Feed Annif a controlled vocabulary (subject headings, thesaurus or classification) and existing metadata - Annif then knows how to assign subjects for new documents!
Annif is built upon a combination of existing natural language processing and machine learning tools including Maui, fastText and Gensim. It is designed to be multilingual and it can support any subject vocabulary (in SKOS or a simple TSV format). It can be used either via a command-line interface or a microservice-style REST API. It also comes with a mobile web app that can analyse physical documents such as printed books.
Check out the Annif REST API for accessing Annif as a service!
The annif-users mailing list and web forum is available on Google Groups. This list is used for general discussion about Annif, asking for help, and announcements for new versions. All messages are public and anyone is welcome to join!
The SWIB18 presentation explains the basics of Annif and shows examples of where it is being used.
The video below demonstrates the first prototype of Annif which is no longer being maintained.