Inspiration
We wanted to build something related to Machine Learning and cybersecurity so we decided to experiment with packet sniffing and eventually came up with this idea.
What it does
Goes through any webpage you desire, sniffs all requests while on the webpage, and classifies harmful and non-harmful packets using Machine Learning.
How we built it
A KNN based binary classification model (98.01% accuracy) was developed and Scapy was implemented to sniff packets and the packets were loaded into the model.
Challenges we ran into
We had many problems with the Dataset. We weren't too sure whether many of the columns would impact the model as we aren't too well versed in cybersecurity. We also had problems with creating the model as the dataset was just too large, so we had to work out a way to make sure the model was fast enough and accurate enough.
Accomplishments that we're proud of
Building a high performing Machine Learning model and being able to implement our model into network packets.
What we learned
How to overcome challenges and learning new technologies.
What's next for Intelligently Secured
We hope to share Intelligently Secured with the world as we believe it can be a useful, free resource.
Built With
- python
- scapy
- scikit-learn
