Thales Bertaglia

about me

I am a PhD Candidate at Maastricht University, working jointly with the Institute of Data Science and Studio Europa. My research focuses on using NLP and other computational methods to understand the influence of content creators and content monetisation on social media. I am also part of the Computational Social Media research group. I am supervised by Catalina Goanta, Mathieu Segers, and Michel Dumontier.


From Jun/2019 to Dec/2019 I was an Applied Data Scientist intern at Microsoft’s Search Technology Center (Munich). I was a member of the Intelligent Search team, and worked on approaches for improving the relevance of search results for users of Microsoft Teams.


From Jul/2018 to Jan/2019 I was a Speech Scientist Intern at Amazon Research (Cambridge) at the Natural Language Understanding team. My internship project was focused on multilingual learning for low-shot intent classification.


From Sep/2017 to Jun/2018 I was a Research Student in Natural Language Processing at the University of Sheffield, under the supervision of Lucia Specia. My research was focused on exploring the use of acoustic information to improve the performance of multimodal NLP tasks – focusing particularly on Machine Translation.


I obtained my M.Sc. in Computer Science and Computational Mathematics at the University of São Paulo, under the supervision of Maria das Graças Volpe Nunes. My dissertation topic was Text Normalisation of User-generated Content for Portuguese. I worked on text normalisation methods based mostly on deep learning architectures and word embeddings similarity. As part of my research, I developed a free and open-source tool called Enelvo – a flexible text normaliser for user-generated texts written in Portuguese. I also built an annotated corpus of non-standard words in different domains.


I took my bachelor’s degree in Computer Science at the Western Paraná State University, with one year spent abroad at Katholieke Universiteit Leuven, in Belgium. During this period, I worked on various topics – including computational epidemiology, ensembles of neural networks, and sentiment analysis. Upon graduation, I was awarded the prizes of Best Computer Science Graduate of 2015 and Outstanding Student in 2015, granted by the Brazilian Society for Computing.