DETEKSI SERANGAN DISTRIBUTED DENIAL OF SERVICE (DDoS) PADA APACHE SPARK MENGGUNAKAN METODE K-MEANS CLUSTERING
Total View This Week0
Total View This Week0
Institusion
Universitas Sriwijaya
Author
WICAKSANA, MOCHAMMAD RAFII NANDA
Heryanto, Ahmad
Heryanto, Ahmad
Subject
T1-995 Technology (General)
Datestamp
2023-05-12 05:00:58
Abstract :
Distributed Denial-of-Service (DDoS) is a collection of denial-of-service attacks that are carried out by executing commands from the master computer to a number of botnets which are infected hosts to attack certain targets. Because of this, DDoS attack detection is the first and most important way to counter DDoS attacks. The basis for carrying out the detection approach is using machine learning. K-means clustering is the simplest and most well-known clustering analysis algorithm in solving clustering problems. This algorithm is known to be efficient for large datasets. This paper proposes detecting DDoS attacks using an unsupervised learning method, namely K-means clustering on Apache Spark. This study used the CIC-DDoS2019 dataset from the University of New Brunswick (UNB) to train and perform experiments on the detection system used. This model produces the best evaluation results with the recall of 99,99%, precision of 99,99%, specificity of 88.24%, the accuracy of 99.98%, and F1 score of 99.99%.
Distributed Denial-of-Service (DDoS) is a collection of denial-of-service attacks that are carried out by executing commands from the master computer to a number of botnets which are infected hosts to attack certain targets. Because of this, DDoS attack detection is the first and most important way to counter DDoS attacks. The basis for carrying out the detection approach is using machine learning. K-means clustering is the simplest and most well-known clustering analysis algorithm in solving clustering problems. This algorithm is known to be efficient for large datasets. This paper proposes detecting DDoS attacks using an unsupervised learning method, namely K-means clustering on Apache Spark. This study used the CIC-DDoS2019 dataset from the University of New Brunswick (UNB) to train and perform experiments on the detection system used. This model produces the best evaluation results with the recall of 99,99%, precision of 99,99%, specificity of 88.24%, the accuracy of 99.98%, and F1 score of 99.99%.
File :
RAMA_56201_09011281823053.pdf
RAMA_56201_09011281823053_TURNITIN.pdf
RAMA_56201_09011281823053_0022018703_01_front_ref.pdf
RAMA_56201_09011281823053_0022018703_02.pdf
RAMA_56201_09011281823053_0022018703_03.pdf
RAMA_56201_09011281823053_0022018703_04.pdf
RAMA_56201_09011281823053_0022018703_05.pdf
RAMA_56201_09011281823053_0022018703_06_ref.pdf
RAMA_56201_09011281823053_0022018703_07_lamp.pdf
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
