• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.6
• /
• pp.1081-1090
• /
• 2022

The face recognition model is used for identity recognition of smartphones, providing convenience to many users. As a result, the security review of the DNN model is becoming important, with adversarial attacks present as a well-known vulnerability of the DNN model. Adversarial attacks have evolved to decision-based attack techniques that use only the recognition results of deep learning models to perform attacks. However, existing decision-based attack technique[14] have a problem that requires a large number of queries when generating adversarial examples. In particular, it takes a large number of queries to approximate the gradient. Therefore, in this paper, we propose a method of generating adversarial examples using orthogonal space sampling and dimensionality reduction sampling to avoid wasting queries that are consumed to approximate the gradient of existing decision-based attack technique[14]. Experiments show that our method can reduce the perturbation size of adversarial examples by about 2.4 compared to existing attack technique[14] and increase the attack success rate by 14% compared to existing attack technique[14]. Experimental results demonstrate that the adversarial example generation method proposed in this paper has superior attack performance.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.15 no.4
• /
• pp.217-222
• /
• 2022

Information plays an important role in modern society. Most of the information is processed and moved in the digital space. In cyberspace, confidential communication based on resistance and security is fundamental. It is essential to protect the information sent and received over the network. However, information may be leaked and forged by unauthorized users. The effectiveness of the existing protection system decreases as an innovative technique is applied to identify the communication contents by a third party. Steganography is a technique for inserting secret information into a specific area of a medium. Stegganography and steganalysis techniques are at odds with each other. A new and sophisticatedly implemented system is needed to cope with the advanced steganalysis. To enhance step-by-step diffusion and irregularity, I propose a hybrid implementation technique of image steganography for Hangul messages based on layered encryption and variable ShiftRows. PSNR was calculated to measure the proposed steganography efficiency and performance. Compared to the basic LSB technique, it was shown that the diffusion and randomness can be increased even though the PSNR decreased by 1.45%.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.2
• /
• pp.267-277
• /
• 2022

Due to the increasing proportion of cloud and remote working environments, various information security incidents are occurring. Insider threats have emerged as a major issue, with cases in which corporate insiders attempting to leak confidential data by accessing it remotely. In response, insider threat detection approaches based on machine learning have been developed. However, existing machine learning methods used to detect insider threats do not take biases and variances into account, which leads to limited performance. In this paper, boosting-type ensemble learning algorithms are applied to verify the performance of malicious insider detection, conduct a close analysis, and even consider the imbalance in datasets to determine the final result. Through experiments, we show that using ensemble learning achieves similar or higher accuracy to other existing malicious insider detection approaches while considering bias-variance tradeoff. The experimental results show that ensemble learning using bagging and boosting methods reached an accuracy of over 98%, which improves malicious insider detection performance by 5.62% compared to the average accuracy of single learning models used.

• Journal of Platform Technology
• /
• v.11 no.4
• /
• pp.3-12
• /
• 2023

In the post-COVID era, the importance of quarantine measures is greatly emphasized, and accordingly, research related to the detection of mask wearing conditions and prevention of other infectious diseases using deep learning is being conducted. However, research on the detection and tracking of visitors to cultural facilities to prevent the spread of diseases is equally important, so research on this should be conducted. In this paper, a convolutional neural network-based object detection model is trained through transfer learning using a pre-collected dataset. The weights of the trained detection model are then applied to a multi-object tracking model to monitor visitors. The visitor detection model demonstrates results with a precision of 96.3%, recall of 85.2%, and an F1-score of 90.4%. Quantitative results of the tracking model include a MOTA (Multiple Object Tracking Accuracy) of 65.6%, IDF1 (ID F1 Score) of 68.3%, and HOTA (Higher Order Tracking Accuracy) of 57.2%. Furthermore, a qualitative comparison with other multi-object tracking models showcased superior results for the model proposed in this paper. The research of this paper can be applied to the hygiene systems within cultural facilities in the post-COVID era.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.4
• /
• pp.687-697
• /
• 2023

Botnets, whose attack patterns are becoming more sophisticated and diverse, are recognized as one of the most serious cybersecurity threats today. This paper revisits the experimental results of botnet detection using autoencoder, a semi-supervised deep learning model, for UGR and CTU-13 data sets. To prepare the input vectors of autoencoder, we create data points by grouping the NetFlow records into sliding windows based on source IP address and aggregating them to form features. In particular, we discover a simple power-law; that is the number of data points that have some flow-degree is proportional to the number of NetFlow records aggregated in them. Moreover, we show that our power-law fits the real data very well resulting in correlation coefficients of 97% or higher. We also show that this power-law has an impact on the learning of autoencoder and, as a result, influences the performance of botnet detection. Furthermore, we evaluate the performance of autoencoder using the area under the Receiver Operating Characteristic (ROC) curve.

