Algoritma LightGBM untuk Deteksi Aktivitas Cyber Espionage Melalui Dataset Serangan Siber
DOI:
https://doi.org/10.70609/jusifor.v4i2.8489Keywords:
Espionage Siber, LightGBM, Algoritma Machine Learning,, Deteksi DiniAbstract
Cyber espionage is a type of cyberattack where hackers try to secretly and continuously steal important information through computer networks. This study suggests using the Light Gradient Boosting Machine (LightGBM) algorithm to spot early signs of digital espionage. The data used comes from the Cybersecurity Intrusion Detection dataset available on Kaggle. The research includes steps like cleaning and organizing the data, dividing it into 80% for training and 20% for testing, and training the model with carefully chosen settings for learning rate and number of leaves. The results show that the LightGBM model performed well, with an accuracy of 89%, an AUC of 0.874, and an average precision of 0.9064. For the attack class, the model had a precision of 1.00 and a recall of 0.75. The most important features that helped identify suspicious behavior were ip_reputation_score, session_duration, and network_packet_size. Early detection happens by looking at unusual patterns in the data to find network activities that look like cyber espionage. When compared to other methods like SVM and Random Forest, LightGBM works better and is faster. Based on these results, the LightGBM model is seen as a good tool for an early warning system to detect cyber espionage.
References
[1] A. Ciluvi, S. Luma-Osmani, E. Rufati, and G. Arifi, “Cyber Espionage: A Growing Threat?,” J. Nat. Sci. Math., vol. 7, no. 13/14, pp. 12–20, 2022, [Online]. Available: http://nuuance.net/work/Cyber Espionage.pdf
[2] M. E. Susila and A. A. Salim, “Cyber Espionage Policy and Regulation : A Comparative Analysis of Indonesia and Germany A . Introduction As technology evolves , it supersedes traditional forms of communication and information technology with newer , often more efficient , media . 1 Amon,” Pjih, vol. 11, no. 1, pp. 9–11, 2024.
[3] “2020 Internet Crime Report,” pp. 1–30.
[4] E. Y. Fitria, K. Mutijarsa, P. Korespondensi, S. Siber, K. Buatan, and D. Learning, “RESEARCH SURVEY ON ARTIFICIAL INTELLIGENCE METHODS FOR DETECTING CYBERATTACK TECHNOLOGY THREATS,” vol. 10, no. 6, 2023, doi: 10.25126/jtiik.2023107341.
[5] M. Atta, O. Ahmed, Y. Abdelsatar, R. Alotaibi, and O. Reyad, “Enhancing Internet of Things security using performance gradient boosting for network intrusion detection systems,” Alexandria Eng. J., vol. 116, no. December 2024, pp. 472–482, 2025, doi: 10.1016/j.aej.2024.12.106.
[6] M. Nidhi, S. Kadam, and M. Gayakwad, “An Ensemble Model for Cyber Attack and Threat Detection in Applications Network Using Random Forest , Lightgbm and Xgboost,” vol. 28, no. 3, pp. 523–534, 2025.
[7] E. Ramadanti, D. Aprilya Dinathi, C. Christianskaditya, and D. R. Chandranegara, “Diabetes Disease Detection Classification Using Light Gradient Boosting (LightGBM) With Hyperparameter Tuning,” Sinkron, vol. 8, no. 2, pp. 956–963, 2024, doi: 10.33395/sinkron.v8i2.13530.
[8] T. G. Laksana, S. Mulyani, I. Komputer, J. Informatika, and U. Bhayangkara, “UNTUK MENCEGAH PEMBOBOLAN DATA PERUSAHAAN Contact : Cite This Article :,” vol. 3, no. 1, pp. 109–122, 2024.
[9] J. N. Sibarani, D. R. Sirait, and S. Ramadhanti, “Intrusion Detection Systems pada Bot-IoT Dataset Menggunakan Algoritma Machine Learning,” vol. 14, no. 1, pp. 38–52, 2023.
[10] J. Nasional and S. Informasi, “Integrasi Algoritma FP-Growth dan K-Means untuk Analisis Keranjang Belanja dan Segmentasi Pelanggan pada Data Transaksi Ritel,” vol. 02, pp. 128–135, 2025.
[11] W. Yustanti and N. Iriawan, “Categorical encoder based performance comparison in pre- processing imbalanced multiclass classification,” vol. 31, no. 3, pp. 1705–1715, 2023, doi: 10.11591/ijeecs.v31.i3.pp1705-1715.
[12] W. Zhu, R. Qiu, and Y. Fu, “Comparative Study on the Performance of Categorical Variable Encoders in Classification and Regression Tasks,” pp. 1–19.
[13] J. Nasional, S. Informasi, E. Mustika, C. Sabila, R. Fakhrizal, and R. Kurniawan, “Analisis dan Prediksi Indeks Kualitas Udara Jakarta : Penerapan Algoritma XGBoost,” vol. 02, pp. 161–169, 2025.
[14] F. B. Dergisi, “A new Intrusion Detection System for Secured IoT / IIoT Networks based on A new Intrusion Detection System for Secured IoT / IIoT Networks based on Güvenli IoT / IIoT Ağları İç in LGBM T abanlı Yeni B ir Saldırı Tespit Sistemi,” 2023, doi: 10.29109/gujsc.1173286.
[15] Y. Cong, T. Guan, J. Cui, and X. Cheng, “LGBM : An Intrusion Detection Scheme for Resource-Constrained End Devices in Internet of Things,” vol. 2022, no. 1, 2022, doi: 10.1155/2022/1761655.
[16] G. Ketepalli, S. Y. M, and K. N. S. Lakshmi, “A Hybrid Intrusion Detection Model Using LSTMAE and LightGBM for Robust Anomaly Detection in Network Systems,” vol. 14, no. 5, pp. 912–922, 2025.
[17] R. Yacouby, “Probabilistic Extension of Precision , Recall , and F1 Score for More Thorough Evaluation of Classification Models,” pp. 79–91, 2020.
[18] J. Crall, “AND RECALL AS TRUE NEGATIVES APPROACH INFINITY .,” no. 2, 2023.
[19] N. Tran, H. Chen, J. Jiang, J. Bhuyan, and J. Ding, “Effect of Class Imbalance on the Performance of Machine Learning-based Network Intrusion Detection,” vol. 17, no. 9, pp. 741–755, 2021, doi: 10.23940/ijpe.21.09.p1.741755.
[20] A. Technology, “UNITEC INSTITUTE OF TECHNOLOGY MASTER THESIS A Comparative Analysis of Machine Learning Algorithms in Network-Based Intrusion Detection Systems for Detecting Advanced Persistent Threats to Enhance Cybersecurity A thesis submitted in partial fulfilment of the requirements for the Degree of Master of Applied Technologies in the School of Computing , Electrical and Applied Technology,” 2024.
[21] T. Ngootip, “Enhancing Network Intrusion Detection in Cloud Computing Using a Deep Boltzmann Machine and LightGBM Ensemble Model : A Performance Evaluation on the NSL-KDD Dataset,” vol. 1, no. 1, pp. 1–7, 2024.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Tasya Wulandari, Yasyfi Farhan Yudisthira, Kayla Chika P.H, Muhammad Afriza Wibowo, Chistofer Andrew
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
