Jakob-Fugger-Zentrum
Italian Language Columns as Linguistic Linked Open Data
The project brings together semantic web technologies and Italian linguistics, with a particular focus on lexical semantics and discourse analysis. The project, which includes Prof. Christian Chiarcos (Applied Computational Linguistics, University of Augsburg) and Prof. Fabio Rossi (Italian Linguistics, University of Messina), aims to convert a corpus of Italian language columns (CronIT) into Linguistic Linked Open Data (LLOD), thereby enabling semantic web access through structured data semantics. The term language column refers to a series of articles addressing language-related issues – particularly in the areas of lexis and morphosyntax – published on a regular basis and over a longer period in the print and online media. In Italy, these texts are almost exclusively written by prominent linguists who are widely recognized as authorities on language issues. Given the long-standing role of these columns in shaping public discourse on language, they represent a vital part of Italy’s cultural heritage. This project seeks to make these texts accessible online for the first time, opening them up to a broader audience for purposes such as education and language consultation. The project pursues the following sub-goals:
- Expanding the digital indexing of the material by integrating multimedia content such as audio and video;
- Publishing the data in the Linked Open Data (LOD) Cloud and enriching it with semantic metadata to facilitate interlinking with external resources (e.g. online dictionaries and ontologies);
- Applying (semi-)automated annotation techniques using few-shot and zero-shot learning methods to conduct sentiment analysis, with a particular focus on epistemic, deontic, and evidential markers related to the linguistic topics discussed in the texts.