Lecture: 'Extraction of Narratives in Text and other NLP Work with a Focus on European Portuguese'

Different types of textual sources contain narratives referring to their participants, events, when they occurred, where they occurred, and how these elements relate to one another. In this lecture, I will describe the ongoing lines of work to automatically extract narratives from text using information extraction, NLP, and linguistics. The result of narrative extraction is a formal structured representation that can be used for different types of visualization and other forms of representation or manipulation. This work mainly stems from the Text2Story and StorySense projects, where computer scientists and linguists work together to address these challenges. Our goal also includes increasing resources available for Portuguese (datasets, models, translators, etc.), especially for its European variant.

About the speaker

Alípio Jorge is a Full Professor at Universidade do Porto and researcher at INESC TEC. He coordinates the Laboratory of Artificial Intelligence and Decision Support (LIAAD). His research career has focused on machine learning, including inductive logic programming, frequent pattern extraction, recommendation systems, and natural language processing. He launched several master’s programs and an undergraduate degree in AI and Data Science. He served as Portugal's Sherpa to the European Commission and was responsible for the first version of the Portuguese Artificial Intelligence Strategy document.