Pius Deski Manalu *, Ammar Yasir Sistem Deteksi Otomatis Berbasis Machine Learning untuk Jejak Digital dalam Analisis Forensik Siber

Abstract Views: 37   PDF Downloads: 32

Authors

  • Ammar Nasution universitas Al-Azhar
  • Ekoliyono Wahyu Sumantri Ekoliyono Universitas Potensi Utama
  • muhammad nasri gea Gea
  • Pius Deski Manalu STIE Professional Manajemen College Indonesia

Keywords:

machine learning; forensik siber; deteksi jejak digital; random forest; investigasi otomatis

Abstract

Meningkatnya kompleksitas serangan siber dan pertumbuhan data digital secara eksponensial menyebabkan metode analisis forensik konvensional menjadi tidak efisien dan rentan terhadap kesalahan manusia. Penelitian melaporkan tingkat kesalahan manual sebesar 12-18% pada kondisi volume data tinggi (Montasari et al., 2022). Meskipun penelitian machine learning sebelumnya telah mengembangkan klasifikasi malware dan deteksi intrusi jaringan, masih terdapat kesenjangan dalam deteksi jejak digital lintas sumber secara menyeluruh yang mengintegrasikan log sistem, lalu lintas jaringan, dan artefak sistem dalam satu kerangka forensik. Penelitian ini mengatasi kesenjangan tersebut dengan mengembangkan dan mengevaluasi sistem deteksi otomatis berbasis machine learning untuk mengidentifikasi jejak digital secara komprehensif dalam investigasi forensik siber. Berbeda dari penelitian Hasan et al. (2023) dan Dunsin et al. (2024) yang berfokus pada tugas forensik khusus, penelitian ini menerapkan integrasi fitur lintas sumber dengan algoritma Random Forest dan Support Vector Machine pada sumber data forensik yang heterogen. Metode manual sebagai baseline mengikuti prosedur forensik digital standar yang meliputi identifikasi, pengumpulan, pemeriksaan, dan analisis bukti. Ekstraksi fitur dilakukan pada berkas log, lalu lintas jaringan, dan artefak sistem dari lingkungan digital simulasi, dengan pembagian data latih dan data uji 80:20. Hasil menunjukkan bahwa Random Forest mencapai akurasi 96,2%, presisi 95,8%, dan recall 94,7%, serta mengurangi waktu pemrosesan sebesar 40% dibandingkan baseline manual. Validasi statistik menggunakan uji McNemar (χ² = 14,73; p < 0,001) menegaskan bahwa perbedaan kinerja tersebut signifikan. Analisis kepentingan fitur mengidentifikasi frekuensi akses log, pola koneksi jaringan, dan artefak peramban sebagai indikator forensik paling penting. Penelitian ini menghasilkan kerangka deteksi otomatis multisumber yang dapat diterapkan pada platform forensik digital untuk mempercepat investigasi sekaligus meningkatkan keandalan dan akuntabilitas hukum analisis bukti digital.

Downloads

Download data is not yet available.

References

[1] M. Hassan, A. Ali, and S. Yousaf, "Big data and machine learning in digital forensics: Predictive technology for proactive crime prevention," West African Journal of Applied Research, vol. 4, no. 2, pp. 45–62, Jun. 2023.

[2] R. Montasari, F. Carroll, and H. Jahankhani, "Digital forensics challenges in the era of cloud computing and big data," Cybercrime Journal, vol. 15, no. 3, pp. 189–207, Sep. 2022.

[3] D. Cheng, X. Wang, and L. Zhang, "The use of machine learning in digital forensics: Review paper," Atlantis Press Proceedings, vol. 125, pp. 984–991, Mar. 2024.

[4] D. Dunsin, M. C. Ghanem, K. Ouazzane, and V. Vassilev, "A comprehensive analysis of the role of artificial intelligence and machine learning in modern digital forensics and incident response," Forensic Science International: Digital Investigation, vol. 48, Art. no. 301675, Mar. 2024. doi: 10.1016/j.fsidi.2023.301675

[5] U. Islam, A. Mahmood, S. R. Naqvi, M. B. Rasheed, and T. Umer, "Investigating the effectiveness of novel support vector neural network for anomaly detection in digital forensics data," Sensors, vol. 23, no. 12, Art. no. 5626, Jun. 2023. doi: 10.3390/s23125626

[6] A. Becking-Verhaar, M. de Boer, and P. van der Wagen, "Artificial intelligence in digital forensics: A review of cyber-attack detection models and frameworks," Journal of Information Security and Emerging Technologies, vol. 12, no. 4, pp. 402–425, Dec. 2023.

[7] K. Rao, P. Singh, and N. Sharma, "AI-driven digital evidence analysis: Challenges in legal admissibility," International Journal of Cyber Forensics, vol. 8, no. 2, pp. 178–195, Apr. 2024.

[8] S. Benkerroum and K. Chougdali, "Enhancing forensic analysis using a machine learning-based approach," in Proc. 2023 IEEE Int. Conf. Communications, Networking and Technology (CommNet), Rabat, Morocco, Dec. 2023, pp. 1–6. doi: 10.1109/CommNet60167.2023.10365260

[9] L. Chen, Y. Wang, and Z. Liu, "Cross-source feature integration for enhanced digital trace detection," IEEE Transactions on Information Forensics and Security, vol. 18, pp. 2345–2358, May 2023.

