• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.05a
• /
• pp.284-287
• /
• 2013

Recently target oriented attacks which target an enterprise and a government agency are increasing. The starting point of APT(Advanced Persistent Threat), called as target oriented attacks, is the spear phishing email that is personalized based on the information collected via Internet of the target personnel. It is known that 94% of target oriented attacks use spear phishing emails. Therefore, in this paper, we analysed spear phishing methods in detail and the characteristics and recent trends of spear phishing threats and proposed the effective prevention method of spear phishing for the end user.

• Korean Journal of School Psychology
• /
• v.17 no.2
• /
• pp.223-241
• /
• 2020

The purpose of this study was to explore the latent profiles of social achievement goals and to investigate the differences in peer status (perceived popularity, social preference) and aggression (overt, relational, cyber) among those profile groups. Social achievement goals and cyber aggression data was acquired through self-reporting, and perceived popularity, social preference, and overt and relational aggression were assessed through peer nomination. Applying the latent profile analysis(LPA) to 1,239 elementary school students, three distinct groups of social achievement goals were identified: a development-oriented achievement goal group, an average social goal group, and a overall-high social achievement goal group. Using logistic regression analysis, the relationships between the latent group, peer status, and aggression were examined. The result indicated that the higher the social preference, the lower the probability of belonging to the 'overall-high social achievement goal group'. And the higher the cyber aggression, the lower the probability of belonging to the 'development-oriented achievement goal group'. In addition, the higher the relational aggression of the second time, the higher the probability of belonging to the 'overall-high social achievement goal group' as compared to the 'average social goal group'.

• Convergence Security Journal
• /
• v.5 no.3
• /
• pp.67-73
• /
• 2005

Security requirements must be defined well to reduce software vulnerabilities in requirement specification phase. In this paper, we show how to specify security requirements in structured manner for object-oriented development methodology. Our method specifies security requirements through four phases: defining security objectives, identifying the threat, construct attack tree, and specifying security function. This method would help developers to specify security requirements and functions which software have to possess clearly and systematically.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.5
• /
• pp.871-883
• /
• 2014

New types of attacks that mainly compromise the public, portal and financial websites for the purpose of economic profit or national confusion are being emerged and evolved. In addition, in case of 'drive by download' attack, if a host just visits the compromised websites, then the host is infected by a malware. Website falsification detection system is one of the most powerful solutions to cope with such cyber threats that try to attack the websites. Many domestic CERTs including NCSC (National Cyber Security Center) that carry out security monitoring and response service deploy it into the target organizations. However, the existing techniques for the website falsification detection system have practical problems in that their time complexity is high and the detection accuracy is not high. In this paper, we propose website falsification detection system based on image and code analysis for improving the performance of the security monitoring and response service in CERTs. The proposed system focuses on improvement of the accuracy as well as the rapidity in detecting falsification of the target websites.

• Journal of Technology Innovation
• /
• v.20 no.3
• /
• pp.93-128
• /
• 2012

Defense technology planning identifies medium-and long-term core technologies to accomplish future defense goals and suggests strategies for future R&D. In order to promote the export-oriented defense industry, planning paradigms should be shifted from technology-oriented planning that focuses on weapon systems to market- oriented R&D planning. This study aims to strategically support 'preliminary technology planning' the Defense Agency for Technology and Quality is pushing ahead with. Through market-orientation analysis models of weapon systems based on defense R&D planning, data research on previous market-oriented research, and the analyses and examples of global defence markets, it evaluates market attractiveness to UAVs and drew methods for exploring markets and enhancing competitiveness of military equipment. The market-oriented analysis model of weapon systems is considered to be a helpful reference as a relevant factor for decision making on establishing and verifying requirement planning. In particular, if a market-oriented defense R&D planning process is established institutionally, it will enable us to make export strategies tailored to different equipment from the planning phase and to support marketing strategically.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06b
• /
• pp.238-240
• /
• 2006

네트워크상에서 발생하는 다양한 형태의 대량의 데이터를 정확하고 효율적으로 분석하기 위해 설계되고 있는 마이닝 시스템들은 목표지향적으로 훈련데이터들을 어떻게 구축하여 다룰 것인지에 대한 문제보다는 대부분 얼마나 많은 데이터 마이닝 기법을 지원하고 이를 적용할 수 있는지 등의 기법에 초점을 두고 있다. 따라서, 점점 더 에이전트화, 분산화, 자동화 및 은닉화 되는 최근의 보안공격기법을 정확하게 탐지하기 위한 방법은 미흡한 실정이다. 본 연구에서는 유비쿼터스 환경 내에서 발생 가능한 문제 중 복잡하고 지능화된 침입패턴의 탐지를 위해 데이터 마이닝 기법과 결함허용방법을 이용하는 개선된 학습알고리즘과 후처리 방법에 의한 RTPID(Refinement Training and Post-processing for Intrusion Detection)시스템을 제안한다. 본 논문에서의 RTPID 시스템은 active learning과 post-processing을 이용하여, 네트워크 내에서 발생 가능한 침입형태들을 정확하고 효율적으로 다루어 분석하고 있다. 이는 기법에만 초점을 맞춘 기존의 데이터마이닝 분석을 개선하고 있으며, 특히 제안된 분석 프로세스를 진행하는 동안 능동학습방법의 장점을 수용하여 학습효과는 높이며 비용을 감소시킬 수 있는 자가학습방법(self learning)방법의 효과를 기대할 수 있다. 이는 관리자의 개입을 최소화하는 학습방법이면서 동시에 False Positive와 False Negative 의 오류를 매우 효율적으로 개선하는 방법으로 기대된다. 본 논문의 제안방법은 분석도구나 시스템에 의존하지 않기 때문에, 유사한 문제를 안고 있는 여러 분야의 네트웍 환경에 적용될 수 있다.더욱 높은성능을 가짐을 알 수 있다.의 각 노드의 전력이 위험할 때 에러 패킷을 발생하는 기법을 추가하였다. NS-2 시뮬레이터를 이용하여 실험을 한 결과, 제안한 기법이 AOMDV에 비해 경로 탐색 횟수가 최대 36.57% 까지 감소되었음을 알 수 있었다.의 작용보다 더 강력함을 시사하고 있다.TEX>로 최고값을 나타내었으며 그 후 감소하여 담금 10일에는 $1.61{\sim}2.34%$였다. 시험구간에는 KKR, SKR이 비교적 높은 값을 나타내었다. 무기질 함량은 발효기간이 경과할수록 증하였고 Ca는 $2.95{\sim}36.76$, Cu는 $0.01{\sim}0.14$, Fe는 $0.71{\sim}3.23$, K는 $110.89{\sim}517.33$, Mg는 $34.78{\sim}122.40$, Mn은 $0.56{\sim}5.98$, Na는 $0.19{\sim}14.36$, Zn은 $0.90{\sim}5.71ppm$을 나타내었으며, 시험구별로 보면 WNR, BNR구가 Na만 제외한 다른 무기성분 함량이 가장 높았다.O to reduce I/O cost by reusing data already present in the memory of other nodes. Finally, chunking and on-line compression mechanisms are included in both models. We demonstrate that we can obtain significantly high-performanc