Sistem Klasifikasi Anomali Jaringan Komputer Bedasarkan Quality of Service Menggunakan Backpropagation Neural Network

Kuswanto, Isaac Ronggo (2023) Sistem Klasifikasi Anomali Jaringan Komputer Bedasarkan Quality of Service Menggunakan Backpropagation Neural Network. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 10311910000021-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (2MB) | Request a copy

Abstract

Suatu industri memiliki beberapa mesin manufaktur, dimana mesin mesin tersebutmemiliki sistem pengambilan data yang berfungsi untuk mengidentifikasi tingkat produktivitasmesin yang terdiri dari data run time, setup time, idle time, breakdown time, support time, actualproduct, good product dan last breakdown time. Mesin mesin tersebut berada pada lokasi kantorA dan mengirimkan data yang telah disimpan menggunakan jaringan internet pada kantor dilokasiB secara berkala. Pada prosses pengiriman data tersebut memiliki permasalahan, bahwa data padalokasi A terkadang tidak terkirim ke kantor lokasi B. Pada Proyek Akhir ini, dibuatlah sebuahsistem yang digunakan untuk mengklasifikasi kondisi dari sebuah jaringan komputermenggunakan Artificial Neural Network (ANN) dengan algoritma Backpropagation yang dilatihdari data Round Trip Time (RTT) atau latensi jaringan. Hasil yang telah didapat setelah ANNdilatih dengan total epoch 10,000, ANN dapat mendeteksi kondisi jaringan abnormal yangdisebabkan oleh banyaknya pengguna diatas kondisi normal yaitu sekitar 300 pengguna dengan presentase 100%, kondisi jaringan abnormal yang disebabkan oleh kerusakan atau masalah padaperangkat keras dengan presentase 100%, dan mendeteksi kondisi jaringan normal dengan presentase 30%, setelah mendapatkan klasifikasi nilai RTT, hasil dari klasifikasi tersebut dapatdiasosiasikan dengan penyebab permasalahan jaringan tertentu yang dipaparkan pada halamanwebsite.
=================================================================================================================================
An industry has several manufacturing machines, where these machines have a data retrieval system that functions to identify the level of machine productivity consisting of run time,setup time, idle time, breakdown time, support time, actual product, good product and last breakdown time data. . These machines are located at office location A and transmit data that has been stored using the internet network at the office at location B on a regular basis. In the process of sending data there is a problem, that data at location A is sometimes not sent to the office of location B. In this final project, a system is created that is used to classify the conditions of a computer network using an Artificial Neural Network (ANN) with a Backpropagation algorithm that trained from Round Trip Time (RTT) data or network latency. The results obtained after ANN was trained with a total of 10,000 epochs, ANN can detect abnormal network conditions caused by the number of users above normal conditions, namely around 300 users with a percentage of100%, abnormal network conditions caused by damage or problems in hardware with a percentage of 100%, and detecting normal network conditions with a percentage of 30%, after obtaining the RTT value classification, the results of the classification can be associated with the causes of certain network problems described on the website page.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Jaringan Komputer, Artificial Neural Network, Klasifikasi Data, Backpropagation, Quality of Service, Computer Networks, Data Classification
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.5 Information technology. IT--Auditing
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Isaac Ronggo Kuswanto
Date Deposited:	07 Dec 2023 02:04
Last Modified:	07 Dec 2023 02:04
URI:	http://repository.its.ac.id/id/eprint/104238

Actions (login required)

View Item