Distributed Machine Learning with Apache Mahout - Dzone Refcardz

Machine learning algorithms, in contrast to regular algorithms, improve their performance after they acquire more experience. The "intelligence" is not hard-coded by the developer, but instead the algorithm learns from the data it receives. A supervised learning task is a task in which the testing data is labeled with both inputs and their desired outputs. These tasks search for patterns between inputs and outputs in test data samples, determine rules based on those patterns, and apply those rules to new input data in order to make predictions on the output. Classification and regression are examples of supervised learning tasks.