Journal of The Korea Internet of Things Society
- P-ISSN : 2799-4791
- Publisher : The Korea Internet of Things Society
18papers in this issue.
This study discusses the necessity of standardizing patient data using blockchain technology and explores its application methods. Existing medical data management systems have limitations in ensuring patient safety and quality of care due to inefficiencies in information sharing and data distribution across multiple medical institutions. Blockchain, as a decentralized ledger technology, guarantees data integrity and security while granting patients the authority to manage their own medical information. This study proposes an approach that integrates medical information systems with blockchain technology to enhance data interoperability, ensuring secure data sharing and efficient management. By doing so, it aims to improve the quality of healthcare services and explores the potential for applications in various medical fields, including patient history management, medication tracking, and telemedicine.
This paper proposes a collision avoidance algorithm using point clouds in a virtual environment created in Unity. First, random rays are fired from the front of a camera to acquire a point cloud, which is then used to extract the floor and other regions. A direction vector towards the target point is generated from the extracted floor, and based on this vector, a path to the farthest point is constructed using the A<sup>*</sup> algorithm. Next, each waypoint along the path is filtered by calculating the distance to nearby obstacles, removing waypoints that are too close. The agent follows the filtered waypoints while emitting rays at set angles to the left and right to detect potential obstacles. If an obstacle is detected, the agent rotates in the opposite direction and continues along the waypoint path to avoid the obstacle and reach the target. This algorithm presents an efficient method for combining obstacle avoidance and path optimization in real-time environments, with strong applicability in autonomous driving and robotic navigation systems.
In response to the recent government's AI and semiconductor talent training policy, this study proposes a method of effectively classifying semiconductor chips and detecting defects in RGBA color space using AI deep learning technology. Quality assurance and defect detection of semiconductor chips are essential to ensure the reliability and performance of electronic devices. However, traditional inspection methods mainly include visual inspection, mechanical measurement, and electrical testing, which are time-consuming, expensive for state-of-the-art equipment, and inefficient for many production environments due to inspection. To solve this problem, image analysis techniques based on deep learning are attracting attention in automated inspection systems. Through this experiment, it was confirmed that the deep learning model using RGBA color space shows excellent performance in defect detection and classification of semiconductor chips. In particular, RGBA color space including alpha channel provides more accurate and precise results for defect detection than conventional RGB color space models with less learning. The results of this experiment suggest that the RGBA color space can play an important role in the deep learning-based defect detection system, and further experiments in various datasets and conditions will expand the scope of the method's use in the future. Such a model is highly likely to contribute to the automation and quality improvement of the semiconductor manufacturing process. This study aims to improve the accuracy and efficiency of the semiconductor chip inspection process by utilizing the advantages of RGBA color space.
The purpose of this study is to clarify the relationship between social support and performance in practical dance. The purpose is to clarify the mediating effects of mindset and grit in this relationship. Through this study, we aim to provide basic data that can contribute not only to the growth of practical dance, but also to the development of the industry and the enhancement of social value by utilizing related digital technology. The research subjects consisted of practical dance majors who had experience participating in competitions and battles. The research method was an online survey, and the survey was conducted from June 18 to July 17, 2024. The research results are as follows. First, the factors of social support affecting practical dance performance were confirmed as parent support, friend support, and leader support. Second, in the influence of social support and performance, fixed mindset was confirmed to have a mediating effect in friend support and leader support. Third, in the influence of social support and performance, perseverance grit was confirmed to have a mediating effect in leader support, and in the influence of social support and performance, passion grit was confirmed to have a mediating effect in parent support, friend support, and leader support. Fourth, in the influence of mindset and performance, perseverance grit was confirmed to have a mediating effect in fixed mindset and growth mindset, and in the influence of mindset and performance, passion grit was confirmed to have a mediating effect in fixed mindset and growth mindset. Based on the research results above, we propose ways to revitalize the practical dance industry by expanding the base of practical dance competitions and battles, utilizing the Internet of Things, and encouraging the participation of the MZ generation.
