Data science is an interdisciplinary area of knowledge that uses scientific approaches, algorithms, and computational systems to extract information from data. It uses analytics and machine learning to investigate hidden knowledge in structured and unstructured data sets, describe and visualize them, make predictions, and support decision making. Applications cover practically all possible areas of society, including Health, Engineering, Logistics, Economics, and Marketing. The Data Science research stream of MagIC is devoted to research and innovation in this vast and popular area, with the ultimate goal of using existing or novel "intelligent" methods to improve the health and wellness of human beings and their social and economic state.

The Data Science research stream relies on studying, developing, and strengthening Machine Learning for most of its scientific and applied research endeavors. Our mission focuses on promoting scientific excellence and innovation by performing high quality, internationally recognized research in the wide range of studied topics. We aim at establishing first quality partnerships with top internationally renowned universities and research centers and recognized collaborations with private and public organizations. We also have the mission of pursuing the dissemination of knowledge and scientific culture in society.

We have two different, but strongly related, main objectives: the study of the theoretical foundation of some of the most known and powerful machine learning algorithms, with the objective of producing improved versions of those algorithms; and the application of those improved versions to complex real-life applications, with the objective of outperforming the state-of-the-art results. These objectives are particularized into the following, more concrete goals:

• Developing novel, more powerful, and efficient versions of genetic programming;
• Developing new versions of artificial neural networks, including the ones generated employing neuro-evolution, to overcome the limitations of the current neural systems;
• Studying the overfitting phenomenon in-depth, both taking into account machine learning algorithms, feature selection, and regularization techniques, and considering data quality;
• Contributing to the area of interpretable AI by investigating new ways of generating readable machine learning data models for the different studied applications;
• Contributing to the area of Health, with particular reference to the study, understanding, and counteracting of diseases such as cancer, covid-19, cardio-vascular diseases, degenerative diseases like Parkinson's disease, and rare diseases;
• Contributing to society by preventing and counteracting social issues such as poverty, over-indebtedness, and various types of addiction (addiction to drugs, gambling, etc.);
• Contributing to areas that can help well-being and sustainability, such as forestal and maritime sciences, energy, civil engineering, and education.