Widya Amellia Putri, . (2026) PENERAPAN STREAMING ANALYTICS PADA DATA HONEYPOT UNTUK PENDETEKSIAN POLA SERANGAN BERBASIS APACHE FLINK. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (219kB) |
|
Text
AWAL.pdf Download (713kB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (299kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (577kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (555kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
BAB 5.pdf Download (228kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (232kB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository staff only Download (105kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (5MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (21MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (502kB) |
Abstract
Cyberattacks against network services continue to increase, while the monitoring mechanism for the Cowrie honeypot at the Cyber Security and Networking Laboratory of the Faculty of Computer Science, UPN "Veteran" Jakarta, is still performed through batch and manual processing, causing delays in detecting and responding to ongoing attack patterns. This condition drives the need for a detection system capable of processing honeypot data continuously and automatically. This research aims to design and implement an Apache Flink-based streaming analytics system that integrates k-Nearest Neighbors (kNN) and Logistic Regression classification models to detect normal and anomalous activities in Cowrie honeypot data, as well as to evaluate the effectiveness (accuracy, precision, recall, F1-score) and efficiency (latency) of both models. This research employs the experimental research method with a quantitative approach. Cowrie honeypot data is collected from the laboratory's MongoDB database, then processed through preprocessing stages including labeling, extraction of 19 features, class imbalance handling with SMOTE, and normalization. Both models were trained with training data and tested with testing data (80:20 ratio), then implemented through a PyFlink pipeline connected to PostgreSQL as the result storage. Hypothesis testing is conducted using descriptive analysis of per-event classification latency
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2210511052] [Pembimbing 1: I Wayan Widi Pradnyana] [Pembimbing 2: Neny Rosmawarni] [Penguji 1: Indra Permana Solihin] [Penguji 2: Ichsan Mardani] |
| Uncontrolled Keywords: | Apache Flink, Cowrie honeypot, k-Nearest Neighbors, Logistic Regression, streaming analytics |
| Subjects: | Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | WIDYA AMELLIA PUTRI |
| Date Deposited: | 25 Aug 2026 15:24 |
| Last Modified: | 25 Aug 2026 15:24 |
| URI: | http://repository.upnvj.ac.id/id/eprint/50972 |
Actions (login required)
 |
View Item |