Analisis Pengenalan Pola Dalam Jaringan Demiliterized Zone Menggunakan Teknik Klasifikasi Decision Tree Data Mining

Gunawan, Jimmy Agung (2008) Analisis Pengenalan Pola Dalam Jaringan Demiliterized Zone Menggunakan Teknik Klasifikasi Decision Tree Data Mining. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 9104205303-Master_thesis.pdf Restricted to Repository staff only Download (8MB)

Abstract

Di zaman yang serba internet ini, kebutuhan akan suatu sistem pengamanan dalam membangun jaringan komputer tidak dapat dielakkan lagi. Sistem administrator wajib memasang pengamanan, baik berupa firewall maupun Intrusion Detection System (IDS), untuk memeriksa aliran data yang keluar dan masuk dari jaringan internal ke jaringan eksternal (internet). Jaringan internal dalam skala pelayanan dan pengaksesan publik itu sendiri dikelompokkan dalam satu lingkup demilitarized zone (DMZ) agar tidak bercampur dengan sistem pada jaringan lokal. Dalam kurun perkembangan waktu, penyimpanan data log lalu lintas data yang semakin bertambah jumlahnya serta menjamurnya teknik-teknik penyerangan dan penyusupan baru sering kali menyebabkan pendeteksian serangan sulit dikontrol oleh analis sistem keamanan. Untuk menanggapi persoalan tersebut, dalam penelitian ini diujicobakan penggunaan teknik data mining dengan algoritma yang termasuk dalam kelompok supervised untuk melakukan klasifikasi pada tiap pola yang terbentuk, khususnya dalam pengenalan pola-pola data normal dan serangan. Data yang digunakan dalam penelitian ini berasal dari Knowledge Discovery in Databases (KDD), yaitu KDDCup'99, dan diaplikasikan algoritma data mining dari decision tree untuk menemukan pola-pola contoh data normal dan serangan.
==================================================================================================================================
The necessity of a security system when building computer networks is increasingly important, especially in today's internet-based computing environment. System administrators have to implement safeguards, such as firewalls or Intrusion Detection Systems (IDS), to monitor data flows in and out of the internal network to the external network (internet). Internal networks that provide public services and access are grouped within a demilitarized zone (DMZ) to prevent them from mixing with systems on the local network. Over time, the amount of stored data traffic logs has continued to increase, while new intrusion and network attack techniques have also emerged. This condition makes it increasingly difficult for security system analysts to control and detect attacks. To address this problem, this research examines data mining techniques using algorithms under the supervised learning method to classify each identified pattern, particularly patterns representing normal and intrusion data. The KDDCup'99 dataset from Knowledge Discovery in Databases (KDD) is used as the network traffic data in this research, and data mining algorithms based on decision trees are applied to identify sample patterns of normal and intrusion data.

Item Type:	Thesis (Masters)
Additional Information:	RTMT 006.312 Gun a
Uncontrolled Keywords:	data mmmg, decision tree, demiliterized zone, pendeteksian intrusi, keamanan jaringan, KDDCup'99, signature-based, data mmmg, intrusion detection, demiliterized zone, network security, signature-based, KDDCup'99
Subjects:	Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
Divisions:	61101-Magister Management Technology
Depositing User:	magang .
Date Deposited:	17 Sep 2026 03:28
Last Modified:	17 Sep 2026 03:28
URI:	http://repository.its.ac.id/id/eprint/144622

Actions (login required)

View Item