• Convergence Security Journal
• /
• v.23 no.5
• /
• pp.65-72
• /
• 2023

This paper proposes a method for detecting malicious domains considering human habitual characteristics by building a Deep Learning model based on LSTM (Long Short-Term Memory). DGA (Domain Generation Algorithm) malicious domains exploit human habitual errors, resulting in severe security threats. The objective is to swiftly and accurately respond to changes in malicious domains and their evasion techniques through typosquatting to minimize security threats. The LSTM-based Deep Learning model automatically analyzes and categorizes generated domains as malicious or benign based on malware-specific features. As a result of evaluating the model's performance based on ROC curve and AUC accuracy, it demonstrated 99.21% superior detection accuracy. Not only can this model detect malicious domains in real-time, but it also holds potential applications across various cyber security domains. This paper proposes and explores a novel approach aimed at safeguarding users and fostering a secure cyber environment against cyber attacks.

• Journal of Society of Preventive Korean Medicine
• /
• v.28 no.1
• /
• pp.31-42
• /
• 2024

Objective : In order to understand the changes in domestic approval regulations applicable to traditional Korean medical device companies, this article will explain the major amendments 「Regulation on Medical Device Approval Report Review, etc.」 from 2005 to the present on a year-by-year basis, and provide a counter plan to the recent changes in approval regulations. Methods : We analysed the changes in approval regulatory amendments related to the traditional Korean medical devices from 2005 to the present. Results : The Ministry of Food and Drug Safety is continuously improving medical device approval regulations to ensure the global competitiveness of domestic medical devices and contribute to the improvement of public health. Recent major approval regulatory amendments include the establishment of a review system for software medical devices and digital therapeutics, the recognition of real world evidence materials, the introduction of a biological evaluation of medical devices within a risk management process and a medical device approval licence renewal system. Conclusions : It is expected that the range of medical devices available to Korean medicine doctors will continue to expand in the future through the provision of non-face-to-face medical services and the development of advanced and new medical devices, as well as wearable medical devices and digital therapeutics. In order to increase the market entry potential of traditional Korean medical devices that incorporate advanced technologies such as digital technology and AI-based diagnosis and prediction technology, it is urgent that the government provide significant support to traditional Korean medical device companies to improve approval regulatory compliance.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.61-68
• /
• 2024

Despite the continuous efforts to mitigate pedestrian accidents at crosswalks, the problem persist. Vulnerable groups, including the elderly and disabled individuals are at a risk of being involved in traffic incidents. This paper proposes the implementation of object detection algorithm using the YOLO v5 model specifically for pedestrians using assistive devices like wheelchairs and crutches. For this research, data was collected and utilized through image crawling, Roboflow, and Mobility Aids datasets, which comprise of wheelchair users, crutch users, and pedestrians. Data augmentation techniques were applied to improve the model's generalization performance. Additionally, ensemble techniques were utilized to mitigate type 2 errors, resulting in 96% recall rate. This demonstrates that employing ensemble methods with a single YOLO model to target transportation-disadvantaged individuals can yield accurate detection performance without overlooking crucial objects.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.2
• /
• pp.76-90
• /
• 2024

Although artificial intelligence (AI) can be utilized in various domains such as smart city, healthcare, it is limited due to concerns about the exposure of personal and sensitive information. In response, the concept of distributed machine learning has emerged, wherein learning occurs locally before training a global model, mitigating the concentration of data on a central server. However, overall learning phase in a collaborative way among multiple participants poses threats to data privacy. In this paper, we systematically analyzes recent trends in privacy protection within the realm of distributed machine learning, considering factors such as the presence of a central server, distribution environment of the training datasets, and performance variations among participants. In particular, we focus on key distributed machine learning techniques, including horizontal federated learning, vertical federated learning, and swarm learning. We examine privacy protection mechanisms within these techniques and explores potential directions for future research.

• Journal of Korean Society of Archives and Records Management
• /
• v.24 no.2
• /
• pp.65-88
• /
• 2024

This pilot study explores a method of extracting metadata values and descriptions from records using named entity recognition (NER), a technique in natural language processing (NLP), a subfield of artificial intelligence. The study focuses on handwritten records from the Guro Industrial Complex, produced during the 1960s and 1970s, comprising approximately 1,200 pages and 80,000 words. After the preprocessing process of the records, which included digitization, the study employed a publicly available language API based on Google's Bidirectional Encoder Representations from Transformers (BERT) language model to recognize entity names within the text. As a result, 173 names of people and 314 of organizations and institutions were extracted from the Guro Industrial Complex's past records. These extracted entities are expected to serve as direct search terms for accessing the contents of the records. Furthermore, the study identified challenges that arose when applying the theoretical methodology of NLP to real-world records consisting of semistructured text. It also presents potential solutions and implications to consider when addressing these issues.

