Responding to the many challenges of Big Data means coming up with specific tactics and tools for specific problems, such as using Hadoop and the MapReduce framework for storing and processing very large data sets. Hadoop embodies one of the fundamental changes in Big Data, bringing the algorithms to the data instead of separating the analysis from the storage.
One specific tactic to bring the algorithms to the data is to embed analytics into the database. This paper presents a walkthrough, with code samples, of how to embed JMSL predictive analytics into an Oracle 12c database, using the Naive Bayes classifier algorithm and an example test data set. The complete Java code for this exercise is presented in the appendix.
Continue Reading Here