Different decision tree algorithms with comparison of complexity or performance

Question

I am doing research on data mining and more precisely, decision trees.

I would like to know if there are multiple algorithms to build a decision tree (or just one?), and which is better, based on criteria such as

• Performance

• Complexity

• Errors in decision making

• and more.

Answer 1

Decision Tree implementations differ along these axes:

• Splitting criterion 

• Training model for regression (continuous variables) as well as classification (discrete variables)

• Technique to reduce over-fitting

• whether it can handle incomplete data or not

The major Decision Tree implementations are:

• ID3, or Iterative Dichotomizer, was the first of three Decision Tree implementations

• CART, or Classification And Regression Trees is often used as a generic acronym for the term Decision Tree

• C4.5 is used commonly because it:

1. accepts both continuous and discrete features

2. handles incomplete data points

3. solves over-fitting problem

Hope this answer helps.