Data modelling and visualisation of a digitised manuscript text enhanced through linked data: from text acquisition to data exploration

Abstract

The digitisation of cultural heritage leads to multiple benefits, including the preservation of artworks, the in-depth computational analysis of their characteristics, and their accessible dissemination to a wide public with the application of standards that promote interoperability and reuse of the resources to which they are linked. The implementation of a digitisation project requires a methodical and articulated approach and the application of a workflow specific to the type of cultural object involved. In the case of the digitisation of a text contained in a medieval manuscript, the digitisation paradigm consists of four basic steps: analysis and acquisition of the text (from source to transformation into machine-readable format); modelling of the collected data; standardisation and serialisation of the textual data; and visualisation of the processed data. This thesis describes the digitisation paradigm applied to the text of the 4th book of the manuscript BnF italien 590: the first prose vernacularisation of the Aeneid based on a Latin reduction of the ancient poem. First the text was acquired from images of the manuscript by applying HTR techniques; in the second phase the text was encoded in XML format by applying the TEI guidelines; subsequently a semantic ontology was designed to represent the characteristics of the text and its contents using some of the ontological vocabularies most in use in the field of cultural heritage, enriching the encoded text with data linked to authorithy files thus expanding the knowledge base of the digitised text; of this ontology, a serialisation in RDF schema has been produced; all the data collected and modelled have been rendered graphically with interactive visualisations implemented from scratch and some appropriate visualisation tools; finally, is proposed the graphic design of a web environment in which to explore and analyse the collected and modelled data of the digitised text and its peculiarities.