Large-Scale Multisources Malware Analysis Dataset using Network Traffic and Memory (BCCC-Mal-NetMem-2025)

The BCCC-Mal-NetMem-2025 dataset comprises over 7.7 million labeled records from controlled experiments involving 15 malware categories and 32 individual malware samples. These categories include ransomware, Trojan downloaders, coin miners, remote access tools (RATs), spyware, backdoors, and worms. The data was collected by executing each malware in isolated Windows environments equipped with real-time network and memory monitoring tools to ensure comprehensive behavioral capture. The dataset integrates memory and network traffic features, offering a multidimensional view of malware behavior for accurate profiling. This hybrid structure allows advanced AI-driven threat detection and malware characterization, consistent labeling, and session-based organization that supports detailed analysis. The BCCC-Mal-NetMem-2025 is a unique benchmark for behavioral malware analysis, bridging gaps between static profiling and real-world execution patterns.