This study was conducted to explore the direction of artificial intelligence education for special education students through the educational experiences of digital sprout camp instructors and teaching assistants. To achieve the purpose of the study, the FGI technique was used. To this end, three areas and six questions were organized through literature analysis. The FGI subjects were selected from those who had teaching experience in both integrated and special school classes and had taught more than 10 times. As a result, three instructors and four teaching assistants were selected. This study conducted interviews through FGI, and extracted 39 meanings in three areas through the teaching experiences of the interview subjects. 19 meanings were extracted for integrated classes, 21 meanings for special schools, 19 meanings for instructors, and 20 meanings for teaching assistants. The educational directions extracted through the results of the study are as follows. Integrated classes are sufficient time and human resources support to operate education, sufficient infrastructure support, and customized education development that reflects the characteristics of students. Next, special schools should design and develop individualized classes that take into account the characteristics of students, provide sufficient infrastructure support, and provide prior training for teachers.
This study analyzed the effectiveness of lessons using metaverses and generative AI for pre-religious teachers through a qualitative case study. Eight third-year students in the Department of Christian Education at A University in Gyeonggi-do, South Korea, designed and conducted a project-based learning (PBL)-based class on 'e-Bible study material design' using metaverses and generative AI. The students analyzed the church school site, identified practical problems, and developed Bible study materials using generative AI to solve the problems. After storyboarding the Bible, they used generative AI to generate pictures, voices, and background music and implemented them into a digital Bible picture storybook and metaverse environment. To analyze the results of the study, we conducted an in-depth analysis of learners' reflection journals and interviews, and found that learning with generative AI and metaverse environments not only increased interest and immersion, but also showed positive changes in their ability to utilize generative AI and metaverse and their awareness of the need for AI convergence education. In addition, we confirmed the possibility of extending the learning experience beyond the classroom into everyday life through self-directed learning and collaborative problem-solving processes. However, the generalizability of the results is limited due to the limited time and number of participants in this study, but it suggests the possibility of education using generative AI and metaverses as an alternative education method in the era of great transformation.
In this paper, AI models for predicting peak power usage were developed and comparatively analyzed using data collected from the Jeju Samdasoo factory through a big data collection system based on IoT sensing technology. The LSTM (Long Short-Term Memory) model demonstrated the highest prediction accuracy for univariate time-series data, achieving an R<sup>2</sup> of 0.98, RMSE of 0.039, and MAE of 0.026. Meanwhile, the XGBoost (eXtreme Gradient Boosting) model effectively handled multivariate data, achieving an R<sup>2</sup> of 0.93, RMSE of 0.018, and MAE of 0.013. Various data preprocessing methods and feature combinations were experimentally applied to optimize model performance, highlighting the significant impact of preprocessing and variable selection on prediction accuracy. The findings suggest that optimized AI models for peak power prediction can reduce power costs and achieve approximately 10-15% reductions in carbon emissions. This study offers companies pursuing ESG (environmental, social, and governance) management practical and specific strategies for achieving sustainability, while demonstrating the applicability of the predictive model across various industries, including manufacturing, logistics, and smart factories.
This study was conducted to explore the direction of AI education for secondary special education students through the educational experiences of Digital Sprout Camp instructors and teaching assistants. The research method used the FGI technique. Through literature analysis, three areas and six questions were composed, and the FGI subjects were selected from those who had teaching experience in both integrated and special school classes and had taught more than 10 times. As a result, three instructors and four teaching assistants were selected. This study conducted interviews through FGI, and extracted 43 meanings in three areas through the teaching experiences of the interview subjects. 19 meanings were extracted from integrated classes and 24 meanings from special schools, and 24 meanings were extracted from instructors and 19 meanings from teaching assistants. The educational directions extracted through the research results are as follows. Integrated classes are learner typology and customized class development, and infrastructure support. Next, special schools are individualized classes according to the characteristics of students, device support and personnel support, and provision of pre-training for teachers.
