| Presentation | 2022-06-23 Discussion about improving a detection accuracy of malware variants using time series differences in latent representation. Atsushi Shinoda, Hajime Shimada, Yukiko Yamaguti, Hirokazu Hasegawa, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Today, computers are used for various purposes to support people's daily lives. Therefore, the existence of malware that targets those computers is a huge threat. Anti-virus software vendors are taking measures to detect and remove malware, but attackers are also searching for avoiding methods and attacking with malware. In particular, if an attacker creates malware variant with concept drift which includes large function expansion or changes in implementation, a detection accuracy of them will decrease largely. Thus, in malware detection, there is a need for a methods that can detect changed malware variants that cannot be detected due to concept drift. In this study, we discuss about methods for improving a detection accuracy of a certain malware variants and its results. We tried to extract a time series differences in collection date for a malware family that concept drift occurred. The extracted difference represents a time series change of malware features, and by adding this to another malware family, we tried to generate a pseudo concept drift. By adding pseudo concept drift data to training data, we aimed to improve a detection accuracy of malware variants. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | malware variants / latent representation / data augmentation |
| Paper # | IA2022-4,ICSS2022-4 |
| Date of Issue | 2022-06-16 (IA, ICSS) |
| Conference Information | |
| Committee | IA / ICSS |
|---|---|
| Conference Date | 2022/6/23(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Univ. of Nagasaki |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Internet Security, etc. |
| Chair | Tomoki Yoshihisa(Osaka Univ.) / Katsunari Yoshioka(Yokohama National Univ.) |
| Vice Chair | Toru Kondo(Hiroshima Univ.) / Yuichiro Hei(KDDI Research) / Hiroshi Yamamoto(Ritsumeikan Univ.) / Kazunori Kamiya(NTT) / Takahiro Kasama(NICT) |
| Secretary | Toru Kondo(Osaka Univ.) / Yuichiro Hei(Kogakuin Univ.) / Hiroshi Yamamoto(NEC) / Kazunori Kamiya(KDDI labs.) / Takahiro Kasama(Okayama Univ.) |
| Assistant | Daisuke Kotani(Kyoto Univ.) / Ryo Nakamurai(Fukuoka Univ.) / Daiki Nobayashi(Kyushu Inst. of Tech.) / Keisuke Kito(Mitsubishi Electric) / Takeshi Sugawara(Univ. of Electro-Comm.) |
| Paper Information | |
| Registration To | Technical Committee on Internet Architecture / Technical Committee on Information and Communication System Security |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Discussion about improving a detection accuracy of malware variants using time series differences in latent representation. |
| Sub Title (in English) | |
| Keyword(1) | malware variants |
| Keyword(2) | latent representation |
| Keyword(3) | data augmentation |
| 1st Author's Name | Atsushi Shinoda |
| 1st Author's Affiliation | Nagoya University(Nagoya Univ.) |
| 2nd Author's Name | Hajime Shimada |
| 2nd Author's Affiliation | Nagoya University(Nagoya Univ.) |
| 3rd Author's Name | Yukiko Yamaguti |
| 3rd Author's Affiliation | Nagoya University(Nagoya Univ.) |
| 4th Author's Name | Hirokazu Hasegawa |
| 4th Author's Affiliation | National Institute of Informatics(NII) |
| Date | 2022-06-23 |
| Paper # | IA2022-4,ICSS2022-4 |
| Volume (vol) | vol.122 |
| Number (no) | IA-85,ICSS-86 |
| Page | pp.pp.19-24(IA), pp.19-24(ICSS), |
| #Pages | 6 |
| Date of Issue | 2022-06-16 (IA, ICSS) |