How to build a geolocated recommender using Spark ML, Cassandra and Akka

Presented by

Natalino Busa, Head of Applied Data Science at Teradata

About this talk

Natalino introduces a collection of machine learning techniques to extract insights from location-based social networks such as Facebook, demonstrating how to combine a dataset of venues’ check-ins with the user social graph using Spark and how to use Cassandra as a storage layer for both events and models before sketching how to operationalize such predictive models and embed them as microservices. In terms of data architecture this processing follows closely the SMACK stack. The proposed data-pipeline is effective at detecting patterns in the sequences of visited venues and recommend relevant venues to visit next, based on the user, and friends location's history as well as the venue popularity graph. Natalino Busa explains how these predictive analytics tasks can be accomplished by using Spark SQL, Spark ML, and just a few lines of Scala and Python code.

Related topics:

More from this channel

Upcoming talks (0)
On-demand talks (462)
Subscribers (6050)
You've got data. It's time to manage it. Find information here on everything from data governance and data quality, to master and metadata management, data architecture, and the thing that was just invented ten seconds ago.