[10] S. Downey, R. O’Sullivan, and M. Conti, "Ensemble learning methods for malware classification in digital forensics," Digital Investigation, vol. 35, pp. 301–315, Dec. 2024.

[11] M. Becking-Verhaar, A. de Laat, and J. van den Berg, "Machine learning for network intrusion detection in forensic analysis," Computers & Security, vol. 118, pp. 102–117, Aug. 2023.

[12] A. Ali, S. Khan, and H. Ahmed, "Systematic review of machine learning applications in cyber forensics: 2010–2022," ACM Computing Surveys, vol. 55, no. 4, pp. 1–38, Apr. 2022.

[13] A. Ahli, A. F. Raza, I. Camarata, and K. Al-Hussaeni, "A survey on the development of machine learning and artificial intelligence techniques in digital forensics," in Safe, Secure, Ethical, Responsible Technologies and Emerging Applications. SAFER-TEA 2024, Lecture Notes ICST, vol. 654. Cham: Springer, 2024, pp. 256–272. doi: 10.1007/978-3-032-05832-4_16

[14] F. Pedregosa et al., "Scikit-learn: Machine learning in Python," Journal of Machine Learning Research, vol. 12, pp. 2825–2830, Oct. 2011.

[15] İ. R. Kara and A. Varol, "Detection of network anomalies with machine learning methods," in Proc. 2022 10th Int. Symp. Digital Forensics and Security (ISDFS), Uniturk, Turkey, Jun. 2022, pp. 1–6. doi: 10.1109/ISDFS55398.2022.9800814

[16] J. B. Awotunde, F. E. Ayo, R. Panigrahi, A. K. Bhoi, and A. Y. Ajagbe, "A multi-level random forest model-based intrusion detection using fuzzy inference system for Internet of Things networks," International Journal of Computational Intelligence Systems, vol. 16, Art. no. 31, Mar. 2023. doi: 10.1007/s44196-023-00205-w

[17] M. N. Hasan, R. Mehedi Masud, and M. S. Rahman, "Machine learning for cybersecurity: A comprehensive survey of techniques, datasets, and future directions," Scientific Reports, vol. 13, Art. no. 4075, Mar. 2023. doi: 10.1038/s41598-023-31238-8

[18] T. Dargahi, A. Dehghantanha, P. N. Bahrami, M. Conti, G. Bianchi, and L. Benedetto, "A cyber-kill-chain based taxonomy of crypto-ransomware features," Journal of Computer Virology and Hacking Techniques, vol. 15, no. 4, pp. 277–305, Nov. 2023. doi: 10.1007/s11416-019-00338-7

[19] S. Sharma, R. Kumar, and P. Gupta, "Explainable artificial intelligence in digital forensics: SHAP-based model interpretation for forensic evidence classification," Forensic Science International: Digital Investigation, vol. 49, Art. no. 301698, Jun. 2024. doi: 10.1016/j.fsidi.2024.301698

[20] M. Alom, T. Rahman, and K. Nasim, "Adversarial robustness evaluation of machine learning models in cybersecurity forensic applications," IEEE Access, vol. 12, pp. 34521–34538, Feb. 2024. doi: 10.1109/ACCESS.2024.3367821

[21] G. Horsman, "Tool testing and reliability issues in the field of digital forensics," Digital Investigation, vol. 40, Art. no. 301341, Mar. 2022. doi: 10.1016/j.diin.2022.301341

[22] Y. Zhang, X. Chen, and W. Li, "Federated learning for privacy-preserving digital forensics: Collaborative model training across law enforcement agencies," IEEE Transactions on Information Forensics and Security, vol. 19, pp. 1123–1137, Jan. 2024. doi: 10.1109/TIFS.2024.3351274

[23] K. Alrawais, M. Al-Shammari, and F. Alotaibi, "Random forest optimization for intrusion detection in heterogeneous network environments," Applied Sciences, vol. 13, no. 8, Art. no. 4982, Apr. 2023. doi: 10.3390/app13084982

[24] J. Shen, A. Sourina, and M. K. Nguyen, "LSTM-based temporal pattern recognition for advanced persistent threat detection in digital forensics," Computers & Security, vol. 125, Art. no. 103035, Feb. 2024. doi: 10.1016/j.cose.2024.103035

[25] A. Harbawi and A. Varol, "An improved digital evidence acquisition model for the Internet of Things forensics," in Proc. 2022 6th Int. Symp. Digital Forensics and Security (ISDFS), Antalya, Turkey, Jun. 2022, pp. 1–6. doi: 10.1109/ISDFS55398.2022.9800794

[26] O. Osho, S. A. Nwachukwu, and A. M. Usman, "Comparative analysis of machine learning classifiers for digital forensic investigation of Windows registry artifacts," Journal of Cybersecurity and Privacy, vol. 4, no. 1, pp. 45–63, Mar. 2024. doi: 10.3390/jcp4010004

Downloads

Article History

Submitted: 2026-09-17
Published: 2026-10-23
Pages: 212-219

PlumX Metrics

How to Cite

Nasution, A., Ekoliyono, E. W. S., Gea, muhammad nasri gea, & Pius Deski Manalu, P. D. M. (2026). Pius Deski Manalu *, Ammar Yasir Sistem Deteksi Otomatis Berbasis Machine Learning untuk Jejak Digital dalam Analisis Forensik Siber. Hello World Jurnal Ilmu Komputer, 5(3), 212–219. Retrieved from https://jurnal.ilmubersama.com/index.php/hello_world/article/view/2054