The GESIS Methods Hub advances social science research by facilitating the sharing, search, and computational replication of methods and tutorials for analyzing digital behavioral data (DBD).
Given the diverse and heterogeneous nature of DBD, complex pipelines of computational methods are required, ranging from tasks like data collection, preprocessing, and cleaning to cutting-edge analysis and the application of AI.
The Methods Hub strives to empower social scientists in integrating state-of-the-art methods from computer science that may involve techniques from AI, network analysis, NLP, and LLMs.
To this end, the Methods Hub extends the three main components of the previous GESIS Notebooks service and other related efforts:
- a Place enabling users to publish and discover content in a FAIR way,
- a collection of content widely following the paradigm of literate programming (in the form of notebooks) including computationally reproducible analysis pipelines as well as related tutorials, and
- execution backends, such as Binder and Jupyter, supporting computational work.
The Methods Hub aims to enhance the reproducibility, reusability, and citation of research related to DBD. It is tailored to the needs of scientists who wish to:
- find methods by providing access to a curated selection of methods,
- obtain guidance on how to use methods, benefiting from tutorials that provide insights into methods and help working with complex computational pipelines,
- increase the visibility and reusability of their own methods by registering and sharing methods and tutorials.