BeScala Meetup: Spark - Talk, Workshop and Hacking

8 March 2017 - 19:00 tot 22:00



Interactive exploration of language classification with Spark, Scala and the Spark Notebook

By Gerard Maas - R&D Lead at Kensu.io

In this interactive presentation, we will explore how we can create a simple language classification model from scratch.

We will be using an interactive format where attendees can follow along with small code challenges that sum up to form our first model. From there, we will explore techniques on how to measure and improve accuracy. You decide the level of participation: from lay-back and listen to heavy keyboard punching. In the process, we will learn about Apache Spark and its data processing capabilities using Scala. Preparation instructions will be sent before the meetup.

How a small company tackles big data at the speed of light

By Niek Bartholomeus - Data engineer at MeetMatch 

During the first part of the presentation we will continue with the interactive format and show how the language classification model that was just created by Gerard can be brought to life. First in batch mode, where we will use Spark to detect the language of a large text, and then in interactive mode where we will use a simple Play/AngularJs application to detect the language of a text that is entered by the user.

During the second part of the presentation I will present a "real life" story of how we used a similar toolset during the past 12 months to build a platform that is capable of finding the core activities of companies based on the information form their websites and use this information for intelligent matchmaking at business speed dating events. I will dive a bit deeper into how we made use of natural language processing, scala and spark to process lots of text in a highly iterative manner and talk about all the hurdles we have had to overcome (or that still exist) to get at where we are now.

Registration via this link.