This study analyzed domestic media reports to determine trends in how public facilities such as childcare centers and Elderly Care Facilities perceive major disasters. To this end, press articles from approximately four years from 2021 to October 2024 were analyzed using BIGKinds. As a result, first, in the keyword trend, there were 209 articles on 'childcare center +serious accident' and 23 articles on 'elderly care + serious accident'. Second, as a result of the relationship analysis, the main keywords for childcare centers were business owners and managers, workers (workers), civil servants, and local government heads, while the main keywords for elderly care facilities were business owners and managers, Gyeongsangbuk-do Governor (Governor), and nursing assistants. Third, as a result of the analysis of related words, the main keywords for childcare centers are public facilities, safety inspections, workplaces, district heads, business owners, management managers, on-site inspections, small business owners, local governments, and workers, while the main keywords for elderly care facilities are public facility managers, daycare centers, social welfare facilities, Gyeonggi-do Suwon-si, countermeasures, Gyeongbuk-do, Jincheon-gun. Based on the research results, suggestions were made to raise awareness of and prevent major disasters by establishing a safety and health management system, organizing a local government organization dedicated to major disasters and supporting a budget for major disaster prevention and response, implementing safety and health education within facilities and establishing procedures for hearing opinions from workers, and developing and distributing a manual for major disaster prevention education and response.
In an IoT environment, various objects operate in an organic relationship and provide a variety of services. This paper proposes a method to utilize shared resources instead of limited resources on a personal mobile device to provide user-requested services in an IoT environment. In order to recommend customized resources among various resources, the paper infers based on the user's sensitive information or usage history on the personal mobile terminal. The public resource management terminal identifies suitable resources, including new and alternative resources, based on environmental information and public information.
This study aims to identify the factors influencing the intention to switch to ChatGPT, which has emerged as a powerful alternative to traditional learning tools and internet searches, using the Push-Pull-Mooring (PPM) framework to establish a research model. The push factors include the limitations of traditional lecture environments and the lack of immediate feedback, while the pull factors are ease of access and the provision of personalized learning experiences. Social influence was selected as a mooring factor and set as a moderating variable to examine its effect on the intention to switch. A survey was conducted among university students in Chungcheongnam-do, and the data were analyzed using SPSS 27 and SmartPLS 4.0. The results revealed that push factors (limitations of traditional lecture environments), pull factors (ease of access and personalized learning experiences), and mooring factors (social influence) significantly affect the intention to switch. Notably, the moderating effect of social influence was significant only for ease of access, reflecting the decision-making characteristics of Generation Z university students. This study is expected to provide important insights into the utilization and development of innovative learning tools and the formulation of educational strategies through the analysis of ChatGPT switching intentions.
This study aimed to empirically verify the relationship between the parenting attitudes of parents of immigrant children and adolescents and their Korean culture Acceptance Attitude(KCAA) for living in Korean society. Additionally, the study sought to demonstrate the mediating effect of self-esteem on this relationship. A survey was conducted with 126 immigrant children and adolescents from 16 districts in Busan Metropolitan City. The survey was conducted online using Google Forms. Due to the difficulty in securing respondents among immigrant children and adolescents, a combination of snowball and random sampling methods was used. The main findings of this study are as follows. First, parenting attitudes had a positive impact on the self-esteem of immigrant children and adolescents. This suggests that when parents exhibit an authoritative style with high demands and responsiveness, children and adolescents are more likely to have higher self-esteem. Second, self-esteem had a positive impact on the KCAA. This finding confirms that self-esteem, which is the emotional feeling of how valuable one is, is crucial for the KCAA, which is necessary for adapting to Korean society. Third, self-esteem had a complete mediating effect on the relationship between parenting attitudes and the acceptance of Korean culture. This result demonstrates that without self-esteem, this model cannot be established in the relationship between parenting attitudes and the acceptance of Korean culture. Based on these findings, this study suggests the activation of various parenting support programs and the integrated implementation of diverse interventions necessary for the growth process of children and adolescents.
Quizzes in the classroom can enhance students' academic achievement by assessing their understanding of the material. However, instructors often hesitate to actively utilize quizzes because of the additional workload required for creating, grading, and analyzing quiz results. With the recent advancements in Large Language Model (LLM) technology and the widespread use of online survey platforms such as Google Forms, a more efficient environment for leveraging quizzes in classes is now available. This paper proposes a system that utilizes ChatGPT, a representative LLM, to automatically generate quiz questions, conduct quiz exams via Google Forms, and provide grading and feedback. Specifically, the paper discusses the system's requirements and design, along with a prototype implementation. The prototype system was developed in a Python environment, employing the ChatGPT API for quiz question generation and the Google Forms API for administering and grading quizzes. Additionally, the system enables instructors to review and select questions generated by ChatGPT, while quiz questions and students' responses are stored in a database to facilitate both collective and individual learning outcome analysis. The automated quiz generation system offers the advantage of automating the processes of quiz question creation, examination, evaluation, and analysis, significantly alleviating instructors' workload and improving students' academic performance.
