The Interface between Data Science, Research Assessment and Science Support - Highlights from the German Perspective and Examples from Heidelberg University

Data Science

“…Data science is warmly embraced by more and more disciplines and domains in which it was traditionally irrelevant, such as law, history, and even nursing [Clancy et al 2014]. Its core driving forces come from data-intensive and data-rich areas such as astronomy [Borne et al 2010 [Siart et al 2015], media and entertainment [Gold et al 2013], Supply Chain Management (SCM) [Hazena et al 2014] and SCM predictive analytics [Schoenherr and Speier-Pero 2015], advanced hierarchical/multiscale materials Gupta et al 2015], and cyberinfrastructure [NSF 2007]. The era of data science presents significant interdisciplinary opportunities [Rudin et al 2014], as evidenced by the transformation from traditional statistics and computing-independent research to cross-disciplinary data-driven discovery combining statistics, mathematics, computing, informatics, sociology, and management.…”

Data Science: Nature and Pitfalls

“…A large proportion of Google searches on these keywords returns results that are irrelevant to their intrinsic semantics and scope, or simply repeat familiar arguments about the needs of data science and existing phenomena. In many such findings [5], [54], [17], [29], [2], [25], [50], [39], [40], [18], [55], [45], [49], [23], [36], [41], [48], [34], [35], [43], [21], [15], [53], [31], [33], big data is described as being simple, data science has nothing to do with the science of data, and advanced analytics is the same as classic data analysis and information processing. There is a lack of deep thinking and exploration of why, what and how these new terms should be defined, developed and applied.…”