This event has ended. Visit the official site or create your own event on Sched.
Thank you for participating in Big Data Tech 2018! Access conference presentation decks here.
View analytic
Tuesday, June 5 • 2:15pm - 3:00pm
Scalable Feature Engineering with Dask & Event-based Count Data

Log in to save this to your schedule and see who's attending!

Feedback form is now closed.
One of the biggest challenges in the creation of high-performance machine learning models continues to be the process of feature extraction and feature engineering.  The hurdles to building features that are informative for modeling become particularly significant when the incoming data includes event counts that are of high cardinality, and/or when the data changes over time.  These scenarios occur widely in real-world statistical learning problems.  Inspiration for strategies to create features that represent the underlying problems to be solved can be drawn from the PPC advertising industry, and extended to related challenges.

avatar for Greg Hayes

Greg Hayes

Director Data Science, Ecolab
Greg is currently a Director of Data Science at Ecolab collaborating with strategic business partners and external stakeholders to unlock value from data, and to define an architecture under which developed models may be deployed to production in a cloud computing environment. Previous... Read More →

Tuesday June 5, 2018 2:15pm - 3:00pm
P1838 Normandale Partnership Center, 9700 France Ave So, Bloomington, MN 55431