INTRODUCTION TO DECISION TREE
Decision tree algorithm is one of the popular machine learning algorithm used in supervised learning(Where the input data and the corresponding output values are provided to the model). In General it is the Graphical representation of possible solution which splits according to a certain parameters. It is a tree like structure with many splits with probable outcomes.
this algorithm can be used in both regression and classification challenges. It is one of the best predictive model with high accuracy and easy interpretation. They can handle non linear data set effectively.
Let‘s understand decision tree with an example.
Consider this scenario, In order to get a ride in a Giant Wheel ,a person need to satisfy the following condition. if he satisfies this condition then he is eligible to take a ride in Giant wheel. if not then, he is not allow inside the giant wheel.
- For Male Weight should be greater than 50kg.
- though, For Male Height should be above 150cm.
- For Female Weight should be greater than 40kg.
- so, For Female height should be above 145cm.
- Age should be greater than 16years.
Terminologies associated with this example
1. Root Node:
The node which is at the highest hierarchical order is the root node or parent node.
2. Internal Node:
The intermediate node with two or more branches is the internal node.
3. Leaf Node:
The node with no branches are the leaf nodes.
4. Child Node:
The node which is at the lowest hierarchical is the child node.
so, The above diagram is the example for the representation. thus, Based on this structure the algorithm works to find the target variable or the output variable.
TYPES OF DECISION TREE ALGORITHM
There are two types of This Algorithm based on the output variable(target variable).They are categorical variable Classification algorithm and continuous variable Regression algorithm.
1. Classification tree:
Classification decision tree works with categorical target variable(output) which usually contains two or more categories eg. Male/Female, Cat/Dog etc.
so, Output variable-depends on categorical data.
2. Regression tree:
In Regression decision tree the target variable (output) usually deals with categorical data. For e.g. which can take unlimited Number of values
APPLICATION OF DECISION TREE ALGORITHM:
Business Management.
Decision Tree algorithms are also used in many organisation to extract useful information from the database. however, Companies use this algorithm commonly in customer relationship management.
Fraudulent Detection
Decision tree algorithm has thus proved that this model has performed well in classifying fraudulent statements with high accuracy.
Energy consumption
Result ‘s from This Algorithm, hence provides deeper insights about the electricity consumption of a household to the electricity companies to make measures to consume electricity.
Healthcare Management:
This algorithm can be used for making predictions in healthcare sectors,such findings helps the doctors to provide better treatment to the patients.
ADVANTAGES AND DISADVANTAGES OF DECISION TREE ALGORITHM:
ADVANTAGES:
- Decision Tree algorithm is easy to understand and visualise the flow of the algorithm.
- It does not require high data pre processing techniques.
- thus, algorithm works well with both numerical and categorical features.
- Normalisation of the data is not required while using this algorithm.
- Decision tree algorithm can also be implemented with missing values in a dataset.
- Perform well with large dataset.
- here, algorithm is very simple when compared with other algorithms.
DISADVANTAGES:
- Decision Tree is more sensitive even a small change in the data can thus result in complex tree structure.
- Training time is relatively large when compared to other models.
- The mathematical calculation may be more complex in some cases.
- Complex tree structure consumes more memory in This Algorithm.
CONCLUSION:
The main objective of this blog is to give an overview of this algorithm and its application.Hope you enjoyed this blog.”Machine learning is the next internet, so start learning…”
Happy Learning!!!
written by: Nikesh Joseph
Reviewed By: Vikas Bhardwaj
If you are Interested In Machine Learning You Can Check Machine Learning Internship Program
Also Check Other Technical And Non Technical Internship Programs
