Skripsi

INTRUSION RESPONSE SYSTEM SERANGAN SIBER MAN-IN-THE-MIDDLE DENGAN METODE RANDOM FOREST

---

The rapid adoption of Cyber-Physical Systems (CPS) has improved operational efficiency across critical sectors but has simultaneously increased exposure to cyber threats, particularly Man-in-the-Middle (MITM) attacks that covertly intercept and manipulate communication. In CPS environments, such attacks pose serious risks to system reliability and operational safety, thereby requiring security mechanisms capable of both accurate detection and automated response. This study proposes an Intrusion Response System (IRS) for detecting and mitigating MITM DNAT Hijacking attacks using a Random Forest classification model. Network traffic was collected under normal and attack scenarios, extracted into flow-based features using CICFlowMeter, and processed through labeling, feature selection, data splitting, and class balancing. The trained Random Forest model was integrated into the IRS to enable real-time responses, including automatic attacker IP blocking. Experimental evaluation shows that the proposed system achieves an accuracy of 0.868, precision of 1.000, recall of 0.827, and F1-score of 0.905, with an average response time of 43.57 ms per flow, a resilience score of 0.8397 based on K-Fold Cross Validation, and consistent inference latency indicating stable system operation. Overall, the results demonstrate that the proposed Random Forest–based IRS provides accurate, resilient, and real-time mitigation of MITM attacks in CPS environments.

---

Availability

Detail Information

Series Title

-

Call Number

T1931252026

Publisher

Indralaya : Prodi Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya., 2026

Collation

xiv, 91 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

005.820 7

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Sistem Komputer
Pengamanan Data Komputer

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment