Events archive

Helwig Hauser

University of Bergen, Norway;

Abstract: Numerical simulation has become indispensable in many fields of science and applications, including engineering, much of natural sciences, and many more. Typically, the computed data are rich and extensive with commonly detailed spatiotemporal resolution and numerous computed variates – we also appreciate a fascinating trend towards ensemble simulation, resulting in parametric and / or stochastic datasets. Visual data science, based on the integration of analytical / computational means with interactive visualization, offers scientists as well as practitioners to better exploit the rich deliveries of their numerical simulation approach. Particular challenges include the increased dimensionality of the data, the associated semantics, which often required a dedicated treatment of different aspects of the data. In this talk, we review visualization research that was done in tight collaboration with domains such as automotive engineering and Earth sciences. We see examples for exploiting modern modeling approaches and we recognize the potential of visual data science to facilitate an exchange between expert users and their challenging data.