Ajalooliste allikate avamine masinöppe vahenditega
Title in English | Unlocking historical sources with machine learning tools |
---|---|
Authors | |
Year of publication | 2024 |
Type | Appeared in Conference without Proceedings |
MU Faculty or unit | |
Citation | |
Description | Web services providing Large Language Models (LLM) have taken the world by storm over the past year, and humanity is still learning about their capabilities. This development is of particular interest in the field of history, where the study of texts described in different historical contexts is a central theme. What can already be done in history with the help of linguistic models, and how will this change the way we do research? This presentation will focus on the research on medieval inquisition protocols being carried out by the DISSINET project, with a focus on the computer-assisted generation of text-structured data (CASTEMO). Based on the results of the project, I will describe how the creation of structured data is carried out, how it can be automated using linguistic machine learning tools, and what results the first experiments with large language models have provided. The opportunities and limitations of large and small language models and how they can change historical research. |
Related projects: |