Online Applied Data and Analytics in partnership with Product Madness

We are pleased to announce our very first meetup event. We are excited this first virtual event will be in partnership with Product Madness.

The great thing about taking it virtual is that you can ask questions throughout the session - in between each talk we will allow some time for the speakers to answer some of your questions.

ZOOM LINK. Join meeting:

https://aristocrat.zoom.us/j/92757574249

----------------------------------------------------------------------------------------------------
Evening Itinerary:

6:00 pm - 6:05 pm - Daniel Day, Consultant at Chi Square Analytics -https://www.linkedin.com/in/daniel28day/

Introduction for the evening

6:05 pm - 6:25 pm - Aleksandrs Gehsbargs, Data Science Tech Lead, Product Madness
www.linkedin.com/in/aleksandrs-gehsbargs-a8836932/

Bayesian bootstrap and AB testing:

The goal is to discuss how to make conclusions from AB tests. There are many inferential procedures in statistics, and the question stands about the most appropriate one for given use-case.

From the business perspective we should use an approach which gives clear and digestible conclusions. We should tackle these issues facing challenges - distributions of data have heavy tails, in a free-to-play world proportion of payers is low as compared to player population and distribution of data is not approximated well enough with parametric models.

Promising way to deal with most of the statistical issues and provide business with flexible conclusions is bootstrap. In my talk I will describe problems we face when we AB test features, will explain what is bootstrap and how we use it.

6:25 pm - 6:45 pm - Davide Anastasia, Expert in Data, Analytics and Engineering
https://www.linkedin.com/in/davideanastasia/

Davide will discuss how leveraging modern cloud data warehouses (like BigQuery or Redshift) can lead to shorter time to market for analytics of any complexity.

Using the ELT approach, modern cloud data warehouses become at the same time data storage and processing engines, where historically you would have needed external additional tools.

6:45 pm – 7.05pm - Dr. Guillermo G Schiava D'Albano, Team lead Nordic Team, Databricks
https://www.linkedin.com/in/g-schiava/

Lakehouse - what it is and how can transform your company:

Traditional EDW are good to store structure data, with a high cost per GB. Traditional Data Lakes on the cloud are good to store unstructured data, not supporting classic ACID operations found on EDW. At the same time Data lakes have lower cost per GB than a traditional EDW.

AI usually requires the combination of different sources structured and unstructured.

The traditional way to solve the shortcomings of these technologies is to combine them with difficult to main ETLs. Lakehouse is a new open architect that combines the best elements of EDW and Data Lakes. On top of open standards that allows costumes to never be vendor lock.

7.05 - 7:30 - Q&A with Panel

Please remember to register to receive the webinar link :)

---------------------------------------------------------------------------------------------------
ZOOM LINK. Join meeting:

https://aristocrat.zoom.us/j/92757574249

----------------------------------------------------------------------------------------------------