Presentation 2021-11-30
Comparison of CNN padding method using a part of malware API call sequence data with RNN.
Shugo Asai, Yuichi Futa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In recent years, as society has become more information-oriented, malware for information has been increasing. In such a situation, methods to automatically classify and analyze malware types are used and researched by learning the characteristics of malware using learning methods such as machine learning and deep learning from data obtained through dynamic analysis. However, since deep learning requires a huge amount of samples, the accuracy may be low when the samples are small. In this report, we suppose the condition that we have few samples and no alternative to utilizing of API calls through dynamic analysis for malware classification. Under the condition, we compare the classification accuracy of CNNs and RNNs to verify their effectiveness.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Malware / APIcall / CNN / RNN / LSTM / Classification / padding
Paper # ICSS2021-55
Date of Issue 2021-11-22 (ICSS)

Conference Information
Committee ICSS
Conference Date 2021/11/29(2days)
Place (in Japanese) (See Japanese page)
Place (in English) KOCHIJYO HALL
Topics (in Japanese) (See Japanese page)
Topics (in English) Security, etc.
Chair Katsunari Yoshioka(Yokohama National Univ.)
Vice Chair Kazunori Kamiya(NTT) / Takahiro Kasama(NICT)
Secretary Kazunori Kamiya(KDDI labs.) / Takahiro Kasama(Okayama Univ.)
Assistant Keisuke Kito(Mitsubishi Electric) / Takeshi Sugawara(Univ. of Electro-Comm.)

Paper Information
Registration To 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) Comparison of CNN padding method using a part of malware API call sequence data with RNN.
Sub Title (in English)
Keyword(1) Malware
Keyword(2) APIcall
Keyword(3) CNN
Keyword(4) RNN
Keyword(5) LSTM
Keyword(6) Classification
Keyword(7) padding
1st Author's Name Shugo Asai
1st Author's Affiliation Tokyo University of Technology(TUT)
2nd Author's Name Yuichi Futa
2nd Author's Affiliation Tokyo University of Technology(TUT)
Date 2021-11-30
Paper # ICSS2021-55
Volume (vol) vol.121
Number (no) ICSS-275
Page pp.pp.55-60(ICSS),
#Pages 6
Date of Issue 2021-11-22 (ICSS)