Malware/Ransomware Analysis and Detection
摘要
This paper analyzes the CIC-MalMem-2022 dataset focusing on obfuscated malware detection through memory analysis. The dataset comprises a balanced collection of benign and malicious memory dumps, encapsulating prevalent real-world malware families such as Trojan Horse, Spyware, and Ransomware. Our methodology involves data preprocessing, feature extraction, and data analysis using Tableau. We generate various visualizations that elucidate the distribution and characteristics of the malware families. We discuss our findings and highlight their practical implications for enhancing real-world malware detection systems. By identifying features with discriminatory power for distinguishing benign and malware instances, our study provides valuable insights that can guide the development and optimization of future cybersecurity practices and tools.