Investigasi log jaringan untuk deteksi serangan Distributed Denial Of Service (DDOS) dengan menggunakan metode general \nREGRESSION neural network
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
INDONESIA: \n \nSalah satu jenis serangan di dunia maya dengan intensitas yang cukup besar yaitu serangan Distributed Denial of Service (DDoS). Dibutuhkan sistem deteksi intrusi yang efektif dan akurat dalam mendeteksi serangan pada data intrusi jaringan untuk mengatasi hal tersebut. Oleh sebab itu, penelitian ini bertujuan untuk mengimplementasikan pendekatan baru dalam mendeteksi serangan pada data intrusi jaringan dengan tingkat akurasi yang baik. Metode yang diusulkan yaitu General Regression Neural Network yang dibantu dengan Random Forest dalam menyeleksi fitur sehingga mampu meningkatkan akurasi deteksi dan mempercepat waktu komputasi. Data latih yang digunakan yaitu CICIDS2017 dari Canadian Institute for Cybersecurity, sedangkan data uji yang digunakan yaitu log jaringan yang didapatkan dari simulasi serangan DDoS pada server web. Percobaan pertama menggunakan 69 fitur, diperoleh tingkat akurasi sebesar 66,41% dengan waktu pelatihan selama 1 jam 45 menit 6 detik. Adapun percobaan kedua menggunakan fitur terpilih yaitu sebanyak 20 fitur, diperoleh tingkat akurasi sebesar 97,21% dengan waktu pelatihan selama 42 menit 27 detik. Dari hasil percobaan tersebut, dapat disimpulkan bahwa General Regression Neural Network memiliki kemampuan deteksi dan klasifikasi yang cukup baik terhadap serangan DDoS pada data intrusi jaringan. \n \n \nENGLISH: \n \n \nOne type of attack in cyberspace with a large enough intensity is the Distributed Denial of Service (DDoS) attack. To overcome this, an effective and accurate intrusion detection system needed to detect attacks on network intrusion data. Therefore, this study aims to implement a new approach in detecting attacks on network intrusion data with a good rate of accuracy. The proposed method is the General Regression Neural Network which is assisted by Random Forest in selecting features so as to improve detection accuracy and speed up computing time. The training data used is CICIDS2017 from the Canadian Institute for Cybersecurity, while the test data used are network logs obtained from a simulation of DDoS attacks on a web server. The first experiment using 69 features, obtained an accuracy rate of 66.41% with training time for 1 hour 45 minutes 6 seconds. As for the second experiment using selected features which is 20 features, obtained an accuracy rate of 97.21% with training time for 42 minutes 27 seconds. From the results of these experiments, it can be concluded that the General Regression Neural Network has a fairly good detection and classification ability against DDoS attacks on network intrusion data.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.003 | 0.004 |
| Meta-epidemiology (broad) | 0.004 | 0.001 |
| Bibliometrics | 0.004 | 0.008 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.000 | 0.004 |
| Open science | 0.010 | 0.003 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it