Annif: DIY automated subject indexing using multiple algorithms

Manually indexing documents for subject-based access is a labour-intensive process. This paper describes Annif, an open source tool and microservice for automated subject indexing developed by the National Library of Finland. After training it with a subject vocabulary and existing metadata gathered from bibliographic databases, Annif can be used to assign subject headings for new documents. The current version is based on a combination of existing natural language processing and machine learning tools.