Rendy Sahrasad, . (2026) ANALISIS KINERJA ALGORITMA LONG SHORT-TERM MEMORY DALAM DETEKSI INTRUSI JARINGAN BERBASIS LOG HONEYPOT COWRIE SSH. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (12kB) |
|
Text
AWAL.pdf Download (594kB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (29kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (405kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (411kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (676kB) |
|
Text
BAB 5.pdf Download (16kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (88kB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository staff only Download (100kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (68kB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (20MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (261kB) |
Abstract
The increasing use of computer networks requires an intrusion detection system capable of identifying suspicious activity patterns in a structured manner. This study aims to implement a Long Short-Term Memory (LSTM)-based Network Intrusion Detection System (NIDS) model to classify activities in Cowrie SSH honeypot logs. The log data were processed through Cowrie SSH data selection, data cleaning, session-based event sequence formation, multiclass label construction, feature engineering, removal of features causing data leakage and proxy labels, tokenization, padding, and data splitting into training, validation, and testing sets. The target classification categories consist of normal, scan_probe, and brute_force. The LSTM model was evaluated using accuracy, precision, recall, F1-score, and a confusion matrix. The evaluation results show that the model achieved an accuracy of 99.30%, with macro average and weighted average precision, recall, and F1-score values of 99.30%. In addition, the model was integrated into a NIDS dashboard prototype using Flask, MongoDB, and Docker. The testing results demonstrate that the system can read Cowrie SSH logs, generate session tokens, perform predictions, store results, and display classifications with a LOGGED status. The developed system operates as a detection-only mechanism and does not perform automatic blocking.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2210511132] [Pembimbing: Supriyanto] [Pembimbing2: Hamonangan Kinantan Prabu] [Penguji 1: Noor Falih] [Penguji 2: Novi Trisman Hadi] |
| Uncontrolled Keywords: | Cowrie SSH, Honeypot, Long Short-Term Memory (LSTM), Multiclass, Network Intrusion Detection System (NIDS). |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4450 Databases |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | RENDY SAHRASAD |
| Date Deposited: | 29 Jul 2026 08:28 |
| Last Modified: | 08 Sep 2026 04:57 |
| URI: | http://repository.upnvj.ac.id/id/eprint/52110 |
Actions (login required)
 |
View Item |