Language technology is a collective term for various technologies that use natural human language as data.
There are mainly two types of language technology: speech technology and natural language processing (NLP).
We work mostly with processing of natural language, which involves interpreting semantic content, context and sentiments in large amounts of text, as well as facilitating dialogue between man and machine. We also develop a methodology for anonymizing sensitive information in text.
Development of language technology models from modest amounts of training data is a focus area for us. It is particularly relevant for language technology for Norwegian, since much of the international development in the field takes place for English.
We are also increasingly using language technology to analyze data that in principle does not come from languages, such as logs or data from industrial processes.