Digitization of Historical Text - Erik Lenas
We begin by examining why Handwriting Recognition (HTR techniques), an AI-driven technology, is of great significance and its potential to transform humanities research. The process behind HTR technology and its remarkable advancements in conjunction with AI development in recent years will be highlighted.
The focus then shifts to the future beyond HTR, where we explore Natural Language Processing (NLP) for historical text. This AI field has its own unique requirements and creates new opportunities, especially within the research world.
Erik Lenas, Lead Data Scientist på Riksarkivet
To me data science is the perfect blend of theory and practice, analysis and creativity. In my work at the Swedish National Archives we use AI in the service of the humanities. Focus right now is on document analysis and recognition, turning millions of scanned documents into digital data. I'm also interested in how NLP can be applied to historical text. Handwritten text recognition together with language models adapted to historical text are going to revolutionize research into our written history, and it's super exciting to be a part of that revolution.