Detection of Ransomware on Cryptocurrency Transactions using Machine Learning Techniques - Mini Project
This mini project involves the complete implementation of a research paper from scratch. The primary goal is to demonstrate the ability to translate academic research into code, identify and resolve errors, and enhance the overall implementation.
In cryptocurrency transactions, ransomware poses a persistent threat, demanding innovative detection solutions. Traditional methods struggle with evolving ransomware variants. This paper tackles the challenge of identifying ransomware in cryptocurrency transactions by using Bitcoinheist dataset. With 28 ransomware families categorized, including Princeton, Montreal, Padua, and a category for legitimate transactions, we propose a novel machine learning framework. Combining supervised and semi-supervised approaches, our framework employs ensemble learning methods to demonstrate significant progress in ransomware detection within cryptocurrency transactions.
Bitcoin Heist Ransomware Address Dataset from UCI machine learning repository.
The dataset consist of 10 attributes and 2916697 instances.There are 2 categorical features and 8 numeric features.