Malware Detection Using Deep Learning & ML
Capstone project focused on building deep learning and machine learning models to detect malware patterns, combining cybersecurity expertise with ML engineering.
Motivation
With the rise of malware attacks, detecting malicious software efficiently is critical. This project applies machine learning and deep learning to automate malware classification and improve security measures.
Tools & Technologies
- Python, PyTorch
- Data preprocessing with pandas & NumPy
- Deep Learning models: GRUs and CNNs / RNNs
- Machine Learning with scikit-learn (Random Forest, SVM, etc.)
- Data visualization with matplotlib & seaborn
Methodology
- Collected and labeled malware datasets.
- Performed data preprocessing and feature extraction.
- Designed and trained deep learning models (GRUs, CNNs/RNNs) for classification.
- Applied classical machine learning methods using scikit-learn (Random Forest, SVM) for comparison.
- Evaluated model performance with accuracy, precision, recall, and F1-score.
Results & Key Takeaways
Achieved high accuracy in malware detection, demonstrating the effectiveness of deep learning and classical ML methods. Learned best practices in model evaluation, dataset handling, and integrating ML approaches for cybersecurity projects.