• Information Systems Review
• /
• v.24 no.3
• /
• pp.23-41
• /
• 2022

This case research investigates on how hearing aids, which is one of disability aids, is becoming a smart device, focusing on the case of Starkey Hearing Technologies(Starkey Inc.). Starkey, founded in 1967, has been a leader in innovating forms and functions of hearing aids, and has recently introduced the world's first hearing aid implemented with AI and biological sensors. In this study, history of disability aids, hearing aids(especially Starkey Inc.'s), smart wearable devices and smart earphones are compared. It has been found that recently, there has been a breakdown of boundaries between hearing aids and smart wearable devices in terms of their functions, since entertainment and life assistant functions are added to hearing aids. Based on this trend, the development model of disability aids and smart wearable devices are derived, and according social changes are discussed.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.3
• /
• pp.625-632
• /
• 2024

This paper analyzed the changes in the curriculum of computer and software-related majors that educate the core ICT technologies needed for the 4th Industrial Revolution, before and after the COVID-19 pandemic. According to the standard classification of university education units, 172 majors classified into Applied Software Engineering, Computer Science·Computer Engineering, and Artificial Intelligence Engineering were targeted, and the curricula of 2023 and 2019 were compared and analyzed. As a result of the analysis, the introduction of the related curriculum for each curriculum group increased by about 2.6%p before and after the COVID-19 pandemic (2023 84.2%, 2019 81.6%). and the 4th Industrial Revolution response index increased by 9.5 points (37.0 in 2023, 27.5 in 2019)

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.3
• /
• pp.515-522
• /
• 2024

Energy consumption is steadily increasing, and buildings in particular account for more than 20% of the total energy consumption around the world. As an effort to cost-effectively manage the energy consumption of buildings, many research groups have recently focused on Smart Building Energy Management Systems (BEMS), which are deepening the research depth by applying artificial intelligence(AI). In this paper, we propose a reinforcement learning-based energy control method for smart energy buildings integrated with V2G station, which aims to reduce the total energy cost of the building. The results of performance evaluation based on the energy consumption data measured in the real-world building shows that the proposed method can gradually reduce the total energy costs of the building as the learning process progresses.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.18 no.6
• /
• pp.191-199
• /
• 2023

The purpose of this study is to examine the impact of ChatGpt's quality on users' satisfaction and intention to continuous use it. For this purpose, a survey was conducted targeting college students in the Busan and Gyeongnam regions, and responses from a total of 155 people were verified using the SPSS 28.0 program. As a result of the study, reliability and stability among ChatGPT quality factors were found to have a positive effect on satisfaction with use and intention to continuous use. Satisfaction with the use of ChatGPT was found to have a positive effect on intention to continuous use.. Satisfaction with use was found to have a positive mediating effect between the reliability and stability of ChatGPT quality and intention to continous use it. As a result of this study, we aim to contribute to suggesting educational and policy directions necessary to promote the use of ChatGPT by presenting factors that affect users' intention to continuous use ChatGPT among the qualities of ChatGPT.

• Convergence Security Journal
• /
• v.23 no.4
• /
• pp.61-70
• /
• 2023

Clustering is an unsupervised learning method that involves grouping data based on features such as distance metrics, using data without known labels or ground truth values. This method has the advantage of being applicable to various types of data, including images, text, and audio, without the need for labeling. Traditional clustering techniques involve applying dimensionality reduction methods or extracting specific features to perform clustering. However, with the advancement of deep learning models, research on deep clustering techniques using techniques such as autoencoders and generative adversarial networks, which represent input data as latent vectors, has emerged. In this study, we propose a deep clustering technique based on deep learning. In this approach, we use an autoencoder to transform the input data into latent vectors, and then construct a vector space according to the cluster structure and perform k-means clustering. We conducted experiments using the MNIST and Fashion-MNIST datasets in the PyTorch machine learning library as the experimental environment. The model used is a convolutional neural network-based autoencoder model. The experimental results show an accuracy of 89.42% for MNIST and 56.64% for Fashion-MNIST when k is set to 10.