New Publication(AI-powered framework for detecting hacking attempts) in JKSCI

We are pleased to announce that our team has published a new paper in the Journal of The Korea Society of Computer and Information (JKSCI), Vol. 24, No. 1, March 2026.

This study proposes a 3-stage lightweight Intrusion Detection System (IDS) framework designed for IoT edge environments. By converting tabular network traffic data into images using the novel LLM-categorized Vortex Feature Positioning (LVFP) technique and pre-training a lightweight CNN encoder via contrastive learning, the model captures complex relationships in network traffic that traditional deep learning approaches struggle to detect. Evaluated on 6 IDS & IoT benchmark datasets, the proposed model outperforms existing methods while maintaining a minimal number of parameters.

"Semantic Tabular-to-Image Conversion and Contrastive Learning for Lightweight Intrusion Detection" Jun Yeong Park, Kunwoo Kang, Hoin Lee, Seungeun Lee, & Yu Rang Park (2026). JKSCI, 24(1).