PENGENALAN POLA SERANGAN PING FLOOD DENGAN ALGORITMA K-MEANS PADA JARINGAN INTERNET of THINGS (IoT)
Total View This Week0
Total View This Week0
Institusion
Universitas Sriwijaya
Author
MEILINDA EKA SURYANI (STUDENT ID : 09011181320033)
Deris Stiawan (LECTURER ID : 0003047905)
Deris Stiawan (LECTURER ID : 0003047905)
Subject
T58.6-58.62 Management information systems
Datestamp
2019-10-01 07:38:20
Abstract :
This research is focused on pattern recognition of ping flood attack on Internet of Things (IoT). The research was conducted on WiFi communication with normal traffic, attack traffic, and the combination of normal and attack traffic. From the scenario, normal dataset, attack dataset, and a combination of normal and attack dataset are generated. The testing was performed by grouping the datasets into two clusters, namely: (i) normal cluster and (ii) attack cluster, using Weka and implementation of K-Means algorithm. The result of clustering with Weka and implementation of K-Means algorithm shows that has average 95.931 packages in attack cluster, and 4.068 packages in normal cluster. The accuracy of clustering result is then calculated using confusion matrix equation. Based on confusion matrix calculation, the accuracy of clustering using K-Means algorithm is very good, reaching 99,94% with 98,62% true negative rate, 100% true positive rate, 0% false negative rate, and 1,38% false positive rate.
This research is focused on pattern recognition of ping flood attack on Internet of Things (IoT). The research was conducted on WiFi communication with normal traffic, attack traffic, and the combination of normal and attack traffic. From the scenario, normal dataset, attack dataset, and a combination of normal and attack dataset are generated. The testing was performed by grouping the datasets into two clusters, namely: (i) normal cluster and (ii) attack cluster, using Weka and implementation of K-Means algorithm. The result of clustering with Weka and implementation of K-Means algorithm shows that has average 95.931 packages in attack cluster, and 4.068 packages in normal cluster. The accuracy of clustering result is then calculated using confusion matrix equation. Based on confusion matrix calculation, the accuracy of clustering using K-Means algorithm is very good, reaching 99,94% with 98,62% true negative rate, 100% true positive rate, 0% false negative rate, and 1,38% false positive rate.
File :
RAMA_56201_09011181320033_0003047905_01_front_ref.pdf
RAMA_56201_09011181320033_0003047905_02.pdf
RAMA_56201_09011181320033_0003047905_03.pdf
RAMA_56201_09011181320033_0003047905_04.pdf
RAMA_56201_09011181320033_0003047905_05.pdf
RAMA_56201_09011181320033_0003047905_06_ref.pdf
RAMA_56201_09011181320033_0003047905_07_lamp.pdf
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
