19 Apr module 21: Github, GDPR and research data management
- module 21: Github, GDPR and research data management
9:30 am - 5:00 pm
Mon 16/09: 9.30 am - 5.00 pm | S8 room 2.036
"Reproducibility in Biomedical Natural Language Processing: Challenges and opportunities" Rigor and reproducibility of biomedical research has been the topic of much debate in recent years. While the increasing complexity of computational environment creates new challenges for reproducibility, cases of replication failures lead to a lack of confidence in published findings. This lecture will review a few examples of reproducibility challenges in Biomedical Natural Language Processing. We will subsequently analyze these experiences to understand what challenges impact reproducibility and which methods can be leveraged to learn from them.
"Tools for Reproducibility & Collaborative Information Management: an introduction to Git."
Reproducibility is challenging in today's complex computational and collaborative environment. This tutorial will present version control, a systematic method for managing and tracking changes to documents, computer programs and other collections of information. After a brief overview of different version control methods and tools, a particular version control system will be presented into more details, with practical exercises to learn the basics of Git.
An interactive lecture on the general topic and a tutorial on git combined with a practical hands-on session.
No specific background required.
Aurélie Névéol is a Senior Staff Scientist at the Centre National pour la Recherche Scientifique (CNRS, LIMSI). She received an MSc in Linguistics in 2002 and a PhD in Computer Science in 2005. She has more than 10 years experience in biomedical Natural Language Processing Research and has addressed the analysis of biomedical text from the literature and from Electronic Health Records in French and in English. Recently, she has been focusing on clinical NLP for languages other than English. She has contributed to the development of representations of clinical information to support information extraction from EHR text, which can then be used for high throughput phenotyping. In the course of her work she has also contributed to the evaluation of research methods and workflows through her participation in the H2020 MiRoR project and international evaluation campaigns such as CLEF and WMT.
Venue: University of Ghent
Venue Phone: +32 9 264 52 39
Venue Website: https://www.flames-statistics.com/summer-school/venue/Address: