LLM-Mining Pre-Stemmatological Philological Literature

Abstract

The current article outlines a new research avenue for the analyses of literature from the time before the advent of the stemmatic method in the nineteenth century using large collections of digitized images and texts of historical philological works. The main aim is to understand the dynamics behind the processes leading to the invention of said method. The proposed steps are object recognition (image analysis) with textual clues and relation extraction (text mining). Proof-of-concept-level experiments demonstrate the applicability.