Dayoung Choi*, Joohong Rheey*, and Hyunggon Park, "Attack-Specific Feature Analysis Framework for NetFlow IoT Datasets," Computers & Security, vol. 157,Β pp. 104536, Oct. 2025. (*: Equal contribution) (Rating: Q1, IF 5.4) π
Joohong Rheey and Hyunggon Park, "SV-SAE: Layer-wise Pruning for Autoencoder Based on Link Contributions," IEEE Access, vol. 13, pp. 75666-75678, May. 2025. (Rating: Q2, IF 3.6) π
Joohong Rheey and Hyunggon Park, "Robust Hierarchical Anomaly Detection using Feature Impact in IoT Networks," ICT Express, vol. 11, no. 2, pp. 358-363, Apr. 2025. (Rating: Q1, IF 4.2) π
Joohong Rheey and Hyunggon Park, "Lightweight deep learning model for edge devices with time-series data," In preparation.Β
Dayoung Choi, Joohong Rheey, and Hyunggon Park, "Unsupervised anomaly detection for tabular data," In preparation.Β
Joohong Rheey, Dayoung Choi, and Hyunggon Park, "Poster: Symmetrical Pruning for Lightweight Network Anomaly Detector," ACM International Conference on Mobile Systems, Applications, and Services (MobiSys 2024), Jun. 2024. (Student Travel Grant Awarded)
Joohong Rheey and Hyunggon Park, "Game-Theoretic Lightweight Autoencoder Design for Intrusion Detection," IEEE Wireless Communications and Networking Conference (WCNC 2024), Apr. 2024. (Student Travel Grant Awarded)
Joohong Rheey and Hyunggon Park, "Lightweighted Sparse Autoencoder based on Explainable Contribution," International Conference on Machine Learning (ICML 2023) (Neural Compression Workshop: From Information Theory to Applications), Jul. 2023.Β
Joohong Rheey and Hyunggon Park, "Low-complexity Anomaly Detection Method based on Feature Importance using Shapley Value," International Conference on Ubiquitous and Future Networks (ICUFN 2023), Jul. 2023.
Joohong Rheey, Dayoung Choi, and Hyunggon Park, "Adaptive Loss Function Design Algorithm for Input Data Distribution in Autoencoder," International Conference on ICT Convergence (ICTC 2022), Oct. 2022.
μ΄μ£Όν, μ λ€μ, λ°νκ³€, "μ λ ₯ λ°μ΄ν° 무μμλ νΉμ±μ΄ κΈ°λ³Έ μ€ν μΈμ½λμ μ μΈ΅ μ€ν μΈμ½λμ νμ΅ μ±λ₯μ λ―ΈμΉλ μν₯ (Impact of Input Data Randomness on Training Performance of Basic Autoencoder and Stacked Autoencoder)," νκ΅ν΅μ ννλ Όλ¬Έμ§, pp. 178-187, 2022λ 1μ. π
μ΅μμ, κΆμν, μ΄μ£Όν, λ°νκ³€, "κ°μ§μΉκΈ°-μμν κ²°ν©μ ν΅ν μ¨λλ°μ΄μ€ κ²½λν μ€μ¦ μ°κ΅¬ (An Empirical Study of On-Device Model Compression via Pruning and Quantization)," μ 6ν μΈκ³΅μ§λ₯ νμ λν, 2025λ 9μ.
μ§λͺ μ€, μ΄μ£Όν, λ°νκ³€, "xLSTM μ€ν μΈμ½λμ μΈμ½λ-λμ½λ ꡬμ±μ λ°λ₯Έ μ΄μνμ§ μ±λ₯ (Anomaly Detection Performance of xLSTM Autoencoder Based on Encoder-Decoder Architecture)," μ 35ν ν΅μ μ 보 ν©λνμ λν(JCCI), 2025λ 4μ.
κΉμν, μ΅λ€μ, μ΄μ£Όν, λ°νκ³€, "μ΄ν μ κΈ°λ° λ€νΈμν¬ νΈλν½ νΉμ§ μ€μλ λΆμ μ°κ΅¬ (Attention Mechanism-based Feature Importance Analysis for Network Traffic)," μ 34ν ν΅μ μ 보 ν©λνμ λν(JCCI), 2024λ 4μ.
μ΅λ€μ, μ΄μ£Όν, λ°νκ³€, "νΉμ§ μ ν λ°©λ²μ λ°λ₯Έ λ€νΈμν¬ κ³΅κ²© μ νλ³ μ€μ νΉμ§ λΉκ΅ μ°κ΅¬ (Important Feature Comparison for Network Attack Types based on Feature Selection Methods)," 2024λ λ νκ΅ν΅μ νν λκ³μ’ ν©νμ λ°νν, 2024λ 1μ.
