PENERAPAN ALGORITMA SUPPORT VECTOR MACHINE UNTUK MENDETEKSI ANOMALI PADA JARINGAN KOMPUTER

REZA PUSPITA SARI POHAN, NPM 2009400118 (2024) PENERAPAN ALGORITMA SUPPORT VECTOR MACHINE UNTUK MENDETEKSI ANOMALI PADA JARINGAN KOMPUTER. Tugas_Akhir (Artikel) Jurnal Ilmu Komputer dan Sistem Informasi (JIKOMSI), 00 (00). pp. 174-182. ISSN 2723- 6129

	Text COVER.pdf Download (214kB)
	Text ARTIKEL.pdf Download (692kB)

Abstract

Anomali jaringan biasanya menunjukkan masalah atau ancaman keamanan potensial. Anomali jaringan dapat menyebabkan kerugian keuangan dan reputasi perusahaan serta merusak integritas, kerahasiaan, dan ketersediaan data. Teknik deteksi anomali tradisional menggunakan algoritma berbasis aturan memiliki keterbatasan dalam menemukan anomali yang beragam dan canggih. Sebaliknya, algoritma machine learning telah menunjukkan hasil yang luar biasa. Akibatnya, semakin banyak orang yang mulai menggunakan machine learning untuk mendeteksi anomali. Penelitian ini bertujuan untuk menerapkan algoritma support vector machine dalam mendeteksi anomali pada jaringan komputer. Tahapan penelitian dimulai dari pengumpulan dataset, prapemrosesan data, penerapan algoritma SVM, dan evaluasi hasil. Penelitian ini telah berhasil menerapkan algoritma support vector machine untuk mendeteksi anomali pada jaringan komputer. Menurut hasil matriks evaluasi kinerja, diperoleh akurasi keseluruhan sebesar 81,50%, presisi 74,66%, recall 100%, dan f1-score 85,49%. Hasil matriks ini menunjukkan bahwa model dapat dengan akurat mengidentifikasi sampel kelas Normal dan Anomali. Secara keseluruhan, temuan ini menunjukkan bahwa model algoritma support vector machine sangat baik untuk menemukan anomali jaringan. Hasil kurva receiver operating characteristic sebesar 0,97 menunjukkan bahwa model support vector machine dengan hiperparameter kernel RBF, C=10 dan gamma=auto yang dipilih sangat akurat dalam membedakan antara contoh positif dan negatif dalam kumpulan data lalu lintas jaringan. Penelitian ini diharapkan dapat membantu administrator jaringan menemukan dan membuat keputusan yang tepat tentang cara mengatasi anomali jaringan komputer. Kata Kunci : Anomali, Deteksi, Jaringan Komputer, Machine Learning, ROC, SVM ================================================================================================ Network anomalies usually indicate potential security issues or threats. Network anomalies can cause financial and reputational losses to companies and damage the integrity, confidentiality, and availability of data. Traditional anomaly detection techniques using rule-based algorithms have limitations in finding diverse and sophisticated anomalies. In contrast, machine-learning algorithms have shown remarkable results. As a result, more and more people are starting to use machine learning to detect anomalies. This study aims to apply the support vector machine algorithm to detect anomalies in computer networks. The research stages start with dataset collection, data preprocessing, application of the SVM algorithm, and evaluation of the results. This study has successfully applied the support vector machine algorithm to detect anomalies in computer networks. According to the results of the performance evaluation matrix, the overall accuracy was 81.50%, precision was 74.66%, recall was 100%, and f1-score was 85.49%. The results of this matrix indicate that the model can accurately identify normal and anomalous class samples. Overall, these findings indicate that the support vector machine algorithm model is very good for finding network anomalies. The result of the receiver operating characteristic curve of 0.97 indicates that the support vector machine model with the selected RBF kernel hyperparameters, C = 10 and gamma = auto is very accurate in distinguishing between positive and negative examples in the network traffic data set. This study is expected to help network administrators find and make the right decisions on how to overcome computer network anomalies. Keywords : Anomaly, Detection, Machine Learning, Network, ROC, SVM

Item Type:	Article
Uncontrolled Keywords:	Anomali, Deteksi, Jaringan Komputer, Machine Learning, ROC, SVM ============================== Anomaly, Detection, Machine Learning, Network, ROC, SVM
Subjects:	Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General)
Divisions:	Fakultas Sains Dan Teknologi > Sistem Informasi
Depositing User:	Unnamed user with email repository@ulb.ac.id
Date Deposited:	06 Jun 2024 07:35
Last Modified:	06 Jun 2024 07:35
URI:	http://repository.ulb.ac.id/id/eprint/763

Actions (login required)

View Item