Today we lack the computational tools for handling semantic changes in language for the use in text-based humanities and social sciences. By handling these changes automatically, and at a large scale, we open up our textual resources (historical and modern) to researchers from a wide range of fields. We can thus facilitate studies of our contemporary and historical societies by identifying and eliminating linguistic barriers caused by language change and variation.

This program has two main aims, firstly to develop corpus-based methods for detecting semantic change (over time) and variation (across social groups and media types). This will create general tools for the study and detection of language change at large-scale and directly benefit historical linguistics and lexicography. Secondly, we will collaborate with researchers from social sciences, gender studies, and literature to answer their research questions. We will together develop methods and research methodology for their specific needs.

The results of the program will advance research both in terms of computation tools and technology for Swedish language data and assert Sweden's leading position in a rapidly growing field, but also contribute to Swedish excellence in text-based humanities and social sciences. The results will benefit the public as they access and interpret historical text, and any researcher that wishes to track concepts over time without manually finding and accounting for language change.