Automatic Feature Engineering with Driverless AI

Dmitry Larko, Kaggle Grandmaster and Senior Data Scientist at H2O.ai, will showcase what he is doing with feature engineering, how he is doing it, and why it is important in the machine learning realm. He will delve into the workings of H2O.ai’s new product, Driverless AI, whose automatic feature engineering increases the accuracy of models and frees up approximately 80% of the data practitioners' time - thus enabling them to draw actionable insights from the models built by Driverless AI.
The webinar will include:
- Overview of feature engineering
- Real-time demonstration of feature engineering examples
- Interpretation and reason codes of final models
Bio: Dmitry Larko is a Senior Data Scientist at H2O.ai. He has more than 10 years of experience in IT. He started with data warehousing and BI, now in big data and data science. He has a lot of experience in predictive analytics software development for different domains and tasks.
He is also a Kaggle Grandmaster who loves to use his machine learning and data science skills in Kaggle competitions.
www.kaggle.com...

[@iyangarsamayal9517]

This is one of the finest videos on this topic I have seen anywhere. In fact, I bet more people would have seen it if it was correctly titled as “Secrets of building Great Machine Learning Models from a Kaggle Grand Master”. Seriously this video should be required viewing for all data scientists graduating from code academies, boot camps and masters degree programs since it is so important and so beautifully explained. Thank you Dmitry Larko!

[@hackstack4933]

It was a good session, thanks Dmitry!

[@NikTuzov]

Thanks for posting this, I learned a lot!

[@victorpalacios1747]

Title is Automatic Feature Engineering. I wanted to learn how to do this in H20. This video does not cover this. The tutorials on the site should allow users to skip steps and get right to the topic they interested in, because we need to see how it all works before we use time trying to implement something we may not want.

[@vidaringa1]

Shouldn't WtdEncode be the same value within each category, because the formula only uses the mean of the whole data set and the mean of the category? So for category A, WtdEncode should be 0.765, right?