Introduction
IntroductionΒΆ
This book gives an example of the use of the Python package semanticlayertools [1], developed in the context of the project ModelSEN, to detect clusters of common research topics at the example of the Max Planck Society.
It makes use of the Research Organization Register to create a corpus based on the DimensionsAI database and then detects temporal clusters of common cocitation using the Leiden algorithm. For found clusters above a given node number, reports are generated displaying most common authors, affiliations and topics, generated using the Textacy framework.