Scribd
Upload
76 views

Decision Tree

The document discusses decision trees, which are a type of classification algorithm used in machine learning. Decision trees use a tree-like model to split a dataset into smaller and smaller subsets while simultaneously an associating a target variable. They can be used for both classification and regression problems. The document outlines key concepts like entropy, information gain, leaf nodes, decision nodes, and root nodes. It also discusses advantages like simplicity and minimal data preparation, and disadvantages such as potential overfitting and instability.

Uploaded by

64-Shashank Sagar
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
76 views

Decision Tree

The document discusses decision trees, which are a type of classification algorithm used in machine learning. Decision trees use a tree-like model to split a dataset into smaller and smaller subsets while simultaneously an associating a target variable. They can be used for both classification and regression problems. The document outlines key concepts like entropy, information gain, leaf nodes, decision nodes, and root nodes. It also discusses advantages like simplicity and minimal data preparation, and disadvantages such as potential overfitting and instability.

Uploaded by

64-Shashank Sagar
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Decision Tree

In our day-to-day life, we interact with various machine learning applications and use it without
knowing it. The best example is buying something from any online shopping portal where we get
several recommendations based on what we are buying.

One type of machine learning algorithm is Decision Tree, which is a type of classification
algorithm that comes under supervised classification.

The decision tree is something that we might have used knowingly or unknowingly. Consider the
case of buying a car. We will choose a car after considering various factors like budget, safety,
color, and price. We first checked the price which is less than X followed by Color followed by
safety and then reached a conclusion.
Image by Author

Looking at the above diagram we can define the Decision tree is a graphical representation of
a tree-shaped diagram that is used to determine the course of action. Each branch of the tree
represents a decision.

The problem decision tree can solve?

It can solve two types of problems.

1. Classification: Classify based on if-then condition. Ex: If a flower color is red then its
rose, if it’s white then lily.
2. Regression: Regression tree is used when there is continuous data.
Advantages of Decision tree

1. Simple to understand.
2. Little effort in data preparation.
3. The non-linear parameter does not affect performance.

Disadvantages:

1. Overfitting: In the case of noise in the data.


2. Instability: Models can become unstable due to variations in data.

Important terminology to be used in the Decision Tree.

1. Entropy: It’s the measure of unpredictability in the dataset. For example, we have a
bucket of fruits. Here everything is mixed and hence it’s entropy is very high.
2. Information gain: There’s a decrease in the entropy. For example, if we have a bucket of
5 different fruits. If all are kept in one place then the information gained is minimal. But if
we keep all 5 fruits separate we see the entropy as min as it’s not mixed and information
gained as maximum.
3. Leaf node: It’s the end of the decision tree that carries the information. In the figure
above we can say “Buy” is the leaf node.
4. Decision Node: It’s the mid node in the decision tree where 2 or more new splits arise. In
the above diagram, color is a decision node because it further splits into red and blue.
5. Root Node: It’s the topmost node of the figure where all the information is stored or has
the highest entropy. In the diagram “Car” is the root node.

Who is a anomaly person?


an anomalous person or thing; one that is abnormal or does not fit in:
What is an example of a anomaly?
An anomaly is an abnormality, a blip on the screen of life that doesn't fit with the rest of the
pattern. If you are a breeder of black dogs and one puppy comes out pink, that puppy is
an anomaly.

You might also like