Numerous libraries and affiliated organizations are investigating automated techniques for generating metadata, and the role of AI is ever increasing. This workshop provides an overview of the open-source multilingual automated indexing tool, Annif ([https//annif.org https//annif.org), which can be incorporated e.g. into a library’s metadata creation system. We are also working on a new solution for automated bibliographic metadata extraction, BIBRA ([GitHub - NatLibFi/BIBRA: Tool for bibliographic metadata extraction · GitHubhttps://github.com/NatLibFi/BIBRA/\]), which we will present.
The target audience of this workshop are people working with metadata, automation or vocabulary services and knowledge organization systems in general. We welcome current Annif users as well as those who would be interested in automation of content description.
We will invite users around the world to showcase their systems and to share experiences and questions. We will also share the latest news and developments. During the workshop we will also cover aspects of using Large Language Models for metadata extraction as implemented in the BIBRA tool.
No previous experience with Annif or metadata extraction is necessary. However, familiarity with subject vocabularies (like thesauri or classification systems) and corresponding subject metadata is expected.
Mona Lehtinen (NLF): Introduction to Annif and the workshop
8:10-9:50 Annif use case presentations
Diana Slawig, Eva Seidlmayer (TIB): Annif use cases at TIB: from small vocabularies to large ontologies
Lynda Bunting (Getty Conservation Institute): AATA’s Implementation of Annif for Semi-Automated Indexing
Parthasarathi Mukhopadhyay (University of Kalyani): Automated Knowledge Organization in Koha: A Practical Implementation using the Annif Framework
MJ Suhonos (Toronto Metropolitan University): Wiki Core: Million-scale vocabulary indexing with Annif and Wikidata
[additional short 5 min use case presentations?]
9:50-10:10 break
10:10-10:50 New developments
Lakshmi Rajendram Bashyam (ZBW): Recent experiments for AutoSE at ZBW: What are the effects of data and vocabulary on the output of our models?
Lakshmi Rajendram Bashyam (ZBW): Combining thesaurus graph structure and LLM approaches for automated subject indexing
Osma Suominen, Juho Inkinen, Unni Kohonen, Mona Lehtinen (NLF): New Annif developments & BIBRA
Maximilian Kähler (DNB): Evaluating descriptive cataloguing and intended use of BIBRA at the DNB
10:50-11:00 Closing and discussion
Practicalities
The workshop will be held online using the Zoom platform.
If presenters give their consent, the workshop presentations will be recorded and the recordings and slides will be published online. The discussion part at the end will not be recorded.
Invitation to present use cases
The programme still has room for short use case presentations. If you are interested in presenting your own Annif use case, even briefly, you are welcome to contact the workshop organizers!