Workshop: Applied Swedish NLP (SLTC 2020)

The goal of the workshop is to highlight and discuss synergies and common challenges, which can form the basis of new collaborations and projects, or contribute to new research topics. 

The workshop will start with presentations on current initiatives in applied Swedish NLP. To the second part, we invite participants to present. The projects that will be presented in the first part all share the overall aim of developing and making available language models and data for Swedish applications, including investigating the various means of distributing these in legally, ethically, and technically permitting ways. 

Project partners:
National Library of Sweden, Peltarion, Region Halland, Sahlgrenska Universitetssjukhuset, Folktandvården Västra Götalandsregionen, Recorded Future, Talkamatic, Språkbanken Text at Gothenburg University, Gavagai, SKR, Swedish Tax Agency, Swedish Public Employment Service, Swedish Agency for Economic and Regional Growth, Luleå University of Technology

Agenda

13.00-13.05 Intro 
13.05-13.30 Swedish Language Data Lab 
13.30-13.50 Swedish Medical Language Data Lab 
13.50-14.10 Language Models for Swedish Authorities   
14.10-14.30 SuperLim: A Swedish Test Dataset for Language Models
14.30-15.00 Break (jointly coordinated with the other SLTC workshops)
15.00-16.00 Short use case presentations (max 3 min) from participants, and discussion

The speakers will give high-level presentations of the work that has been done so far in the above mentioned projects, focusing on targeted and identified needs and use cases. We look forward to your contributions in sharing your use cases and related challenges in the short presentations or discussions.

Confirmed speakers

• Staffan Larsson, Amelie Åstbom, José Cano Santin, Talkamatic
• Danila Petrelli, Simon Westlund, Fredrik Moller, Recorded Future
• Olof Mogren, RISE Research Institutes of Sweden
• Agrin Hilmkil, Peltarion
• Magnus Sahlgren, RISE Research Institutes of Sweden
• Aleksandrs Berdicevskis, University of Gothenburg

The moderator of the workshop is Magnus Sahlgren, Head of Natural Language Processing Group at RISE Research Institutes of Sweden. 

Preparatory questions

• What challenges do you see when it comes to applying NLP on your data/for your use cases?
• What resources (data, code, hardware, ...) would facilitate your specific application?
• Are there aspects of the application that you could and would want to collaborate with other participants/organisations on?

What should our use case presentation focus on? 

We would like you as participants to view this as an opportunity and a forum to bring up needs, challenges, and possibilities, which you would find valuable to discuss and collaborate with others on.

In the presentation, the focus should therefore be on challenges and needs that you have encountered, for example when working with a specific use case. With that said, this could be interpreted in broad terms. Contact us if unsure!

Registration 

Update 18 November: The registration is now closed. 

If you plan to present a use case, please notify us by sending an email to isabelle.johansson@ai.se.

Note that participation in the workshop will not require registration for the SLTC 2020 conference (but is of course encouraged).

Questions?

Contact isabelle.johansson@ai.se 

About the conference

University of Gothenburg is the organiser of this year’s Swedish Language Technology Conference (SLTC), on 25-27 November. The conference is arranged every second year.

SLTC will take place entirely online, including this workshop. 

Read more about the conference here