KLASIFIKASI SERANGAN BOTNET MENGGUNAKAN METODE BI-DIRECTIONAL LONG SHORT-TERM MEMORY

TAUFIK, M. and Heryanto, Ahmad (2022) KLASIFIKASI SERANGAN BOTNET MENGGUNAKAN METODE BI-DIRECTIONAL LONG SHORT-TERM MEMORY. Undergraduate thesis, Sriwijaya University.

[thumbnail of RAMA_56201_09011281823073.pdf] Text
RAMA_56201_09011281823073.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (3MB) | Request a copy
[thumbnail of RAMA_56201_09011281823073_TURNITIN.pdf] Text
RAMA_56201_09011281823073_TURNITIN.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (5MB) | Request a copy
[thumbnail of RAMA_56201_09011281823073_0022018703_01_front_ref.pdf]
Preview
Text
RAMA_56201_09011281823073_0022018703_01_front_ref.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.

Download (1MB) | Preview
[thumbnail of RAMA_56201_09011281823073_0022018703_02.pdf] Text
RAMA_56201_09011281823073_0022018703_02.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (1MB) | Request a copy
[thumbnail of RAMA_56201_09011281823073_0022018703_03.pdf] Text
RAMA_56201_09011281823073_0022018703_03.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (665kB) | Request a copy
[thumbnail of RAMA_56201_09011281823073_0022018703_04.pdf] Text
RAMA_56201_09011281823073_0022018703_04.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (1MB) | Request a copy
[thumbnail of RAMA_56201_09011281823073_0022018703_05.pdf] Text
RAMA_56201_09011281823073_0022018703_05.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (201kB) | Request a copy
[thumbnail of RAMA_56201_09011281823073_0022018703_06_ref.pdf] Text
RAMA_56201_09011281823073_0022018703_06_ref.pdf - Bibliography
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (330kB) | Request a copy
[thumbnail of RAMA_56201_09011281823073_0022018703_07_lamp.pdf] Text
RAMA_56201_09011281823073_0022018703_07_lamp.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (631kB) | Request a copy

Abstract

Botnet attacks have become a major threat on the internet in recent years. Because a botnet is a collection of programs in which there is malware, both of which are connected to each other within the scope of the internet network, which can communicate with a collection of similar malware bot programs to do work that is detrimental and targets the intended victim. There are three objectives in this research, among others, to build a model of the Bi-Directional LSTM method for the ability to classify botnet attacks on the CIC-IDS 2018 dataset. Second, apply PCA feature selection to optimize the classification of botnet attacks. And thirdly Knowing the results of the classification performance of Botnet attacks seen from the results of accuracy, specificity, recall, precision. Therefore, to overcome the previous problem, the deep learning method was used. The Deep Learning method used is the BI-Directional LSTM method which is a branch of LSTM which has the advantage of having two layers, namely the forward layer and the backward layer so that it allows additional information enhancement and improves memory capabilities. This research has three benefits, including applying the Bi-Directional Long Short-Term Memory method for classifying Botnet attacks. The second is to optimize the Bi-Directional Long Short Term Memory method so as to get a high accuracy value. The third is to find out the performance of Bi-Directional Long Short-Term Memory results to classify Botnet attacks. This research was conducted by training the 2018 CIC-IDS dataset on machine learning with the provision of tuning hyperparameters and comparing results with different ratios of training data and test data so that the best evaluation results were obtained with an accuracy value of 99.82% accuracy, 99.76% precision, 99.89 recall. %, and a specificity of 99.82%.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Botnet, LSTM, Bi-Directional LSTM, PCA, Dataset CIC-IDS-2018
Subjects: Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA75 Electronic computers. Computer science
Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA75.5.A142 Computer science. Information society. Information technology.
Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA76.9.A25 Computer security. Systems and Data Security.
Q Science > QA Mathematics > QA75-76.95 Calculating machines > QA76.9.E94 Computer system performance. Computer Communication Networks. Computer science. Logic design. Operating systems (Computers).
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5101-6720 Telecommunication Including telegraphy, telephone, radio, radar, television > TK5105.5.S72 Computer networks
Divisions: 09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User: Users 26759 not found.
Date Deposited: 09 Jan 2023 04:13
Last Modified: 09 Jan 2023 04:13
URI: http://repository.unsri.ac.id/id/eprint/85403

Actions (login required)

View Item View Item