• Journal of Platform Technology
• /
• v.11 no.4
• /
• pp.35-49
• /
• 2023

Recently, real-time cyber threats are constantly occurring for various reasons. Most companies have the characteristic of digitizing important internal information and storing it centrally, so it can be said that the impact is very high when an Computer Security Incident occurs. All electronic device information collected and analyzed in the process of responding to an Computer Security Incident has the characteristic of being subject to change at any time. Submission of related evidence is required in future investigations and courts. At this time, the basic principles of digital forensics, such as the principle of integrity and the principle of chain of custody, must be followed to ensure legitimacy and accuracy of the evidence. In this paper, we propose a digital forensic-based Computer Security Incident Inspection and Response procedure in the Windows 10 environment to secure the legitimacy and accuracy of digital evidence collected and analyzed when an intrusion occurs, prevent intrusion in advance, and quickly recognize it.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.5
• /
• pp.65-75
• /
• 2009

The development of IT technology, Internet popularity is increasing geometrically. However, as its side effect, the intrusion behaviors such as information leakage for key system and infringement of computation network etc are also increasing fast. The attack traffic detection method which is suggested in this study utilizes the Snort, traditional NIDS, filters the packet with false positive among the detected attack traffics using Nmap information. Then, it performs the secondary filtering using nessus vulnerability information and finally performs correlation analysis considering appropriateness of management system, severity of signature and security hole so that it could reduce false positive alarm message as well as minimize the errors from false positive and as a result, it raised the overall attack detection results.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.7
• /
• pp.179-187
• /
• 2023

Consumer trust plays a crucial role in the development of the sharing economy. This study primarily focuses on the factors influencing consumer trust and examines the case of ofo, a former leader in China's bike-sharing industry. This paper analyzes the decline in consumer trust in ofo, which can be attributed to internal management issues and the near-bankruptcy situation. The "difficulty in refunds" issue faced by ofo since December 2018 has been growing continuously, and this study explores the factors influencing trust and distrust in this context. By considering product factors (quality), platform factors (payment security, privacy protection, reputation), and social factors (social norms, government regulation) as independent variables, the study analyzes the factors affecting consumer trust. The analysis results revealed that as consumers' distrust towards shared bikes increases, their switching intention also increases. The company's reputation and social norms were found to influence both trust and distrust, while government regulation was found to influence trust. The research findings provide insights relevant to sharing economy platforms and offer guidance for future studies.

• Journal of Service Research and Studies
• /
• v.11 no.3
• /
• pp.55-74
• /
• 2021

In the global era, the tourism industry is a major profitable business, and the state and local governments are making various efforts to provide quality tourism experiences. The purpose of this study in a variety of tourism contents is to study barrier-free subway services in line with the global era, and to expand from the existing rapid, safe, and ordered transportation to the provision of high-quality comfort and all-round service experiences. This study compared and analyzed the subway service systems of Korea, China and both countries through the service design method, and presented a barrier-free subway service system to improve the user's satisfaction with the subway service system by improving the user's service experience. As a result, research results showed that 20 attractive quality attributes in 17 fields, such as convenience facilities, language issues, security equipment, and riding environment, play an important role in improving the quality and experience of subway services. In addition, through the construction of a Service Spatial Scenario, an optimized subway service system can be visualized to help understand this so that it can be used as a reference for creating a strategic application.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.1
• /
• pp.51-61
• /
• 2023

Messengers such as KakaoTalk, LINE, and Facebook Messenger are universal means of communication used by anyone. As the convenience functions provided to users and their usage time increase, so does the user behavior information remaining in the artifacts, which is being used as important evidence from the perspective of digital forensic investigation. However, for security reasons, most of the data is currently stored encrypted. In addition, cover-up behaviors such as intentional manipulation, concealment, and deletion are increasing, causing the problem of delaying digital forensic analysis time. In this paper, we conducted a study on the data decryption and artifacts analysis in a Windows environment for KakaoTalk, the messenger with the largest number of users in Korea. An efficient way of obtaining a decryption key and a method of identifying and decrypting messages attempted to be deleted are presented, and thumbnail artifacts are analyzed.