A Digital Edition between Stylometry and OCR:
Simone Rebora · Textual Cultures · 2019
This article presents the digital edition of Robert Musil’s work (Klagenfurter Ausgabe) and its role in a digital humanities project aimed at reconstructing Musil’s activity in the WWI journal Tiroler Soldaten-Zeitung. First, the article reviews the ways in which the compu- tational methods of stylometry are applied to attribute the anonymous texts published in the Klagenfurter Ausgabe. Second, it explores how optical character recognition (OCR) soft- ware is employed to expand the corpus. At the core of this methodology two machine learn- ing algorithms are trained and revised using the transcriptions of the Klagenfurter Ausgabe, to reach an accuracy of about 99.9% in the digitization of the Tiroler Soldaten-Zeitung texts. The work of this project offers not only the possibility of expanding stylometric analysis to the whole journal, but also of improving the transcriptions of the Klagenfurter Ausgabe.