Elvira Cámara Aguilera had the pleasure of carrying out a research stay at the School of Computing and Communications at Lancaster University (United Kingdom). During this research stay, she achieved the planned goals. Above all, the main achievement is the compilation of a specialised representative corpus in the domain of wine. This corpus is essential for carrying out all sorts of planned research and analysis from a terminological point of view, and for the extraction of terms and neologisms, allowing both corpus analysis and the implementation of NLP techniques. This corpus consists of specialised, semi-specialised and informative texts.
The second achievement is the annotation of this corpus and the manual extraction of terms and neologisms. The annotation is designed in a way that it follows a rule-based orientation in order to use it to train a large language model and carry out experiments on the automatic extraction of terms and neologisms from specialised corpus. Elvira Cámara Aguilera collaborated with Dr. Tharindu Ranasinghe and hopefully soon they will be able to share interesting results.
Moreover, the grantee and the host have designed a future research project to include more languages, and specially, low resources languages, orientated towards the extraction of terms and neologisms and its translation into those languages in the specified domain.
Finally, Elvira Cámara Aguilera would like to highlight how this STSM helped her widening her academic contacts. Furthermore, it gave her new insights for future projects and interesting collaboration.