Latest News

Monday, April 17, 2017

DECISION TREE

Decision Tree

What is Decision Tree?

A decision tree is a tree based modeling technique used to determine a class/event given a certain input.

3 Important Parts of Decision Tree

1. Root Node -> has no incoming edge but outgoing edge(s)
2. Leaf or Terminal Node -> has exactly one incoming edge and no outgoing edge.
3. Intermediary Node -> has exactly one incoming edge and two or more outgoing

Structure of Decision Tree

pending: Insert image

Decision Tree Algorithms
·       Hunt’s Algorithm
·       Ross Quinlan
§  i.ID3
§  ii.C4.5
§  iii.C5.0
·       Breiman et al.
o CART (Classification and Regression Trees)
·       Gordan Kass
§  i.CHAID (chi-square automatic interaction detector)

Hunt’s Algorithm
Hunt’s algorithm is recursive.
Let
Si =set of training records, that reach a node i
 Steps
1. If Sicontains records that belong the same class yi, then ‘i’ is a leaf node labeled as yi
2. If Sicontains records that belong to more than one class, use an attribute test to split the data into smaller subsets. Recursively apply the procedure to each subset. 






  • Google+
  • Pinterest
« PREV
NEXT »

No comments

Post a Comment