This document discusses different types of decisions, decision trees, and how decision trees are used for classification. It describes strategic, administrative, and operating decisions. It defines a decision tree as a graphical representation of possible solutions to a decision based on conditions, and discusses decision nodes, chance nodes, and end nodes. It explains how decision trees are used for classification, with training and test datasets containing attributes like weather conditions to predict if people will play tennis or not. Finally, it provides an example of rules generated from a decision tree to classify days based on weather.
2. Types of Decision
Strategic Decision: Concerned with structuring and acquisitor of the
organization.
Administrative Decision: Concerned with structure and acquisitor of
the organization's resources so as to optimize the performance of the
organization.
Operating Decision: Concerned with day to day operations of the
organization such as pricing, production scheduling, inventory levels
etc.
3.
4. Decision Tree
Definition: A Decision tree is a graphical representation of
possible solution to a decision based on certain conditions.
It’s called a decision tree because it starts with a single
box(or root), which then branches off into a number of
solutions, just like a tree.
OR
it is the process of chosing a course of action from among
alternatives to achieve a desired goal.
5. Nodes for making Decision
tree
Decision Nodes: Commonly Represented by squares.
Change nodes: Represented by circle.
End Nodes: Represented by triangles.
6. DECISION TREE
• Classification scheme
• Generates a tree and a set of
rules
• Set of record divide into two
subsets
Training set
Test set
• Attributes are divide into 2
types
Numerical attribute
Categorical attribute
8. DECISION TREE
Decision tree to represent learned
target functions
Each internal node tests an attribute
Each branch corresponds to attribute
value
Each leaf node assigns a classification
Rules are easier for humans to
understand
10. Example
So, let me explain this to you with an example. So, this is what I was
mentioning that this table that we see represents the training set of the
training examples. Let us see what this table means. In this in this toy
example the instances are objects that you are talking about and nothing,
but it a day, a day of the week some day of some season. So, each row in
this table describes a day. So, there are there are D 1 to D 14, there are 14
days which are previous examples. And each day belongs to one of the two
categories one of the two categories. If you look at the table in the slide,
there are two categories whether people prefer to play tennis and outdoor
sports on that day or does not prefer to play tennis on that day. So, each
day belongs to two categories whether people play tennis or do not.
11. Rule 1: If it is sunny and humidity is
high then don’t play
Rule 2: If it is sunny and humidity is
normal then play
Rule 3: If it is overcast then play
Rule 4: If it is rainy and windy then
don’t play
Rule 5: if it is rainy and not windy
then play
12. Tree Construction Principle
Generally building a tree involves two
steps:
Tree construction- recursively split the tree according to selected
attributes
Tree pruning- identify and remove the irrelevance braches