Working with Decision Trees
Jason Bell · Machine Learning · 2014
This chapter shows how decision trees work. The examples use Weka data-mining tool to create a working decision tree that will also create the Java code for you. Every tree is comprised of nodes. Decision trees always start with a root node and end on a leaf. The chapter explains where decision trees are used along with some of the advantages and limitations. It also shows how a decision tree is calculated manually. The chapter then demonstrates how to manually work through an algorithm with category values; the example walkthrough uses numerical data. Finally, it covers a lot of ground in a short space of time: putting an.arff file together to creating a classifier, and generating the Java code with Weka and testing it with more unclassified data. The chapter helps to work on a full project to create a working classifier based on the C4.5.