Autoren:
Hajra, Arben
Pianos, Tamara
Tochtermann, Klaus
Zusammenfassung:
Author-related data, e.g., affiliations, collaborations or biographical facts, represents essential information in scholarly communication. For a comprehensive overview, scholars usually need to visit several sources to gather this information. In addition, the ambiguity of author names adds another level of complexity. Information about authors is scattered among authoritative sources such as libraries as well as open knowledge bases like Wikidata. In many cases, the data provided by these sources is highly structured by following linked data principles and semantic web technologies. Our approach uses this data - its identification and aggregation - to generate the most complete profiles for the authors. At the same time, this avoids the creation of further isolated silos of data and benefits from multi-stakeholder collaboration in data generation and curation. Furthermore, by analysing the research output through NLP, text mining and ML methods, an extended profile including the author's most important research topics is achieved.