PENGEMBANGAN MODEL REINFORCEMENT LEARNING DENGAN ALGORITMA DEEP Q NETWORK (DQN) DAN PROXIMAL POLICY OPTIMIZATION (PPO) UNTUK MULTIKLASIFIKASI SERANGAN SIBER

FIRJATULLAH, FARREL and Heryanto, Ahmad (2025) PENGEMBANGAN MODEL REINFORCEMENT LEARNING DENGAN ALGORITMA DEEP Q NETWORK (DQN) DAN PROXIMAL POLICY OPTIMIZATION (PPO) UNTUK MULTIKLASIFIKASI SERANGAN SIBER. Undergraduate thesis, Sriwijaya University.

Preview	Image RAMA_56201_09011382126164_cover.jpg - Cover Image Available under License Creative Commons Public Domain Dedication. Download (615kB) | Preview
	Text RAMA_56201_09011382126164.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_56201_09011382126164_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (15MB) | Request a copy
	Text RAMA_56201_09011382126164_0022018703_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (871kB)
	Text RAMA_56201_09011382126164_0022018703_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (828kB) | Request a copy
	Text RAMA_56201_09011382126164_0022018703_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_56201_09011382126164_0022018703_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_56201_09011382126164_0022018703_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (298kB) | Request a copy
	Text RAMA_56201_09011382126164_0022018703_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (278kB) | Request a copy
	Text RAMA_56201_09011382126164_0022018703_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy

Abstract

In digital era right now, with the rapid advancement of time, system and network security is very important in the digital communication environment. Machine learning techniques are currently one of the many methods used to address this, including a method known as Reinforcement Learning (RL), which can be used to solve classification problems. In this study, we discuss the application of the Deep Q Network (DQN) and Proximal Policy Optimization (PPO) algorithms from the Reinforcement Learning model to detect multiclass classification of cyber attacks. The impleme’tation was conducted using the ISCX 2012, CICDDoS 2019, KDDCup 1999, and NSL-KDD datasets, and RL successfully learned attack patterns effectively. To evaluate the effectiveness of the model, performance measurements were carried out using the DQN algorithm and achieved an accuracy of 89.27% on the ISCX 2012 dataset, 97.49% on the CICDDoS 2019 dataset, 94.48% on the KDDCup 1999 dataset, and 86.83% on the NSL-KDD dataset. Meanwhile, the PPO algorithm achieved 84.00% on the ISCX 2012 dataset, 87.00% on the CICDDoS 2019 dataset, 95.00% on the KDDCup 1999 dataset, and 85.19% on the NSL-KDD dataset.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Reinforcement Learning, Multiklasifikasi, Deep Q Network, Proximal Policy Optimization
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Farrel Firjatullah
Date Deposited:	12 Jun 2025 01:44
Last Modified:	12 Jun 2025 03:32
URI:	http://repository.unsri.ac.id/id/eprint/175393

Actions (login required)

View Item