Write short note on Decision Tree based Classification Approach

 

  • Training dataset should be class-labeled for learning of decision trees in decision tree induction.
  • A decision tree represents rules and it is very a popular tool for classification and prediction.
  • Rules are easy to understand and can be directly used in SQL to retrieve the records from the database.
  • To recognize and approve the discovered knowledge acquired from decision model is a crucial task.
  • There are many algorithms to build decision tree:
    • ID3 (Iterative Dichotomiser)
    • C4.5 (Successor of ID3)
    • CART (Classification and Regression Tree)
    • CHAID (Chi-square Automatic Interaction Detector)
Decision Tree representation
  • A decision tree classifier has a tree type structure which has leaf-nodes and decision nodes.
  • A leaf node is that last node of each branch and indicates the class label or value of a target attribute.
  • A decision node is the node of a tree which has leaf node or sub-tree. Some test to be carried on each value of decision node to get the decision of class label or to get next sub-tree.
Decision Tree represents for play tennis