This study aims to outline knowledge translation as a method for practising interdisciplinarity in a domain-analytic context which uses new machine learning technologies to improve interoperability between knowledge organisation systems (KOSs).
Through conceptual analysis of topics from translation studies and natural language processing (NLP), a theoretical synthesis is performed which applies functionalist theories of translational action to how word embeddings can be used to increase interoperability.
Theories from translation studies and recent work in context can inform how information science approaches word embeddings and large language models (LLMs) as tools for furthering interoperability.
This method for knowledge integration puts concepts like interoperability in a new context and responds to debates about interdisciplinarity in the field of knowledge organisation by proposing a method using machine learning to explore the contexts of different vector spaces.
