cyberai-classwork
The content of this repository is for a TU Special Topics course on CyberAI.
Science Score: 44.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (6.9%) to scientific vocabulary
Repository
The content of this repository is for a TU Special Topics course on CyberAI.
Basic Info
- Host: GitHub
- Owner: Cwagne17
- Language: Jupyter Notebook
- Default Branch: main
- Size: 17.8 MB
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
Hello There
The dataset I selected is the CIC-MalMem-2022 from the University of New Brunswick. The dataset is tailored for assessing memory-based obfuscated malware detection techniques. It simulates real-world scenarios by featuring common malware types like Spyware, Ransomware, and Trojan Horses, offering a well-balanced dataset for testing obfuscated malware detection systems. Notably, it employs debug mode during memory dump processes to emulate the conditions typical users might encounter during a malware attack, enhancing its realism.
The significance of addressing obfuscated malware in the realm of cybersecurity cannot be overstated. As malicious actors continually adapt and refine their tactics to evade detection, the development of robust and innovative methods for malware detection is an ongoing imperative. The CIC-MalMem-2022 dataset emerges as a pivotal tool in addressing this pressing concern. Notably, the initial research paper utilizing this dataset primarily concentrated on feature engineering and selection to attain optimal results in binary classification models. However, it did not explore the dataset's potential for classifying malware types and families.
The objectives of my experiment encompass two primary goals. First, I aim to recreate the binary classification model originally presented in the research paper. Second, I seek to explore the dataset's potential for classifying malware by type and family. To achieve these objectives, I will adhere to the feature set developed in the original paper, and employ the most effective algorithms, including K-Nearest Neighbor (KNN), Random Forest, Decision Tree, and a Stacked Ensemble employing KNN, Random Forest, and Decision Tree as base learners, with a Logistic Regression model as the metalearner.
Owner
- Name: Christopher M Wagner
- Login: Cwagne17
- Kind: user
- Location: MD
- Company: SecurEd Inc.
- Website: www.linkedin.com/in/christopher-wagner-8a8043186
- Repositories: 7
- Profile: https://github.com/Cwagne17
Citation (CITATION.cff)
cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
- family-names: Wagner
given-names: Christopher
orcid: https://orcid.org/1234-5678-9101-1121
title: "CyberAI Coursework"
date-released: 2023-10-25
url: https://github.com/Cwagne17/CyberAI-Classwork