Overview / Usage

Given the Gender , Age , Estimated Salary of a person the decision tree model will tell whether the customer will buy the car or not. I did this small project in my 2nd year of engineering. I am still learning and completing courses online and brushing my skills. This project aims to t decide if the customer will buy the car or not based on the data it is trained on. If the company has the data set of customers who have showed interest in purchasing the car then the company can follow and get in touch with the costumers that the model predicts . Which in turn will save their time and they will have the idea on the customers they need to follow.

The accuracy of my model was 91%.

Methodology / Approach

I have used the method of decision trees for classification in which each internal node corresponds to a test. Each branch corresponds to a result of the test. Each leaf node assigns a classification. If the entropy of the given attribute is high, then it is a bad splitting attribute. The lower the entropy, the less uniform distribution, the purer the node.

Also the tree with higher information gain after splitting is the winner.

Entropy=p(A)log(p(A))-p(B)log(p(B)), where A and B are the classifications

Information Gain=entropy before split-weighted entropy after the split

Technologies Used

used numpy, matplotlib, pandas and scikit learn

​