Table 3 Program flow of BuildTree

Require: NodeQueue NQ, TreeModel TM, Training record

(x,y) ∈D

1. For each T_curr∈NQ do

2. If JudgeLeaf(T_curr) is false then

3. bestSplit=FindBestSplit(T_curr)

4. T_curr→splitAtt=bestSplit→splitAtt

5. If bestSplit→splitAtt is category then

6. T_curr→leftAttSet=bestSplit→leftAttSet

7. T_curr→rightAttSet=bestSplit→rightAttSet

8. Else

9. T_curr→splitValue=bestSplit→splitValue

10. parationTrainingSet(T_curr→D, leftD, rightD)

11. remove(T_curr→splitAtt)

12. Create new nodes T_left, T_right

13. Initiate(T_left, leftD,Att)

14. Initiate(T_right, rightD,Att)

15. T_curr→left=T_left

16. T_curr→right=T_right

17. NQ.push_back(T_left)

18. NQ.push_back(T_right)

19. Else

20. T_curr→isLeaf = true

21. T_curr→label =y //y is the most common label