RISS 검색 - 국내학술지논문

무료
기관 내 무료
유료

내보내기
내책장담기
한글로보기

정확도순

내림차순

내림차순

10개씩 출력

1
Compact feature hashing for machine learning based malware detection

Damin Moon,JaeKoo Lee,MyungKeun Yoon 한국통신학회 2022 ICT Express Vol.8 No.1
Machine learning can detect variant malware files that can evade signature-based detection. Feature hashing is used to convert features into a fixed-length vector. In this paper, we study the appropriate vector size for feature hashing for a large dataset of malware files. Through exhaustive experiments on more than 280,000 real malware and benign files, we find for the first time that the default vector size of current feature hashing practices is unnecessarily large. We experimentally explore the appropriate vector size, which not only reduces memory space by 70% but also increases the detection accuracy, compared with the state-of-the-art scheme.
2
Direct successive ionic layer adsorption and reaction (SILAR) synthesis of nickel and cobalt hydroxide composites for supercapacitor applications

Lee, Damin,Xia, Qi Xun,Mane, Rajaram S.,Yun, Je Moon,Kim, Kwang Ho ELSEVIER SCIENCE 2017 JOURNAL OF ALLOYS AND COMPOUNDS Vol.722 No.-
- 원문보기
Abstract Mesoporous nickel and cobalt hydroxide composites are directly grown onto 3D macro-porous Ni foam as a binder-free electrode for supercapacitors by using the successive ionic layer adsorption and reaction (SILAR) method. This method is the cheapest and simplest among several deposition processes for supercapacitor applications. An as-obtained porous NiCo(OH)2 electrode exhibits a remarkable specific capacity (1113.6 mAh g−1 at a current density of 3 A g−1) and excellent cycling stability (85.62% capacity retention after 5000 cycles). Furthermore, an asymmetric supercapacitor assembled with NiCo(OH)2 as a positive electrode and graphene as a negative electrode shows a high energy density of 20.07 W h kg−1 at a power density of 2302.73 W kg−1 and excellent cycling stability (76.46% retention after 5000 cycles). As a result, it shows that the NiCo(OH)2 fabricated by the SILAR method can be a promising electrode towards energy-storage devices with high energy and power densities. Highlights <UL> <LI> SILAR method is a process conducted under mild conditions for 30 s. </LI> <LI> NiCo(OH)2 composite by SILAR method is a 3D flower-like porous nanostructure. </LI> <LI> The synthesized NiCo(OH)2 electrode shows high electrochemical performances. </LI> <LI> The specific capacity of the NiCo(OH)2 electrode is 1113.6 mAh g−1 at 3 A g−1. </LI> </UL> Graphical abstract [DISPLAY OMISSION]
3
머신러닝 기반 보안데이터 분석 연구

이식(Lee, Seek),김동훈(Kim, DongHoon),조영훈(Cho, YoungHun),명준우(Myung, JoonWoo),문다민(Moon, DaMin),이재구(Lee, JaeKoo),윤명근(Yoon, MyungKeun) 한국정보보호학회 2019 情報保護學會誌 Vol.29 No.3
- 원문보기 2
  ScienceON
  
  DBpia
최근 머신러닝 기술이 비약적으로 발전하고 있다. 하드웨어 성능이 향상되고 머신러닝 활용 도구가 오픈소스로 사용 편리하게 개발되어 대중화됨으로써 보안데이터 분석 분야에서도 머신러닝을 이용한 기술 개발이 활발히 진행되고 있다. 본 논문에서는 보안 분야의 악성코드 데이터와 보안관제 로그 데이터를 주요 대상으로 머신러닝 기술을 적용할 때 고려되어야 할 기술적 사항들과 최신 연구 동향, 데이터 셋 특징, 그리고 머신러닝 기반의 보안데이터 분석 기술의 기대 효과 및 현재 기술의 한계점 등을 다루도록 한다.