| Presentation | 2017-12-15 A study of gini importance-based header feature selection methods for cyber attack detection Yuta Kazato, Yuichi Nakatani, Takeshi Okamoto, Akira Shibata, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Growing the expansion of network services and IoT devices, the risk of cyber attacks is increased by high-frequency and sophistication. Therefore, network service providers must protect their users and services against the huge attacks. Network header feature based machine learning methods for cyber attack detection are effective from the viewpoint of data payload encryption, but non-optimal feature selection result affect both learning time and accuracy. In this paper, we propose the Gini importance-based header feature selection methods for general machine learning algorithms that can detect cyber attacks with performances and efficiencies. We evaluate our feature selection methods in UNSW-NB15 dataset, a comprehensive dataset for network IDSs. As a result, we show the high-accuracy and average 99% detection rates compared with the previous methods. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Cyber attack detection / Feature selection / Gini importance / Machine learning |
| Paper # | IN2017-61 |
| Date of Issue | 2017-12-07 (IN) |
| Conference Information | |
| Committee | IA / IN |
|---|---|
| Conference Date | 2017/12/14(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Hiroshima City Univ. |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Performance Analysis and Simulation, Robustness, Traffic and Throughput Measurement, Quality of Service (QoS) Control, Congestion Control, Overlay Network/P2P, IPv6, Multicast, Routing, DDoS, etc. |
| Chair | Katsuyoshi Iida(Hokkaido Univ.) / Katsunori Yamaoka(Tokyo Inst. of Tech.) |
| Vice Chair | Rei Atarashi(IIJ) / Hiroyuki Osaki(Kwansei Gakuin Univ.) / Tomoki Yoshihisa(Osaka Univ.) / Takuji Kishida(NTT) |
| Secretary | Rei Atarashi(Tokyo Metropolitan Univ.) / Hiroyuki Osaki(TOYOTA-IT) / Tomoki Yoshihisa(NTT) / Takuji Kishida(NTT) |
| Assistant | Kenji Ohira(Tokushima Univ.) / Ryohei Banno(NTT) / Toshiki Watanabe(NEC) |
| Paper Information | |
| Registration To | Technical Committee on Internet Architecture / Technical Committee on Information Networks |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A study of gini importance-based header feature selection methods for cyber attack detection |
| Sub Title (in English) | |
| Keyword(1) | Cyber attack detection |
| Keyword(2) | Feature selection |
| Keyword(3) | Gini importance |
| Keyword(4) | Machine learning |
| 1st Author's Name | Yuta Kazato |
| 1st Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
| 2nd Author's Name | Yuichi Nakatani |
| 2nd Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
| 3rd Author's Name | Takeshi Okamoto |
| 3rd Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
| 4th Author's Name | Akira Shibata |
| 4th Author's Affiliation | Nippon Telegraph and Telephone Corporation(NTT) |
| Date | 2017-12-15 |
| Paper # | IN2017-61 |
| Volume (vol) | vol.117 |
| Number (no) | IN-353 |
| Page | pp.pp.91-96(IN), |
| #Pages | 6 |
| Date of Issue | 2017-12-07 (IN) |