The Data Science Seminars are held in the scope of the MSc in Data Science from the Faculty of Sciences of the University of Lisbon.

Title: Why analyze biomedical text if we have data?
Speaker: Francisco M. Couto (LASIGE/DI-FCUL)
Date: December 18, 14h30
Where: https://videoconf-colibri.zoom.us/my/tjvguerreiro
Organized by: Tiago Guerreiro

Abstract: Our daily quest for knowledge usually starts in a search engine with a text query, not a SQL query. Text is still the main tool that we humans use to communicate and publish our findings and the context in where they were generated. Biomedical data repositories lack the flexibility to fully represent most of our hypotheses, theories, deductions, proofs, assumptions, observations and experimental evaluations that support the stored scientific facts. Moreover, absolute truth is not part of science, so biomedical text is crucial for enabling proper scientific inquiry and interpretation of facts for a particular context. However, text analysis is a tedious and time-consuming task for humans and a hard and prone to error task for machines. This talk will give a brief introduction to how text mining coupled with ontologies can address this challenge, using technologies that range from regular expressions to recent deep learning algorithms, and from natural language processing techniques to semantic similarity. The talk will end with an overview of the most recent contributions made by LASIGE researchers in this topic, namely in named-entity recognition and linking, relation extraction, question and answering, recommender systems, and multilingual international evaluations.

Bio: Francisco M. Couto is currently an associate professor with habilitation at Universidade de Lisboa (Faculty of Sciences) and a researcher at LASIGE. His main research contributions cover several key aspects of bioinformatics and knowledge management, namely in proposing and developing various text mining solutions that explore the semantics encoded in ontologies. Until November 2020, he published 2 books; was co-author of 8 chapters, 52 journal papers (41 Q1 Scimago), and 32 conference papers (10 core A and A*); and was the supervisor of 8 PhD theses and of 45 master theses. He received the Young Engineer Innovation Prize 2004 from the Portuguese Engineers Guild, and an honorable mention in 2017 and the prize in 2018 of the ULisboa/Caixa Geral de Depósitos (CGD) Scientific Prizes.