Trending Technology Machine Learning, Artificial Intelligent, Block Chain, IoT, DevOps, Data Science

Recent Post

Codecademy Code Foundations

Search This Blog

Rule Induction and Decision Trees in AI

Decision Trees

Decision tree is a classifier in the form of a tree structure. Each node is either
- A leaf node - indicates the value of the target attribute (class)
- A decision node - specifies some test to be carried out usually on a single attribute value with one branch for each outcome of the rest.



Decision Tree Example :




Each internal node tests an attribute

Each leaf node assigns a classification

Each branch corresponds To an attribute value

Decision Tree Defn :

A decision tree is a tree in which -
  • Each non-leaf node has associated with it an attribute (feature)
  • Each leaf node has associated with it a classification (+ or -)
  • Each arc corresponds to one possible value of the attribute of its parent node.
Decision tree for classification

A decision tree used to classify an example.
  • Start from the root node.
  • Follow the appropriate decision branches
  • On reaching a leaf node the predicted class is obtained
Classification us9ing Decision tree



What type of functions can decision trees represent ?

Disjunction of conjunctions
Try representing the following functions:
  • AND
  • OR
  • XOR
  • majority
Decision tree induction
  • Top down greedy search through the space of decision trees.
  • Use training data 
  • Quinlan developed ID3 in 1973 based on the CLS algorithm.
Top-Down Induction of Decision Trees : ID3
  1. A ← the "best" decision attribute for next node
  2. Assign A as decision attribute for node
  3. For each value of A create new descendant
  4. Sort training examples to leaf node according to attribute value of the branch
  5. If all training example are perfectly classified (same value of target attribute) stop, else iterate over new leaf nodes.
 ID3

1. Input :
  • Atts - A set of non-target attributes
  • Q - target attribute
  • S - training set
2. Returns a decision tree

If S is empty, return a single node with value Failure

If S consists of records of the same class, return a single leaf node with that value.

If Atts is empty, then return a single node with the value of the most frequent value of Q in S.

Begin
End

No comments:

Post a Comment

Popular Posts