Data Science @ Regensburg: Corona Edition

Dear all,

We are very pleased to announce a special edition of our Data Science Meetup that is dedicated to looking at COVID-19 from a data science perspective. Did I say "a" data science perspective? Well, it's certainly not just one ...

This is again a double bill but with a twist as you will be drawn into the session a lot more than at previous meetings. So hold on tight, get ready and here we go ...

We will start with Jimmy Lin (yes, THAT Jimmy Lin!) who will join us to answer YOUR questions. For many years Jimmy has been a key figure in a variety of areas ranging from search to natural language processing. His team at the University of Waterloo was the first to release a state-of-the-art search engine, Neural Covidex, that would allow search over the recently released COVID-19 Open Research Dataset (CORD-19). He is also actively involved in running the TREC-COVID challenge that has just moved into the next round. We welcome Jimmy to answer questions you might have on any of these activities. He is happy to share insights, experiences and findings. Get ready and have these questions coming in. A chance not to be missed! For those of you new to the field, here are some links that get you started:

(1) TREC-COVID: https://ir.nist.gov/covidSubmit/index.html
(2) Press release on Neural Covidex: https://cs.uwaterloo.ca/news/new-ai-powered-search-engine-makes-it-easier-to-find-information-covid-19
(3) Rapidly Deploying a Neural Search Engine for the COVID-19 Open Research Dataset: Preliminary Thoughts and Lessons Learned: https://arxiv.org/pdf/2004.05125.pdf
(4) Jimmy on Twitter: https://twitter.com/lintool
(5) News just in (today): Waterloo tops the MS MARCO passage ranking leaderboard: https://microsoft.github.io/msmarco/

But that's not all! We are equally happy to welcome Christian Schuhegger, who most of you will have met at past meetings, and he will give us a hands-on tour using Jupyter Notebooks to equip us with enough knowledge so that we can all go away and do our own data science analysis of infection numbers, predicted reproduction rates etc.

We hope you will join us next Wednesday (Zoom details to follow next week). Details below ...

Best,
Udo / David / Bernd

Details:

Session 1:
Q&A with Jimmy Lin (University of Waterloo). Send us your questions and we will pass them on.

Session 2:
Christian Schuhegger (ex. 360T/Deutsche Börse Group, Data-Hobbyist) will talk about "Covid-19 Infection Dynamics"

Abstract:
DIY: Data-Sources, Forecasting, R-Value, Survival Analysis in Jupyter Notebooks

Bio:
Christian Schuhegger started his career as a physicist at CERN/Geneva, but soon thereafter changed into software development. From 2008 to 2018 he was responsible for the development and operation of the 360 Treasury Systems AG foreign-exchange trading infrastructure (acquired in 2015 by Deutsche Börse Group). His main focus over the past years has been low-latency high-throughput (GEO-scale) distributed computing. In 2017 he re-discovered his love for data, data-science, statistics and machine learning and describes himself as a data hobbyist. He is looking forward to have the honor to speak at Data Science @ Regensburg!