A Multilingual NLP Framework for Offshore Installations

This paper presents a novel multilingual NLP framework tailored for Brazilian offshore operations, combining data engineering, alignment techniques, and information extraction to offer safety recommendations based on a decade of Brazilian offshore experience. NLP models provide platform staff with relevant historical incident lessons, integrated into a microservice architecture for streamlined access. The paper also discusses future work using LLMs for improvement.