Going data driven: learnings from building a real time recommender system on a 100m user base
Spil Games is accustomed to using data. With our internally developed event tracking, we monitor and analyze the gaming behavior of our 100m user base across our domains and in our apps. But in order to go data driven we had to setup an infrastructure that is able to handle this big data in real time. Building a recommender system using Apache Spark & Scala was our first project to build expertise in this area. The lecture gives valuable insights on the challenges we encountered, solutions found, expertise needed and benefits gained from building this real time recommender system for our 100m user base and 10k game portfolio.
This lecture is powered by Anchormen.