Global Azure Bootcamp: Talk on HDInsight and Recommendation Engines

Not more than a week after the 50-year anniversary of Moore’s Law, the 2015 edition of Global Azure Bootcamp is taking place at more than 183 locations around the world. Speak about global computing at scale! It’s a truly amazing effort driven by the local Azure communities across the planet — and I have been fortunate enough to participate in the Copenhagen camp.

Ever since I started grad school in Madrid, I have been very interested in statistics and quantitative methods in general. With Microsoft I have dived even more into the world of big data and machine learning — topics on which I have given a ton of presentations ever since I started as a Tech Evangelist.

At the Global Azure Bootcamp I gave a presentation on how to build your own recommendation engine by using Hadoop in Azure through the service named HDInsight. The case is to build a song recommender that can suggest new songs to users based on the songs they have listened to in the past and the songs that other users have listened to (i.e., user-based collaborative filtering). Essentially, the case is to build what Spotify, YouTube and other popular media services are doing.

I recorded the presentation and uploaded it to YouTube. Check it out below. Also, check out the PowerPoint presentation below the YouTube video if your are particularly interested in the slides used in the talk. If you have any questions to this topic, do not hesitate to drop a comment. I will try to be quick at getting back to you! 🙂

Here are the slides: