| Presentation | 2020-07-17 Study on which data we should label in a few-shot learning for service identification over encrypted web services Shouta Yoshida, Yutaka Eguchi, Kohei Shiomoto, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | It is very important to monitor and control the communication traffic to cope with theincreasing communication traffic.However, due to the recent standardization of encryption, it is not possible to distinguish the type of traffic.Therefore, in this paper, we use the observable packet data as a feature for encrypted communications. Using the supervised learning method, Few-shot Learning, to describe the types of web services We propose a method for classification.The unlabeled dataset is dimensionally reduced by t-SNE and then k-means method to label the characteristic data near the center of gravity of the cluster. The method improves classification accuracy in a small number of data sets by 11%. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Traffic classification / Machine learning / Deep learning / Few-shot Learning |
| Paper # | ICM2020-14 |
| Date of Issue | 2020-07-09 (ICM) |
| Conference Information | |
| Committee | ICM |
|---|---|
| Conference Date | 2020/7/16(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Virtual Conference |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Kazuhiko Kinoshita(Tokushima Univ.) |
| Vice Chair | Yoichi Sato(Open Systems Laboratory) / Haruo Ooishi(NTT) |
| Secretary | Yoichi Sato(NTT) / Haruo Ooishi(Bosco) |
| Assistant | Tetsuya Uchiumi(Fujitsu Lab.) |
| Paper Information | |
| Registration To | Technical Committee on Information and Communication Management |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Study on which data we should label in a few-shot learning for service identification over encrypted web services |
| Sub Title (in English) | |
| Keyword(1) | Traffic classification |
| Keyword(2) | Machine learning |
| Keyword(3) | Deep learning |
| Keyword(4) | Few-shot Learning |
| 1st Author's Name | Shouta Yoshida |
| 1st Author's Affiliation | Tokyo City University(TCU) |
| 2nd Author's Name | Yutaka Eguchi |
| 2nd Author's Affiliation | Tokyo City University(TCU) |
| 3rd Author's Name | Kohei Shiomoto |
| 3rd Author's Affiliation | Tokyo City University(TCU) |
| Date | 2020-07-17 |
| Paper # | ICM2020-14 |
| Volume (vol) | vol.120 |
| Number (no) | ICM-109 |
| Page | pp.pp.37-42(ICM), |
| #Pages | 6 |
| Date of Issue | 2020-07-09 (ICM) |