Implementasi Algoritma Neutrosophic C-Means dan Firefly untuk Deteksi Intrusi pada Jaringan Internet of Things

Farindra, Laurensia Vira (2026) Implementasi Algoritma Neutrosophic C-Means dan Firefly untuk Deteksi Intrusi pada Jaringan Internet of Things. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5002211082-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Perkembangan pesat Internet of Things (IoT) meningkatkan pemanfaatan perangkat pintar sekaligus memperluas potensi serangan siber. Karakteristik lalu lintas jaringan IoT yang dinamis, heterogen, dan sarat ketidakpastian menuntut pendekatan deteksi intrusi yang mampu menangani data kompleks dan noisy. Penelitian ini mengimplementasikan Neutrosophic C-Means (NCM) sebagai metode deteksi intrusi berbasis anomali, yang memanfaatkan konsep himpunan neutrosophic melalui tiga komponen keanggotaan, yaitu truth, indeterminacy, dan falsity, untuk merepresentasikan tingkat kepastian, ambiguitas, serta kemungkinan data sebagai outlier. NCM diintegrasikan dengan Firefly Algorithm (FA) sebagai mekanisme optimisasi parameter klaster, meliputi nilai fuzziness, bobot komponen neutrosophic, dan parameter δ, sehingga membentuk pendekatan FA-NCM. Pengujian dilakukan menggunakan ACI IoT Network Traffic Dataset 2023 pada skenario klasifikasi biner dan multikelas, dengan kondisi data seimbang dan tidak seimbang. K-Means dan Fuzzy C-Means (FCM) digunakan sebagai metode pembanding. Hasil evaluasi menunjukkan bahwa pada skenario klasifikasi biner dengan data seimbang, FA-NCM menghasilkan kinerja terbaik dengan accuracy 80,26%, precision 95,20%, recall 77,14%, dan F1-score 85,22%. Dengan demikian, penelitian ini menunjukkan bahwa NCM dan FA-NCM efektif diterapkan sebagai pendekatan deteksi anomali pada sistem deteksi intrusi IoT, serta relevan untuk sistem keamanan jaringan yang memerlukan sensitivitas tinggi terhadap serangan.
================================================================================================================================
The rapid development of the Internet of Things (IoT) has increased the adoption of smart devices while simultaneously expanding the potential for cyberattacks. The dynamic, heterogeneous, and uncertainty-prone characteristics of IoT network traffic require intrusion detection approaches capable of handling complex and noisy data. This study implements the Neutrosophic C-Means (NCM) algorithm as an anomaly-based intrusion detection method, which leverages neutrosophic set theory through three membership components (truth, indeterminacy, and falsity) to represent certainty levels, ambiguity, and the likelihood of data being outliers. NCM is integrated with the Firefly Algorithm (FA) as a cluster parameter optimization mechanism, including fuzziness values, neutrosophic membership weights, and the δ parameter, forming the FA-NCM approach. Experiments are conducted using the ACI IoT Network Traffic Dataset 2023 under binary and multiclass classification scenarios with balanced and imbalanced data conditions. K-Means and Fuzzy C-Means (FCM) are employed as baseline comparison methods. Evaluation results indicate that under the balanced binary classification scenario, FA-NCM achieves the best performance, with an accuracy of 80.26%, precision of 95.20%, recall of 77.14%, and an F1-score of 85.22%. These findings demonstrate that NCM and FA-NCM are effective for anomaly detection in IoT intrusion detection systems and are particularly relevant for network security environments that require high sensitivity to cyberattacks.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	internet of things, deteksi intrusi, Neutrosophic C-Means, internet of things, intrusion detection, Neutrosophic C-Means
Subjects:	Q Science > QA Mathematics > QA278.55 Cluster analysis Q Science > QA Mathematics > QA9.58 Algorithms Q Science > QA Mathematics > QA248_Fuzzy Sets
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Laurensia Vira Farindra
Date Deposited:	04 Feb 2026 01:48
Last Modified:	04 Feb 2026 01:48
URI:	http://repository.its.ac.id/id/eprint/132074

Actions (login required)

View Item