In this chapter we learned that classification and regression are supervised learning problems, which require labeled data. Predictor variables and output variables should be defined to come up with a model during the training phase.

We also saw that supervised learning can be achieved using different techniques, namely model-based, regression-based and tree-based techniques.

Regression can be divided into two categories based on the outcome of the algorithm, that is linear regression and logistic regression.

We saw that text classification is an important application and this is explained using the Naïve Bayes algorithm.

In the next chapter, we will discuss the recommendation techniques using Apache Mahout.