Inspiration
The inspiration for this project was the increasing amount of scams we encounter on a daily basis which makes it important to detect and identify
What it does
Our phishing URL detector is designed to calculate the probability that a URL is a phishing site
How we built it
Our build process included finding and cleaning datasets to train our model and creating a neural network model from scratch
Challenges we ran into
Our team was very new to machine learning and artificial intelligence. We had to research the creation of models ourselves
Accomplishments that we're proud of
We're proud of the fact that, despite having almost no prior knowledge of machine learning, we created a neural network model. Despite having a small dataset, our model had a good accuracy in identifying phishing sites.
What we learned
A lot about building a machine learning model, including a neural network, and random forest, and integrating it into a website
What's next for Phishing URL Detector
We hope to turn the Phishing URL Detector into an extension that immediately warns the user if they attempt to navigate to a phishing website. We also hope to expand the extension to be able to identify phishing in different forms, such as text, images, emails, and calls.
Log in or sign up for Devpost to join the conversation.