Phishing Website detection using Machine Learning

Inspiration

I have seen many people getting fooled by phishing websites and losing their lives. People need to have a tool that they can use to distinguish between legitimate and phishing websites. I wanted to do this project as I felt that Machine Learning can be used to tackle such need of the hour issues and help people in staying safe in the cyber world.

What it does

My project is the building of a model, that can be integrated into websites or even developed in chrome extensions so that the user can enter the website and check if it is legitimate or not before proceeding.

How we built it

I built this model using python and took the help of some great libraries like NumPy,sklearn. I have used a Decision tree classifier to train the model.

Challenges we ran into

The problem was building the dataset. Since I didn't find datasets with both legitimate as well as phishing websites, I had to preprocess and use feature Extraction, to build new datasets.

Accomplishments that we're proud of

I have successfully built and trained the model and checked for its prediction too. The model provides 0.84 accuracy, which is quite good.

What we learned

I learned to use a lot of new libraries and since it was my own project, I had to deal with all the errors and rectify them on my own. So it was more of a learning experience.