Data Science Meetup Hamburg

=== Doors open ===

=== Small intro ===

=== Talk 1 ===

End-to-End Machine Learning Stack Under the Microscope @ enfore Speaker: Miroslav Zoričák, Nils Grabbert

“With history as a guide, our key observation is that most of the effort in industrial ML applications is not spent in devising new learning algorithms or models but is instead spent in other areas that are especially in need of better tools and infrastructure: data preparation, feature selection and extraction, and productionization” [ https://arxiv.org/pdf/1705.07538.pdf - Infrastructure for Usable Machine Learning] We will start with a zoom from the big picture of Machine Learning to the problems we experience first-hand at Enfore. How we enable data scientists to build their own production-quality data products without requiring a large team of experts in big data, or distributed systems, and without understanding the latest hardware. How to assemble an end-to-end Machine Learning stack. What approaches exist and how we evaluate them. We will finish by presenting our approach to the end-to-end learning and prediction and maybe there will even be a demo.

=== Talk 2 ===

Data Science for Online Marketing @ OTTO

Speaker: Dirk Petzoldt

Following the success of launching our own dynamic attribution model in 2012, OTTO Business Intelligence developed many machine learning models supporting our online marketing deliver relevant online adds effectively on large scale. Get an overview on what we work on including some of our learnings and best practices.

Dirk is a trained data scientist and leads OTTO’s BI department for online marketing and search.

=== Networking ===