Fingerprinting using Machine Learning
In a world of ever-growing networks, it is crucial not just to build models, but to understand what they reveal. From simple to advanced, the students explored different paths to the same truth: smarter, safer digital environments.


As digital networks continue to grow in scale and complexity, monitoring them for suspicious activity is critical to protect essential infrastructure such as power plants, factories, and bridges. The project focused on detecting potentially malicious actors by analyzing their digital fingerprints using AI-based methods.
The students developed multiple AI models that leverage network data and inter-relational structures to fingerprint devices. The project delivered a scalable solution, with the best-performing model achieving a recall of 99%, significantly improving the detection of suspicious activity compared to existing approaches.