• Journal of Korea Multimedia Society
• /
• v.25 no.10
• /
• pp.1433-1447
• /
• 2022

Recently, as the demand for physical security technology to prevent leakage of technical and business information of companies and public institutions increases, the high tech companies are operating X-ray security checkpoints at building entrances to protect their intellectual property and technology. X-ray security checkpoints are operated to detect cameras and storage media that may store or leak important technologies in the bags of people entering and leaving the building. In this study, we propose an X-ray security checkpoint system that automatically detects a storage medium in an X-ray image using a deep learning based object detection method. The proposed system consists of an edge computing unit and a cloud-computing unit. We employ the RetinaNet for automatic storage media detection in the X-ray security checkpoint images. The proposed approach achieved mAP of 95.92% on private dataset.

• International Journal of Computer Science & Network Security
• /
• v.22 no.9
• /
• pp.358-368
• /
• 2022

Social media is a window for everyone, individuals, communities, and companies to spread ideas and promote trends and products. With these opportunities, challenges and problems related to security, privacy and rights arose. Also, the data accumulated from social media has become a fertile source for many analytics, inference, and experimentation with new technologies in the field of data science. In this chapter, emphasis will be given to methods of trend analysis, especially ensemble learning methods. Ensemble learning methods embrace the concept of cooperation between different learning methods rather than competition between them. Therefore, in this chapter, we will discuss the most important trends in ensemble learning and their applications in analysing social media data and anticipating the most important future trends.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.5
• /
• pp.1055-1062
• /
• 2021

As face-to-face education becomes difficult due to the spread of COVID-19, the use of e-learning content and virtual training is increasing. In the case of information security education, practice to learn response techniques is important, so simulation hacking and vulnerability analysis activities have been supported as virtual training for a long time. In order to increase the educational effect, contents should be designed similar to real situation, and learning activities to achieve the learning goals should be designed. In addition, excellent functions and scalability of the system supporting learning activities are required. The researcher developed an LMS evaluation index that supports non-face-to-face education by considering the key elements of non-face-to-face education and training. The developed evaluation index was applied to the information security education platform to verify its practical utility.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.702-723
• /
• 2020

The application of machine learning (ML) in intrusion detection has attracted much attention with the rapid growth of information security threat. As an efficient multi-label classifier, kernel extreme learning machine (KELM) has been gradually used in intrusion detection system. However, the performance of KELM heavily relies on the kernel selection. In this paper, a novel multiple kernel extreme learning machine (MKELM) model combining the ReliefF with nature-inspired methods is proposed for intrusion detection. The MKELM is designed to estimate whether the attack is carried out and the ReliefF is used as a preprocessor of MKELM to select appropriate features. In addition, the nature-inspired methods whose fitness functions are defined based on the kernel alignment are employed to build the optimal composite kernel in the MKELM. The KDD99, NSL and Kyoto datasets are used to evaluate the performance of the model. The experimental results indicate that the optimal composite kernel function can be determined by using any heuristic optimization method, including PSO, GA, GWO, BA and DE. Since the filter-based feature selection method is combined with the multiple kernel learning approach independent of the classifier, the proposed model can have a good performance while saving a lot of training time.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.18 no.6A
• /
• pp.115-127
• /
• 2008

The Native API is a system call which can only be accessed with the authentication of the administrator. It can be used to detect a variety of malicious codes which can only be executed with the administrator's authority. Therefore, much research is being done on detection methods using the characteristics of the Native API. Most of these researches are being done by using supervised learning methods of machine learning. However, the classification standards of Anti-Virus companies do not reflect the characteristics of the Native API. As a result the population data used in the supervised learning methods are not accurate. Therefore, more research is needed on the topic of classification standards using the Native API for detection. This paper proposes a method for re-grouping malicious codes using fuzzy clustering methods with the Native API standard. The accuracy of the proposed re-grouping method uses machine learning to compare detection rates with previous classifying methods for evaluation.

• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.297-302
• /
• 2021

The development of various phishing websites enables hackers to access confidential personal or financial data, thus, decreasing the trust in e-business. This paper compared the detection techniques utilizing URL-based features. To analyze and compare the performance of supervised machine learning classifiers, the machine learning classifiers were trained by using more than 11,005 phishing and legitimate URLs. 30 features were extracted from the URLs to detect a phishing or legitimate URL. Logistic Regression, Random Forest, and CatBoost classifiers were then analyzed and their performances were evaluated. The results yielded that CatBoost was much better classifier than Random Forest and Logistic Regression with up to 96% of detection accuracy.

• KEPCO Journal on Electric Power and Energy
• /
• v.6 no.4
• /
• pp.399-404
• /
• 2020

There are many attempts to apply AI technology to diagnose facilities or improve the work efficiency of the power industry. The emergence of new machine learning technologies, such as deep learning, is accelerating the digital transformation of the power sector. The problem is that traditional power systems face security risks when adopting state-of-the-art AI systems. This adoption has convergence characteristics and reveals new cybersecurity threats and vulnerabilities to the power system. This paper deals with the security measures and implementations of the power system using machine learning. Through building a commercial facility operations forecasting system using machine learning technology utilizing power big data, this paper identifies and addresses security vulnerabilities that must compensated to protect customer information and power system safety. Furthermore, it provides security guidelines by generalizing security measures to be considered when applying AI.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.3
• /
• pp.357-367
• /
• 2020

According to Symantec's 2018 internet security threat report, The number of new mobile malware variants increased by 54 percent in 2017, as compared to 2016. And last year, there were an average of 24,000 malicious mobile applications blocked each day. Existing signature-based technologies of malware detection have limitations. So, malware detection technique through machine learning is being researched to detect malware variant. However, even in the case of applying machine learning, if the proper features of the malware are not properly selected, the machine learning cannot be shown correctly. We are focusing on feature selection method to find the features of malware variant in this research.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4625-4647
• /
• 2020

The Generative Adversarial Networks, as one of the most creative deep learning models in recent years, has achieved great success in computer vision and natural language processing. It uses the game theory to generate the best sample in generator and discriminator. Recently, many deep learning models have been applied to the security field. Along with the idea of "generative" and "adversarial", researchers are trying to apply Generative Adversarial Networks to the security field. This paper presents the development of Generative Adversarial Networks. We review traditional generation models and typical Generative Adversarial Networks models, analyze the application of their models in natural language processing and computer vision. To emphasize that Generative Adversarial Networks models are feasible to be used in security, we separately review the contributions that their defenses in information security, cyber security and artificial intelligence security. Finally, drawing on the reviewed literature, we provide a broader outlook of this research direction.

• Journal of Information Processing Systems
• /
• v.17 no.3
• /
• pp.571-585
• /
• 2021

Applications for object detection are expanding as it is automated through artificial intelligence-based processing, such as deep learning, on a large volume of images and videos. High dependence on training data and a non-transparent way to find answers are the common characteristics of deep learning. Attacks on training data and training models have emerged, which are closely related to the nature of deep learning. Privacy, integrity, and robustness for the extracted information are important security issues because deep learning enables object recognition in images and videos. This paper summarizes the security issues that need to be addressed for future applications and analyzes the state-of-the-art security studies related to robustness, privacy, and integrity of object detection for images and videos.