• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.286-289
• /
• 2020

정보통신 기술의 발달로 무분별한 사이버 공격에 노출되어 있기 때문에 정보보안의 기술이 중요해지고 있다. 이중 침입 탐지 시스템은 방화벽과 더불어 시스템 및 네트워크 보안을 위한 대표적인 수단으로, 현재까지 네트워크 기반인 NIDS와 호스트 기반인 HIDS에 대한 많은 연구가 이루어졌다. 이러한 침입 탐지에 대한 CTI(Cyber Threat Intelligence)를 공유하기 위해 다양한 CTI 프레임워크를 사용하여 CTI 정보를 공유하는 연구가 진행되고 있다. 이에 본 논문에서는 CVE기반의 OpenIOC와 Snort 및 OSSEC에서 생성된 Raw Data를 결합하여 새로운 CTI 프레임 워크를 제안한다. 제안된 시스템을 테스트하기 위해서는 CVE 분석을 기반으로한 Kali Linux로 공격을 진행한다, 이를 통해 생성된 데이터는 시간이 지남에 따라 축적된 데이터를 저장 및 검색을 위해 대규모 분산 처리 시스템과도 결합이 필요할 것으로 예상되며 추후 딥러닝 기술을 활용하면 지능형 지속 위협을 분석하는데 용이할 것으로 예상된다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.775-778
• /
• 2021

악성 댓글은 언어폭력이며 사이버 범죄의 일종으로 인터넷상에서 상대방이 올린 글에 비방이나 험담을 하는 악의적인 댓글을 말한다. 악성 댓글을 단순히 차단하는 다른 프로그램들과는 달리 해당 영상의 악성 댓글의 비율을 알려주고 악플러들의 닉네임과 그 빈도를 나타내주는 것으로 차별화를 두었다. 따라서 많은 유튜버들이 겪는 악성 댓글 문제들을 탐지하여 유튜브에 달리는 악성 댓글들을 탐지하고 시각화하여 제공한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.535-536
• /
• 2022

본 논문 원격교육환경에서 학습자가 남긴 개별 데이터에 대한 장기적 관리 및 효율적 학습자 관리를 위한 데이터 전처리 방법으로 전통적인 ARIMA를 수정하여 연구하였다. ARIMA는 과거시점 데이터에 대한 회귀식과 변화율을 현 시점 데이터에 반영하는 방식이며 본 연구에서는 ARIMA 처리과정에서 딥러닝 알고리즘인 RNN의 변형방법인 LSTM을 적용하여 부분 데이터셋의 전처리과정에 대한 정확성과 재현율을 높이도록 하였다. 본 연구의 결과 전통적인 ARIMA 적용시와 대비하여 7~9%의 성능향상을 확인하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.801-802
• /
• 2023

시큐어 코딩은 해킹 등 사이버 공격의 원인인 보안 취약점을 제거해 안전한 소프트웨어를 개발하는 SW 개발 기법을 의미한다. 개발자의 실수나 논리적 오류로 인해 발생할 수 있는 문제점을 사전에 차단하여 대응하고자 하는 것이다. 그러나 현재 시큐어 코딩에는 오탐과 미탐의 문제가 발생한다는 단점이 있다. 따라서 본 논문에서는 오탐과 미탐이 발생하는 단점을 해결하고자 머신러닝 알고리즘을 활용하여 AI 기반으로 개발자의 실수나 논리적 오류를 탐지하는 시큐어 코딩 도구를 만들고자 한다. 다양한 모델을 사용하여 보안 취약점을 모아놓은 Juliet Test Suite를 전처리하여 학습시켰고, 정확도를 높이기 위한 과정 중에 있다. 향후 연구를 통해 정확도를 높여 정확한 시큐어 코딩 점검 도구를 개발할 수 있을 것이다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.918-919
• /
• 2023

딥러닝 기술이 발전함에 따라 모델의 복잡성 역시 증가하고 있다. 본 연구에서는 모델 경량화를 위해 Knowledge Distillation 기법을 Crowd Counting Model에 적용했다. M-SFANet을 Teacher 모델로, 파라미터수가 적은 MCNN 모델을 Student 모델로 채택해 Knowledge Distillation을 적용한 결과, 기존의 MCNN 모델보다 성능을 향상했다. 이는 정확도와 메모리 효율성 측면에서 많은 개선을 이루어 컴퓨팅 리소스가 부족한 기기에서도 본 모델을 실행할 수 있어 많은 활용이 가능할 것이다.

