The eld of scientometrics is concerned with measuring and analyzing the impact of science in its broadest sense. All but the most basic measures require a substantial amount of data retrieval, cleaning, and integration, and therefore are performed semi-automatically on relatively small subsets of data. Furthermore, many of the more interesting and advanced questions in the eld of scientometrics require a substantial amount of domain modeling before any measures can be applied. Examples of such questions include studies on how research elds evolve over space and time and what it means for a research eld to grow, shrink, or radiate into other elds. In conjunction with machine learning techniques such as topic modeling, Linked Data and Semantic Web technologies are well suited to form the underpinning of modern scientometrics.