Back to All Events

UCL Data Science Society w/ Sia Togia

Meet Sia at the UCL Data Science Society where she will be talking about Extracting facts from biomedical literature using different machine learning methods.

Extracting structured facts from the biomedical literature is an important component in AI-based drug discovery. In this talk, I will cover the following topics: 1) What are biomedical facts and how are they best represented in a structured form?, 2) What are the different algorithms used for extracting biomedical facts from the literature?, 3) How do we ensure fact extraction methods remain interpretable? 4) How do we reliably evaluate a fact extraction algorithm's performance?


Sia is a Senior machine learning researcher specialising in Natural Language Processing. She obtained a PhD in 2015 from the University of Cambridge and previously an MSc in Speech and Language Processing and an MPhil in Linguistics (from the University of Edinburgh and University of Cambridge respectively). For the last five years, she has been working in the tech industry focusing on information extraction, information retrieval and knowledge representation.



Biomedical relation extraction with pre-trained language representations and minimal task-specific architecture.