University of Central Missouri · Computer Science & Cybersecurity
NetML Lab combines network measurement, machine learning, optimization, explainable AI, and large language models to build practical defenses and more efficient retrieval-augmented generation.
Latest research news
Jasurbek Ibragimov and Ahmet Aksoy
Across four question-answering datasets, JMAP removed 81–87% of context tokens while improving Qwen3-8B answer F1 over unpruned context.
Workshop on Long Context Foundation Models (LCFM) @ NeurIPS 2026 · Atlanta, GA · Explore the project
What we study
Our work focuses on making security analytics more automated, adaptive, interpretable, and useful in real network environments.
Machine learning systems for detecting malicious traffic, botnets, denial-of-service attacks, fast-flux networks, and scanning activity.
Passive identification of devices, hosts, and operating systems from packet-level protocol headers and encrypted traffic metadata.
Genetic algorithms, Artificial Bee Colony optimization, feature selection, quantization, and SHAP-based model analysis.
Large language models for security tasks, alongside retrieval-augmented generation and adaptive context pruning.
How we work
Students work across the full research cycle: framing a security problem, building data-driven methods, testing them against realistic traffic, and communicating results through publications and presentations.
The lab
NetML brings together students who want to build strong technical foundations, ask useful questions, and contribute to publishable research.
Associate Professor · Lab Director
LLM optimization and long-context RAG
LLM-based incident classification
LLM-based incident classification
User behavior fingerprinting
User behavior fingerprinting
Build with us
We welcome motivated students with strong programming skills, analytical thinking, and curiosity about machine learning, networks, and security. Students engage in hands-on experiments and are expected to contribute to research and publication.
Learn how to apply