Using a comprehensive list of job titles, we propose a framework to automatically generate job descriptions in Italian. This synthetic data is then used in a Large Language Model to detect inclusive language in job postings. Finally, we compare the results of this synthetic dataset with real data. Our study demonstrates that the data format and prompting method signif-icantly impact performance. Additionally, we identify limitations and key considerations for unifying synthetic data with real data for fine-tuning purposes. We also propose improvements to the framework and provide guidelines for effectively integrating these two types of data. The novelty of our work is generating and integrating synthetic data due to the scarcity of annotated Italian job descriptions, thereby improving the training of Large Language Models (LLMs) tailored specifically for Italian.
Synthetic Data for Identifying Inclusive Language (Case Study: Job Descriptions in Italian) / T. Romano, F. Mohammadi, P. Ceravolo (PROCEEDINGS OF THE ... IEEE INTERNATIONAL CONFERENCE ON CYBER SECURITY AND RESILIENCE (CSR)). - In: Proceedings of the 2024 IEEE International Conference on Cyber Security and Resilience (CSR) / [a cura di] S. Shiaeles, N. Kolokotronis, E. Bellini. - [s.l] : IEEE, 2024 Sep. - ISBN 979-8-3503-7536-7. - pp. 737-742 (( convegno IEEE International Conference on Cyber Security and Resilience, CSR tenutosi a London nel 2024 [10.1109/csr61664.2024.10679398].
Synthetic Data for Identifying Inclusive Language (Case Study: Job Descriptions in Italian)
F. Mohammadi
Secondo
;P. CeravoloUltimo
2024
Abstract
Using a comprehensive list of job titles, we propose a framework to automatically generate job descriptions in Italian. This synthetic data is then used in a Large Language Model to detect inclusive language in job postings. Finally, we compare the results of this synthetic dataset with real data. Our study demonstrates that the data format and prompting method signif-icantly impact performance. Additionally, we identify limitations and key considerations for unifying synthetic data with real data for fine-tuning purposes. We also propose improvements to the framework and provide guidelines for effectively integrating these two types of data. The novelty of our work is generating and integrating synthetic data due to the scarcity of annotated Italian job descriptions, thereby improving the training of Large Language Models (LLMs) tailored specifically for Italian.File | Dimensione | Formato | |
---|---|---|---|
Synthetic_Data_for_Identifying_Inclusive_Language_Case_Study_Job_Descriptions_in_Italian.pdf
accesso riservato
Descrizione: Conference Paper
Tipologia:
Publisher's version/PDF
Dimensione
871.5 kB
Formato
Adobe PDF
|
871.5 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.