Machine Learning with Spark

This is a really book to help people get started with Spark. It doesn't require much knowledge about machine learning nor Spark.The book goes through different examples of business needs and explains the process of constructing very basics machine learning systems form end to end.It's well written and with a clear and logical structure and easy to follow: it starts form data preparation and it ends with real-time machine learning with Spark Streaming.The code (mainly Scala, but also Java and Python) it's quite simple and explained in details.I would have liked more on scaling but it was a really good read: I was familiar with the machine learning concepts but not with Spark, this book helped me understand how to build machine learning systems in a very efficient way, so I would definitely recommend it.

I got the ebook , "Machine Learning with Spark" by Nick Pentreath from the Publishers. The book gives the foot print of Machine learning for Cluster/Cloud based applications. The basics of Spark and Scala are explained well in detail. The book gives new scope to Machine learning beyond the scientific application, especially to data driven utilities. Machine learning aspects like,data "classification" , "Regression" and "Dimensionality Reduction" are explained well with sufficient examples. But, "scalable" is not justified in any manner with in the examples/illustrations. The design aspects of data preparation and mining architecture mentioned in the book are very basic. Competitive comparisons like R could have been mentioned. Though Spark has the potential of dealing "Big Data", none of instances high lights it. All in all, the book is good for the people in cloud/cluster domains.