Patent & Scientific Information

In the Patent & Scientific Information (PSI) program unit, we develop and operate information services to support research, innovation and patenting processes. Core activities are the analysis, structuring and indexing of information from technical literature and patents. We develop innovative, AI-based concepts and methods in research projects to make this knowledge accessible and interlink it. This involves high data volumes of very heterogeneous provenance. In addition, we provide scientific and technical consulting services for our international customers.

This program unit was freshly establihed in 2022. The working areas include:

• Tasks for the development and operation of the information service STN in cooperation with our US-American partner CAS as well as projects for the evaluation and development of new methods and services
• Establishment of an innovative Patents4Science information infrastructure geared to the scientific use of patents
• Activities in the Bibliometrics Competence Network to build bibliometric databases and develop bibliometric analysis tools.

In Patents4Science, the core part of this program unit, we are dedicated to building an innovative information infrastructure geared to the scientific use of patents on the basis of knowledge-based approaches. The aim is to provide simple, efficient and sustainable access to patents and the knowledge they contain. The main target group are researchers in university and non-university research environments. We use the procedures and methods developed by us for the development and analysis of patent information to link heterogeneous sources of information and knowledge. By using text and data mining as well as machine learning methods, various innovative approaches for extracting specific information from patent documents have been evaluated and further developed to a proof of concept, e.g., to annotate patent claims or to identify references. In addition, we pursue the development of user-centered solutions in the context of collaborations with other disciplines.

The scientific publication system is one of the central sources through which the dissemination and validation of scientific knowledge are organized. Bibliometrics uses information about authors, journals, institutions, sponsors and topics that accumulates in the publication process and processes it into indicators, structural representations and key figures. For this purpose, the Competence Network Bibliometrics has built up a quality-assured in-house data infrastructure based on the databases Scopus (Elsevier) and Web of Science (Clarivate Analytics) and uses these to develop and refine analysis methods and indicators. Since the beginning of the first funding phase (2008), we have been involved in setting up the Competence Network Bibliometrics, which has now expanded to 22 consortium partners, with the role of hosting and development partner. Our activities include:

• Providing the central infrastructure and services for bibliometric research and
• analyses
• Development of standardized and quality-assured bibliometric databases based on the data products Web of Science from Clarivate and Scopus from Elsevier
• Performing data analysis and data management for the above data products.