LaBackDoor

Projects

Explore my innovative cybersecurity research projects focusing on advanced neural networks, explainability, and cutting-edge threat detection. My work bridges the gap between machine learning and network security to address emerging challenges in the digital landscape.

NetSpeech
NetSpeech Transforming Packet Whispers into Actionable Intelligence

A transformer-based NLP approach that translates network traffic data (PCAP) into natural language for enhanced security analysis. This project enables security analysts to interpret network patterns more intuitively through advanced NMT techniques.

VulnScout
VulnScout Advanced Memory Composition in Multi-Recurrent Neural Networks

A revolutionary approach to Multi-Recurrent Neural Networks (MRNs) focusing on autonomous memory composition through self-learning link ratios, enhanced pruning techniques, and adaptive activation functions. This research improves forecasting of software vulnerabilities while developing deeper MRN architectures with reduced computational overhead.

More

TempoExplain
TempoExplain Illuminating the Black Box of Temporal Neural Networks

Making temporal Komogorov-Arnold neural networks transparent through advanced explainability techniques. This project provides insights into how these complex models process time-series data for security applications.

MalwareRadar
MalwareRadar System Call Profiling with Stratos Shark

Revolutionary baseline analysis system for detecting malicious behavior patterns in system calls. This framework establishes new benchmarks for malware detection through behavioral analysis techniques.