• Journal of Digital Convergence
• /
• v.12 no.4
• /
• pp.21-32
• /
• 2014

Recently, the number of Koreans who use smartphones has increased drastically; many use smartphones to learn English. In this study, one hundred Korean adult ELT (English language teaching) learners were surveyed to investigate their use of smartphones and factors influencing such use. For comparison, sixty-two students of a Korean cyber university were surveyed; these students were able to study using their smartphones in a smart campus environment. The research results showed that both groups positively used smartphones frequently, and that many intended to continue using them. With regard to ELT, both groups intended to learn English using their smartphones. Furthermore, they preferred certain types of ELT content: thirty-minute or less learning sessions, receptive English skills that focused on listening and reading, and short units of framed language items such as pronunciation and vocabulary. However, few of the respondents in both groups installed ELT apps on their smartphones, and few of the ELT apps satisfied them. The cyber university students responded similarly about smartphone use, although their responses regarding smartphone use for ELT purposes were less positive. These results indicate that the goal of cyber universities in achieving optimum learning outcomes through smart learning and the smart campus has not yet been realized.

• Journal of Technology Innovation
• /
• v.12 no.3
• /
• pp.1-24
• /
• 2004

E-Learning is another way of teaching and learning. E-learning is a networked phenomenon allowing for instant revisions and distribution, and goes beyond training and instruction to the delivery of information and tools to improve performance. The benefits of e-learning are many, including cost-effectiveness, enhanced responsiveness to change, consistency, timely content, flexible accessibility, and providing customer value. The proponents of e-learning stress the importance of using communities of interest to support and enhance the learning process. They also emphasizes that people learn more effectively when they interact and are involved with other people participating in similar endeavors. Although the role of e-learning in higher education has significantly increased, the resistance to new technology by professors and lecturers in university and colleges worldwide remains high. The purpose of this study is to identify the determinants of attitude and planned behavior toward e-learning class in universities. A survey methodology was used to investigate a proposed model of influence, and structural equation modeling was used to analyze the results. The hypothesized model was largely supported by this analysis, and the overall results indicate that attitude toward e-learning systems is mostly influenced by the perceived ease of use as well as the level of perceived usefulness, where both factors are influenced by years of experiences in using cyber system and the technical support level. As in other TAM related research, it can be concluded that the perceived ease of use and perceived usefulness contribute to the future use of e-learning system.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.2
• /
• pp.201-211
• /
• 2022

According to the recent change in the cybersecurity paradigm, research on anomaly detection methods using machine learning and deep learning techniques, which are AI implementation technologies, is increasing. In this study, a comparative study on data preprocessing techniques that can improve the anomaly detection performance of a GRU (Gated Recurrent Unit) neural network-based intrusion detection model using NGIDS-DS (Next Generation IDS Dataset), an open dataset, was conducted. In addition, in order to solve the class imbalance problem according to the ratio of normal data and attack data, the detection performance according to the oversampling ratio was compared and analyzed using the oversampling technique applied with DCGAN (Deep Convolutional Generative Adversarial Networks). As a result of the experiment, the method preprocessed using the Doc2Vec algorithm for system call feature and process execution path feature showed good performance, and in the case of oversampling performance, when DCGAN was used, improved detection performance was shown.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.177-188
• /
• 2024

As technology evolves, Internet usage continues to grow, resulting in a geometric increase in network traffic and communication volumes. The network path selection process, which is one of the core elements of the Internet, is becoming more complex and advanced as a result, and it is important to effectively manage and analyze it, and there is a need for a representation and visualization method that can be intuitively understood. To this end, this study designs a framework that analyzes network data using BGP, a network path selection method, and applies it to the cyber common operating picture for situational awareness. After that, we analyze the visualization elements required to visualize the information and conduct an experiment to implement a simple visualization. Based on the data collected and preprocessed in the experiment, the visualization screens implemented help commanders or security personnel to effectively understand the network situation and take command and control.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.28 no.4
• /
• pp.847-857
• /
• 2018

As the number of malicious code increases steeply, cyber attack victims targeting corporations, public institutions, financial institutions, hospitals are also increasing. Accordingly, academia and security industry are conducting various researches on malicious code detection. In recent years, there have been a lot of researches using machine learning techniques including deep learning. In the case of research using Convolutional Neural Network, ResNet, etc. for classification of malicious code, it can be confirmed that the performance improvement is higher than the existing classification method. However, one of the characteristics of the target attack is that it is custom malicious code that makes it operate only for a specific company, so it is not a form spreading widely to a large number of users. Since there are not many malicious codes of this kind, it is difficult to apply the previously studied machine learning or deep learning techniques. In this paper, we propose a method to classify malicious codes when the amount of samples is insufficient such as targeting type malicious code. As a result of the study, we confirmed that the accuracy of 97% can be achieved even with a small amount of data by applying the Memory Augmented Neural Networks model.