Scholarly communication is facing a radical change with the introduction of Artificial Intelligence (AI) methods. A disruptive change of the whole system seems possible. Since publication is still the most important criterion for evaluation and promotion in academia, the individual pressure to publish is constantly increasing in this spiral. The question of "relevance" as a filter function for controlling the volume of publications is becoming more and more pressing, and an integer scientific practice requires knowledge of all "relevant" previous research results. At the same time, the differences between disciplinary cultures are not adequately reflected. Those who only write successful monographs instead of articles quickly fall through the cracks when evaluated according to metrics.

The interdisciplinary combination of library and information science expertise with mathematically based solutions for the effective use of AI offers enormous potential for supporting researchers in publishing. FAN integrates the current state of research and scientific practice in the fields of Data Science, Statistics, AI, Operations Research, Network and Graph Theory. The starting point for the project work is KOBV's digital databases of bibliographic information. Statistical and AI methods will be used to automatically determine the data quality of the publication databases. Methods for merging and supplementing the content of different publication databases will be investigated.

Appropriate metrics for articles and entities will be calculated based on large graphs. Finally, hypergraphs will be used to model and analyse publication data. As a result, the project explores the potential of AI methods for their application in libraries.