Showing posts with label Information Theory. Show all posts
Showing posts with label Information Theory. Show all posts

Thursday, 28 January 2010

Machine Learning - Decision Trees and Entropy

Decision Trees


If anyone requires further explanation, please get in touch with me (or anyone else for that matter)!

In machine learning, it can be desirable to come up with meaningful if-then rules in order to predict what will occur in the future. For example, "if this and if that then this will probably happen". Decision trees can be built automatically, which can then be used to come up with these if-then rules.

A decision tree is used to investigate huge amounts of data and come up with the most probable outcomes. Each condition is an internal node on the tree. Each outcome is an external node.