1
Random Forest
2
Introduction
3
Supervised Learning
4
Why Random Forest
5
What is Random Forest?
6
Random Forest
7
Applications of Random Forest
8
Important Terms
Entropy
Information Gain
Leaf node
Decision Node
Root Node
9
How does a Decision Tree Work?
10
How does a Decision Tree Work?
11
How does a Decision Tree Work?
12
How does a Decision Tree Work?
13
How does a Decision Tree Work?
14
How does a Decision Tree Work?
15
How does a Decision Tree Work?
16
How does a Decision Tree Work?
17
How does a Decision Tree Work?
18
How does a Decision Tree Work?
19
How does a Decision Tree Work?
20
How does a Decision Tree Work?
21
How does a Decision Tree Work?
22
How does a Decision Tree Work?
23
How does a Decision Tree Work?
24
How does a Decision Tree Work?
25
How does a Decision Tree Work?
27
Random Forest Prediction Pseudocode
To perform prediction using the trained random forest algorithm uses the below pseudocode:
1. Takes the test features and use the rules of each randomly created decision tree to predict the outcome and
stores the predicted outcome (target)
2. Calculate the votes for each predicted target.
3. Consider the high voted predicted target as the final prediction from the random forest algorithm.
28
How Random Forest Works
Let this be Tree 1
29
How Random Forest Works
Let this be Tree 2
30
How Random Forest Works
Let this be Tree 3
Results Compared 31
Tree 1 Tree 2 Tree 3
• This concept of voting is known as majority voting.
• Highest voted is the final outcome.
32
Missing Data Example
1. Diameter =3
2. Color =Orange
3. Grows in Summer = Yes
4. Shape = Circle
Tree 1 classifies it
as an Orange
33
Missing Data Example
1. Diameter =3
2. Color =Orange
3. Grows in Summer = Yes
4. Shape = Circle
Tree 2 classifies it
as Cherries
34
Missing Data Example
1. Diameter =3
2. Color =Orange
3. Grows in Summer = Yes
4. Shape = Circle
Tree 3 classifies it
as Oranges
Results Compared 35
Tree 1 Tree 2 Tree 3
36
Conclusions
Random forests are an effective tool in prediction.
Random inputs and random features produce good results in classification.
For larger data sets, we can gain accuracy by combining random features
The problem is
same as before
You are given data by your manager of customers who have previously bought some older makes of
your
company’s SUV. The data includes a total of 400 instances
Independent variables
• Age
• Estimated salary
Dependent Variable
• Purchased (0 = no SUV purchased, 1 = SUV purchased)
Your company has just introduced a new SUV vehicle.
You are asked to predict who will buy the new SUV vehicle
Import the libraries
Import the dataset
Split the dataset into training and testing
samples Feature scaling
Training the model
Predicting the new results
Predicting the Test set results
Making the confusion matrix
Visualize the training set results
Visualize the test set results
n_estimators=10
Criterion=entrop
y Random
state=0
The confusion matric shows that 91 samples are correctly classifies and 9 are mis
classified.
The true negative values are 63 i.e., expected value is 0 and predicted value is 0.
The false negative values are 5 i.e., expected value is 1 and the predicted value is
0. The false positive values are 4 i.e., expected value is 0 and the predicted
value is 1. The true positive values are 28 i.e., expected value is 1 and the
predicted value is 1.