κΉλ―Όμ£Ό, κΆκ²½λ―Ό, μ΅μκ²½, μ΄μ£Όν, λ°νκ³€, "μ λμ 곡격μ κ°κ±΄ν νΉμ§ μ€μλ κΈ°λ° κ³μΈ΅μ λ€νΈμν¬ μ΄μνμ§ λ°©λ² (Robust Feature Importance-based Hierarchical Network Anomaly Detection Method against Adversarial Attacks)," μ 4ν νκ΅ μΈκ³΅μ§λ₯ νμ λν, 2023λ 9μ. (νλΆ λ Όλ¬Έ κ²½μ§λν μ΅μ°μμ μμ)
μ΅λ€μ, μ΄μ£Όν, λ°νκ³€, "μνΈμ 보λ κΈ°λ° κ°μ§μΉκΈ° κΈ°λ²μ μ΄μ©ν λ€νΈμν¬ νΈλν½ μ΄μνμ§ (Network Traffic Anomaly Detection using Mutual Information-based Pruning)," 2023λ λ νκ΅ν΅μ νν νκ³μ’ ν©νμ λ°νν, 2023λ 6μ.
μ΅λ€μ, μ΄μ£Όν, λ°νκ³€, "μ μ§μ νΉμ§ μΆκ°λ₯Ό ν΅ν λ€νΈμν¬ νΈλν½ μ΄μ νμ§ μ±λ₯ ν₯μμ κ΄ν μ°κ΅¬ (A Study on Performance Improvement of Network Traffic Anomaly Detection via Progressive Feature Addition)," 2023λ λ νκ΅ν΅μ νν λκ³μ’ ν©νμ λ°νν, 2023λ 2μ. (νλΆ λ Όλ¬Έ κ²½μ§λν μ₯λ €μ μμ)
μ΅λ€μ, μ΄μ£Όν, λ°νκ³€, "λ€νΈμν¬ νΈλν½μ νΉμ§ κ·Έλ£Ήμ μ΄μ©ν μ€ν μΈμ½λ κΈ°λ° μ΄μνμ§μ κ΄ν μ°κ΅¬ (Autoencoder-based Anomaly Detection using Network Traffic Feature Grouping)," μ 3ν νκ΅ μΈκ³΅μ§λ₯ νμ λν, 2022λ 9μ.Β (νλΆ λ Όλ¬Έ κ²½μ§λν μ°μμ μμ)
νμλ―Ό, μ΄μ£Όν, λ°νκ³€, "μ€ν μΈμ½λμ μ μΆλ ₯ λΆν¬ μ μ¬λ ν₯μμ μν μ μ²λ¦¬ μκ³ λ¦¬μ¦ (The Preprocessing Algorithm for Improved Similarity on Input and Output Data Distribution in Autoencoder)," 2022λ λ νκ΅ν΅μ νν νκ³μ’ ν©νμ λ°νν, 2022λ 6μ.
μνλ¦°, μ΄μ£Όν, λ°νκ³€, "μ€ν μΈμ½λμ νμ΅ μ±λ₯ ν₯μμ μν κ°μ λ μμ€ν¨μ μ€κ³ (Improved Loss Function for Autoencoder Performance Enhancement)," μ 32ν ν΅μ μ 보 ν©λνμ λν(JCCI), 2022λ 4μ.
μ΄μ£Όν, μ λ€μ, λ°νκ³€, "μ λ ₯ λ°μ΄ν° μ°ν¬λμ μ€ν μΈμ½λ νμ΅ μ±λ₯ μ°κ΄μ±μ κ΄ν μ°κ΅¬ (A Study of Relation Between Input Data Dispersion and Autoencoder Training Performance)," 2021λ λ νκ΅ν΅μ νν μΆκ³μ’ ν©νμ λ°νν, pp.435-436, 2021λ 11μ.
μ΄μ£Όν, μ λ€μ, λ°νκ³€, "μ λ ₯ λ°μ΄ν°μ 무μμλμ λ°λ₯Έ μ€ν μΈμ½λμ νμ΅ μ±λ₯μ κ΄ν μ°κ΅¬ (Impact of Input Data Randomness on Training Performance of Autoencoder)," 2021λ λ νκ΅ν΅μ νν νκ³μ’ ν©νμ λ°νν, 2021λ 6μ. (μ°μλ Όλ¬Έμ μμ)