Authors: Leo Martinez III - LinkedIn, John Cavazos, Anthony Martinez, Jalen Williams
Contact: Leo: leo.martinez@students.tamuk.edu, Anthony: anos.martinez24@gmail.com
Created: Fall 2023 - Spring 2024
Please see necessary dependencies at the top of each script if you choose to run it locally; some (not all) of the major dependencies are Scikit-Learn, TensorFlow, Keras, XGBoost, NumPy, and Pandas.
- The machine learning models were developed in Spyder IDE using Python 3.18.
- The Android app was developed in AndroidStudio using Java/XML.
- The website was developed in VS Code using HTML/CSS.
M.T.D. (Multi-Threat Detection) is a comprehensive cybersecurity solution designed to combat the ever-evolving landscape of cyber threats. The project employs a holistic approach, addressing common vulnerabilities with a focus on real-world relevance and practicality. It provides an Android app utilizing state-of-the-art pretrained machine learning models. A full detailed report for those interested in the specific details of the design for each machine learning model.
- Machine Learning Models: Four ML models are included, saved individually in folders:
- Network Intrusion Detection
- Network Intrusion Classification
- Malware Detection
- Malware Classification
- Scoring Scripts: Pickle (.pk1) files with scoring scripts allow customization for specific needs.
- Android App: A mobile Android app is provided to utilize the models and empower users to be more aware of cybersecurity threats.
- Website: An accompanying website offers easy access to information and the app.
Each ML model has been rigorously tested against various research papers to prove its effectiveness. The project integrates diverse datasets to ensure solutions resonate with real-world scenarios. Various algorithms, including Random Forest, Support Vector Machines, and Artificial Neural Networks, are meticulously assessed for effectiveness and robustness.
In conclusion, M.T.D. represents a proactive defense against cyber threats, underpinned by cutting-edge technology and a commitment to ongoing enhancement. As we navigate the complexities of the digital landscape, our project stands as a beacon of resilience and innovation in the fight for cybersecurity.
- android_app: Folder containing source code and necessary files for the Android App.
- images: Contains images for visualization purposes.
- machine_learning_models: Contains subfolders for each ML task written in Python3.
- website: Contains necessary files for viewing the website written in HTML/CSS.
- MTD_Detailed_Report.docx: Highly detailed report for those interested in intricate details.
- README.md: Contains context information about the project (you are here!).
- LICENSE: Contains license information (MIT) for the GitHub repository.