NACSOS is a django site for managing collections of documents, screening or coding them by hand, and doing NLP tasks with them like topic modelling or classifiation.
It was built for handling collections of scientific document metadata, but has extensions that deal with twitter data and parliamentary data.
It currently contains many experimental, redundant or unsupported features, and is not fully documented.
The part that deals with topic modelling is a fork of Allison J.B Chaney’s tmv repository. It extends this by managing multiple topic models and linking these with various document collections.
NACSOS is research software produced by the APSIS working group at the Mercator Research Institute on Global Commons and Climate Change (MCC), and some parts of the repository are instution specific. We are in an ongoing process of generalising, and documenting.
Refer to the documentation for a (partial) guide to using the app.
If you use NACSOS in academic work, you can cite it as
Max Callaghan, Finn Müller-Hansen, Jérôme Hilaire, & Yuan Ting Lee. (2020). NACSOS: NLP Assisted Classification, Synthesis and Online Screening. Zenodo. http://doi.org/10.5281/zenodo.4121525