Stacked Hidden Markov Models (HMM) and Support Vector Machines (SVM) to classify malware samples based on opcode sequences. Implemented the HMM from scratch using Java and used sklearn’s SVM implementation to get an AUC of 85%-100% depending on the malware family.
Java, Python, Scikit-Learn, Pandas, Matplotlib
Built a maze game where the user moves a dog carrying a pizza through a series of mazes. Game features include in-game notifications, reset/time traps, and increases in difficulty throughout the mazes. The project was structured using the MVC design pattern and tested using JUnit.
Java, Java Swing, JUnit