EliNor

EliNor: Electronic Infrastructure Resources for Research and AI Applications on Norwegian

EliNor aims to develop three cutting-edge digital resources to advance innovative linguistic research on understudied Norwegian language patterns (constructions), support the development of Large Language Models (LLMs) for Norwegian, and foster international collaboration.

• First, we create a professional interface for the Norwegian Blog Corpus (NBC), a large collection of authentic, up-to-date texts that addresses the lack of openly available data representing informal Norwegian.

• Second, we develop the Norwegian Constructicon (Språknett), a large searchable database of 2,000 thoroughly described and illustrated conventionalized and prominent language patterns (constructions) of Norwegian.

• Third, leveraging these resources, we build the first Learner Chatbot for domain-specific workplace Norwegian (SnakKIs), an AI-driven tool designed to help learners master these patterns in practical contexts.

Each resource has currently been developed to a level below Technology Readiness Level 6, and EliNor aims to advance them to levels 6-7 in 2026. Together, these innovative open-access resources advance Norwegian language research, enhance language technology, and promote student-centered, accessible language learning.