develop better models and to design accurate and efficient global optimization methods for solving problems from these models using tools from convex analysis. Numerical results and comparisons affirm that our strategies are efficient approaches. |
