Siraj, Muhammad Jaya (2021) Pengaruh Seleksi Fitur Terhadap Performa Sistem Deteksi Intrusi. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 05111740000088-Undergraduate_Thesis.docx]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
05111740000088-Undergraduate_Thesis.docx - Accepted Version Download (813kB) |
|
Text
05111740000088-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (810kB) | Request a copy |
Abstract
Kemajuan teknologi mengakibatkan mudahnya informasi untuk diakses oleh kalangan publik. Tetapi ada banyak ancaman termasuk malware, peretas yang dapat menyebabkan kerusakan serius jika tidak ditangani dengan benar. Untuk mengatasi masalah ini Intrution Detection System atau IDS diimplementasikan. Intrusion Detection System dapat memberi tahu sistem komputer apabila terjadi serangan pada jaringan tersebut. Tetapi karena banyaknya fitur yang ada pada traffic jaringan, untuk mendeteksi apakah data tersebut serangan atau tidak membutuhkan banyak waktu. Salah satu cara untuk melakukan optimasi adalah dengan seleksi fitur. Dengan menghapus fitur yang tidak diinginkan, performa secara keseluruhan dapat ditingkatkan sekaligus mengurangi waktu komputasi. Pada penelitian ini diusulkan seleksi fitur baru menggunakan uji ANOVA-f dan Sequential Feature Selection (SFS). Dengan menggunakan uji ANOVA-f kita akan mendapatkan jumlah fitur yang optimal, dan kita akan menentukan fitur terbaik yang dipilih menggunakan SFS. Hasil percobaan akan diukur dengan menggunakan metrik evaluasi seperti akurasi, spesifisitas, dan sensitivitas pada beberapa dataset.
=====================================================================================================
The advancement of technology has made resource sharing easier using the internet. But there are a lot of threat including malware, hackers that can lead to a serious damage if not handled properly. To addres this problem Intrution Detection System or IDS is implemented. Intrusion Detection System help computer system notify when there is an attack on a network. But because of the number of features that is present on a network capture data classifying the data takes a lot of time. One of the ways to optimize is by feature selection. By removing unwanted features, the overall performance can be increase while decreasing computational time. In this research we proposed a new feature selection using ANOVA-f test and Sequential Feature Selection (SFS). By using ANOVA-f test we will get the optimal number of features, and we will determine the best feature selected using SFS. The result will be evaluated using evaluating metric such as accuracy, specificity, and sensitivity over several dataset.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Uncontrolled Keywords: | Intrusion Detection System, Network Security, Feature Selection, Anova-F Test, Sequential Feature Selection |
| Subjects: | Q Science > QA Mathematics > QA76.9.A25 Computer security. Digital forensic. Data encryption (Computer science) |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Muhammad Jaya Siraj |
| Date Deposited: | 04 Aug 2021 23:46 |
| Last Modified: | 04 Aug 2021 23:46 |
| URI: | http://repository.its.ac.id/id/eprint/84863 |
Actions (login required)
 |
View Item |