Summary

We began this chapter by introducing you gently to the rich and abundant world of machine learning algorithms and open-source tools which facilitate their application of large datasets.

We then moved on to practical tutorials during which we presented you with three different machine learning methods run on a multi-node Microsoft Azure HDInsight cluster with Hadoop, Spark, and RStudio Server installed. In the first example you learnt how to perform a logistic regression through the Spark MLlib module using the SparkR package for R with HDFS as a data source.

In two further tutorials, we explored the powerful capabilities of H2O-an open-source, highly-optimized platform for Big Data machine learning models run through the h2o package for R. We applied the Naive Bayes algorithm to predict the classes of the outcome variable and then we compared the achieved performance and accuracy metrics with two models generated by the Neural Networks and Deep Learning techniques.

In the final chapter...