The core of blockchain technology lies in the fact that instead of a centralized approach, individual nodes autonomously create blocks, but go through a consensus process so that eventually, all nodes have the same blockchain ledger. The technology used in this process is the consensus algorithm. Additionally, the consensus algorithm also helps solve the issue of double spending. To address this, the consensus algorithms used, such as Proof of Work (PoW), Proof of Stake (PoS), and Delegated Proof of Stake (DPoS), have been examined. Therefore, this paper aims to examine the types of double-spending attacks in blockchain and propose a Delegated Proof of Stake (DPoS) consensus algorithm incorporating Counting Bloom Filters (CBF) to prevent double-spending attacks and collusion among delegates.
In the past, making a right turn at intersections relied entirely on the driver's judgment. The Act on the Passage of Intersections was revised in 2023 to require drivers to yield or pause when there are pedestrians crossing or about to cross the crosswalk. Despite the revision, many drivers are unaware of the new law and continue to make right turns as before. Moreover, some drivers have a complacent attitude, thinking "nothing will happen anyway" when they follow the old method. Even drivers who know the correct way to pass often make sudden stops due to pedestrians being in blind spots. This creates dangerous situations for pedestrians. This paper proposes a system that warns drivers about pedestrians' presence using the YOLOv8x model. The system is activated based on the presence of pedestrians, creating a light induction effect that helps drivers naturally recognize pedestrians, preventing accidents.
Recent advancements in virtual reality (VR) and robotics have increased interest in integrating physical and virtual environments. Point cloud data from cameras and LiDAR are essential for capturing 3D environmental details. This study proposes a ray-casting-based method for generating real-time point cloud data, enabling efficient virtual environment exploration. The agent prioritizes unexplored regions by analyzing point cloud density, separates ground and obstacles using RANSAC, and plans paths with the A<sup>*</sup> algorithm and Catmull-Rom splines. Implemented in Unity, the method supports real-time processing without complex preprocessing. Experiments confirm its effectiveness, highlighting its potential for various applications.
The purpose of this study is to study ways to eliminate social exclusion through the use of digital devices by people with disabilities in the digital age. The research results are as follows. First, all social exclusion factors had an effect on quality of life. Second, except for health, family, social participation, finance, culture, and employment in the social exclusion factors had an effect on digital device usage attitude. Third, except for health, family, social participation, finance, culture, and employment in the social exclusion factors had an effect on digital device usage self-efficacy. Fourth, digital device usage attitude had a partial mediating effect in the relationship between family, social participation, finance, culture, and employment factors of social exclusion and quality of life. Fifth, digital device usage self-efficacy had a partial mediating effect in the relationship between family, social participation, finance, culture, and employment factors of social exclusion and quality of life. Sixth, differences by gender were found in the family factor of social exclusion. Seventh, differences by age were found in social exclusion, quality of life, digital device usage attitude, and digital device usage self-efficacy. Eighth, differences by disability type were shown in social exclusion, quality of life, digital device usage attitude, and digital device usage self-efficacy. Through this study, we presented effective ways to use digital devices. We hope that these ways will contribute to eliminating social exclusion and improving quality of life for people with disabilities, and to achieving sustainable social integration while increasing digital inclusion for people with disabilities.
The planning of underground logistics pipeline networks is a crucial component of urban underground logistics systems, aiming to find the optimal construction path for the logistics network, improve logistics efficiency, and reduce operational costs. However, due to the complexity and uncertainty of the underground environment, traditional planning methods often fall short. This paper proposes a improved underground logistics pipeline network planning method based on the White Shark Optimization(WSO) algorithm, referred to as LGWSO(White Shark Algorithms Combining Logistic Maps and Gaussian Variations). The proposed method first establishes an underground space model and then uses the LGWSO algorithm for path planning. By adopting chaos initialization method and Gaussian mutation strategy, the performance of the algorithm has been effectively improved. Through simulation experiments, the algorithm has demonstrated significant advantages in optimization accuracy, convergence speed, and robustness. Compared to traditional planning methods, the proposed approach is better suited to handle the complex underground environment, providing an optimized strategy for the construction of urban logistics system underground networks.