Roman medieval documents meet machine learning An experience of research and teaching.

This contribution is dedicated to recounting our experience of researching the doc- uments of medieval Rome, initiated in 2016 thanks to In codice ratio2. In codice ratio is a research project conceived by Paolo Merialdo that aims to develop innovative methodologies and tools for the analysis and study of manuscript sources. In its ini- tial phase, the project involved the collaboration of the Department of Humanistic Studies of Roma Tre University and the Vatican Apostolic Archive in the appli- cation of machine learning techniques to create software capable of automatically recognizing and transcribing manuscript sources from the medieval period.