Automatic features generation and news from Spark Summit

First meetup after Spark Summit East, we will start with a short review of news from people who attended. Then we have a formal presentation about data science/machine learning.

1] News from Spark Summit East

• Format: Discussion + Presentation

• Discussion Leader(s): Jean-Francois Rajotte (https://ca.linkedin.com/in/jfraj) & Luis Da Costa (https://www.linkedin.com/in/luisdacosta)

• Summary: Overview of selected topics from Spark Summit. Anyone who attend are welcome to share what interested them.

1] Automatic feature generation & Amdahl's Law

• Format: Presentation

• Presenter: Simon Ouellette (http://www.simonouellette.com/)

• Summary:

Automatic feature generation & discovery has existed on paper for over 20 years, but has been computationally unfeasible until recently with the advent of Big Data technologies such as Apache Spark.

In the first part we will discuss the concept itself, and demo Nectarine: the most advanced application that currently exists in this area.

In the second part we will focus on algorithmic challenges relating to Amdahl's Law that were tackled during the development of Nectarine, in the context of a Spark framework.

Door open at 6PM Presentation starts at 6:30PM