ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술지 Design of Network Threat Detection and Classification based on Machine Learning on Cloud Computing
Cited 20 time in scopus Download 12 time Share share facebook twitter linkedin kakaostory
저자
김현주, 김종현, 김영수, 김익균, 김귀남
발행일
201901
출처
Cluster Computing, v.22 no.Supp.1, pp.2341-2350
ISSN
1386-7857
출판사
Springer
DOI
https://dx.doi.org/10.1007/s10586-018-1841-8
협약과제
18HH1400, 맞춤형 보안서비스 제공을 위한 클라우드 기반 지능형 보안 기술 개발, 김종현
초록
To respond to recent network threats that are using increasingly intelligent techniques, the intelligent security technology on cloud computing is required. Especially it supports small and medium enterprises to build IT security solution with low cost and less effort because it is provided as Security as a Service on a cloud environment. In this paper, we particularly propose the network threat detection and classification method based on machine learning, which is a part of the intelligent threat analysis technology. In order to improve the performance of detection and classification of network threat, it was built in a hybrid way such as applying an unsupervised learning approach with unlabeled data, naming clusters with labeled data, and using a supervised learning approach for feature selection.
키워드
Cloud computing, Machine learning, Network threat detection and classification, Security as a Service, Virtual security function
KSP 제안 키워드
Classification method, Cloud Computing, Feature selection(FS), Intelligent security, Intelligent techniques, It security, Learning approach, Low-cost, Network Threat, Security as a Service, Security function