ANALISIS SELEKSI FITUR DALAM MENENTUKAN SERANGAN MENGGUNAKAN ALGORITMA DECISION TREE

• Main Author: Hidayah Husnul Khotimah, (Author)
• Format: Book
• Published: 2020-06-17.
• Subjects: Thesis; NonPeerReviewed
• Online Access: Link Metadata
• Notes: http://repository.upnvj.ac.id/6834/1/ABSTRAK.pdf; http://repository.upnvj.ac.id/6834/2/AWAL.pdf; http://repository.upnvj.ac.id/6834/3/BAB%201.pdf; http://repository.upnvj.ac.id/6834/5/BAB%202.pdf; http://repository.upnvj.ac.id/6834/4/BAB%203.pdf; http://repository.upnvj.ac.id/6834/6/BAB%204.pdf; http://repository.upnvj.ac.id/6834/7/BAB%205.pdf; http://repository.upnvj.ac.id/6834/8/DAFTAR%20PUSTAKA.pdf; http://repository.upnvj.ac.id/6834/9/RIWAYAT%20HIDUP.pdf; http://repository.upnvj.ac.id/6834/10/LAMPIRAN.pdf; http://repository.upnvj.ac.id/6834/11/ARTIKEL%20KI.pdf
• Keywords: T Technology (General)

Cybercrime is a crime in cyberspace that utilizes technology and telecommunications as its medium. One of the cybercrime is cyber attack. Cyber attack is an attack on information systems, infrastructure, computer networks, and computer security systems. Therefore, there needs to be a tool that can be used to detect an attack called the Intrusion Detection System (IDS). The result of this IDS is in the form of an attack log which will later be analyzed using Security Event Management (SEM) by the network security administrator. To make network security administrators easier to analyze SEM, a feature selection method is needed that functions to determine what features are influential in detecting attacks. The method used in feature selection is importance ranking with the decision tree classification algorithm as an accuracy calculation. From the research results obtained important features, which consist of source ip, source port, signature id, rev, signature, severity, source city name, destination regional name, and destination city name with classification accuracy of 99.97%, training time for 0.021208 s, as well as the testing time of 0.001554 s.