The main objective of this project was to develop scalable and flexible methods for author disambiguation and to significantly improve author identification within dblp and zbMATH, the two world leading scholarly databases for computer science and mathematics, respectively. Both databases serve as prototypical data repositories in two different scientific fields, offering heterogeneous metadata, including such of historical value.
The cooperation with international partners, namely with the operating institution of the Russian mathematics portal www.mathnet.ru, provides a unique opportunity to develop methods for author disambiguation for one very important but extremely difficult language in this context. The results of this project will be of interest to digital libraries in general.
The project was conducted together with two project partners (see below) and funded by the Leibniz Association within the scope of the Leibniz Competition.