Project Description
A detailed overview of the research phases, corpus work, linguistic annotation, psycholinguistic experimentation and human-centered AI tool developed by HCAI4Gend.
The HCAI4Gend project investigates the use of gender-sensitive language in university institutional communication in Italy, Germany, Switzerland and the United Kingdom.
In the initial phase of the project, we analysed Italian legislation on transparency and the right of access, European Union legislation, and the national legal frameworks of Germany, Switzerland and the United Kingdom from a legal-linguistic perspective. In parallel, we examined guidelines on gender-sensitive language developed by Italian and international universities in Italian, German and English.
On this basis, we built a comparable corpus of institutional texts — regulations, statutes, calls, forms and contracts — available on university websites in the three languages. This was followed by a phase of linguistic annotation based on annotation schemes developed within the project.
This made it possible to carry out a systematic comparative analysis of the use of gender-sensitive language and supported the rewriting of passages that did not comply with existing guidelines, particularly in Italian institutional texts.
As a result, we produced a parallel corpus of original and rewritten texts in German, English and Italian. This corpus formed the basis for extracting sentences used in a psycholinguistic experiment designed to assess the processing difficulty associated with different linguistic strategies. It also supported the Natural Language Processing component of the tool for gender-oriented linguistic simplification.
The different research phases, integrated and closely interconnected, ultimately led to the development of an open-source software tool based on Human-Centred Artificial Intelligence. The tool is designed to detect and analyse the use of grammatical and social gender in institutional texts, and to generate contextualised linguistic solutions supporting gender-sensitive institutional communication.