• Journal of Korean Tunnelling and Underground Space Association
• /
• v.23 no.6
• /
• pp.549-558
• /
• 2021

In order to keep infrastructures such as tunnels and underground facilities safe, cracks of concrete lining in tunnel should be detected by regular inspections. Since regular inspections are accomplished through manual efforts using maintenance lift vehicles, it brings about traffic jam, exposes works to dangerous circumstances, and deteriorates consistency of crack inspection data. This study aims to provide methodology to automatically extract cracks from tunnel concrete lining images generated by the existing tunnel image acquisition system. Specifically, we train a deep learning based semantic segmentation model with open dataset, and evaluate its performance with the dataset from the existing tunnel image acquisition system. In particular, we compare the model performance in case of using all of a public dataset, subset of the public dataset which are related to tunnel surfaces, and the tunnel-related subset with negative examples. As a result, the model trained using the tunnel-related subset with negative examples reached the best performance. In the future, we expect that this research can be used for planning efficient model training strategy for crack detection.

• Smart Media Journal
• /
• v.12 no.2
• /
• pp.66-75
• /
• 2023

Recently, the number of cloud web applications is increasing owing to the accelerated migration of enterprises and public sector information systems to the cloud. Traditional network attacks on cloud web applications are characterized by Denial of Service (DoS) attacks, which consume network resources with a large number of packets. However, HTTP DoS attacks, which consume application resources, are also increasing recently; as such, developing security technologies to prevent them is necessary. In particular, since low-bandwidth HTTP DoS attacks do not consume network resources, they are difficult to identify using traditional security solutions that monitor network metrics. In this paper, we propose a new detection model for detecting HTTP DoS attacks on cloud web applications by collecting the application metrics of web servers and learning them using machine learning. We collected 18 types of application metrics from an Apache web server and used five machine learning and two deep learning models to train the collected data. Further, we confirmed the superiority of the application metrics-based machine learning model by collecting and training 6 additional network metrics and comparing their performance with the proposed models. Among HTTP DoS attacks, we injected the RUDY and HULK attacks, which are low- and high-bandwidth attacks, respectively. As a result of detecting these two attacks using the proposed model, we found out that the F1 scores of the application metrics-based machine learning model were about 0.3 and 0.1 higher than that of the network metrics-based model, respectively.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.10a
• /
• pp.562-565
• /
• 2022

Recently, the increasing number of overloaded vehicles on the road poses a risk to traffic safety, such as falling objects, road damage, and chain collisions due to the abnormal weight distribution, and can cause great damage once an accident occurs. However, this irregular weight distribution is not possible to be recognized with the current weight measurement system for vehicles on roads. To address this limitation, we propose to build an object detection-based AI model to identify overloaded vehicles that cause such social problems. In addition, we present a simple yet effective method to construct an object detection model for the large-scale vehicle images. In particular, we utilize the large-scale of vehicle image sets provided by open AI-Hub, which include the overloaded vehicles from the CCTV, black box, and hand-held camera point of view. We inspected the specific features of sizes of vehicles and types of image sources, and pre-processed these images to train a deep learning-based object detection model. Finally, we demonstrated that the detection performance of the overloaded vehicle was improved by about 23% compared to the one using raw data. From the result, we believe that public big data can be utilized more efficiently and applied to the development of an object detection-based overloaded vehicle detection model.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.241-265
• /
• 2023

A corporate insolvency prediction model serves as a vital tool for objectively monitoring the financial condition of companies. It enables timely warnings, facilitates responsive actions, and supports the formulation of effective management strategies to mitigate bankruptcy risks and enhance performance. Investors and financial institutions utilize default prediction models to minimize financial losses. As the interest in utilizing artificial intelligence (AI) technology for corporate insolvency prediction grows, extensive research has been conducted in this domain. However, there is an increasing demand for explainable AI models in corporate insolvency prediction, emphasizing interpretability and reliability. The SHAP (SHapley Additive exPlanations) technique has gained significant popularity and has demonstrated strong performance in various applications. Nonetheless, it has limitations such as computational cost, processing time, and scalability concerns based on the number of variables. This study introduces a novel approach to variable selection that reduces the number of variables by averaging SHAP values from bootstrapped data subsets instead of using the entire dataset. This technique aims to improve computational efficiency while maintaining excellent predictive performance. To obtain classification results, we aim to train random forest, XGBoost, and C5.0 models using carefully selected variables with high interpretability. The classification accuracy of the ensemble model, generated through soft voting as the goal of high-performance model design, is compared with the individual models. The study leverages data from 1,698 Korean light industrial companies and employs bootstrapping to create distinct data groups. Logistic Regression is employed to calculate SHAP values for each data group, and their averages are computed to derive the final SHAP values. The proposed model enhances interpretability and aims to achieve superior predictive performance.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.4D
• /
• pp.565-571
• /
• 2006

This study aimed to analyze access mode choice behavior for KTX Passengers. To fulfill the aims of this study, Dong-Daegu Station was selected as a station for a case study. This study takes place in two stages. These are (i) descriptive statistical analysis of transportation status before and after introduction of the KTX, (ii) empirical model estimation for analyzing access mode choice behavior. This study makes use of the data from travel survey from Daegu metropolitan area. The main part of the survey was carried out in the KTX Dong-Daegu station. The data was collected from a sample of 1,800 individuals. The survey data includes the information on travel from Dong-Daegu station to Seoul. From descriptive statistical analysis of transportation status before and after introduction of the KTX, it is found that revealed demand of the KTX is lower than that expected. Moreover, it is found that the low demand of the KTX stems from high cost for the KTX itself and inconvenience( including travel time and cost) of access mode. In order to analyze mode choice behavior for accessing Dong-Daegu station, multinomial logit model structure is used. For the model specification, a variety of behavioral assumptions about the factors which affect the access mode choice, were considered. From the empirical model estimation, it si found that access travel time and access travel cost are significant in choosing access mode. Given the empirical evidence, we see that improvement of access transportation system for Dong-Daegu station is very important for enhancing the use of KTX.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.10
• /
• pp.461-470
• /
• 2023

While speech animation generation employing deep learning has been actively researched for English, there has been no prior work for Korean. Given the fact, this paper for the very first time employs supervised deep learning to generate Korean speech animation. By doing so, we find out the significant effect of deep learning being able to make speech animation research come down to speech recognition research which is the predominating technique. Also, we study the way to make best use of the effect for Korean speech animation generation. The effect can contribute to efficiently and efficaciously revitalizing the recently inactive Korean speech animation research, by clarifying the top priority research target. This paper performs this process: (i) it chooses blendshape animation technique, (ii) implements the deep-learning model in the master-servant pipeline of the automatic speech recognition (ASR) module and the facial action coding (FAC) module, (iii) makes Korean speech facial motion capture dataset, (iv) prepares two comparison deep learning models (one model adopts the English ASR module, the other model adopts the Korean ASR module, however both models adopt the same basic structure for their FAC modules), and (v) train the FAC modules of both models dependently on their ASR modules. The user study demonstrates that the model which adopts the Korean ASR module and dependently trains its FAC module (getting 4.2/5.0 points) generates decisively much more natural Korean speech animations than the model which adopts the English ASR module and dependently trains its FAC module (getting 2.7/5.0 points). The result confirms the aforementioned effect showing that the quality of the Korean speech animation comes down to the accuracy of Korean ASR.

• The Journal of Bigdata
• /
• v.9 no.1
• /
• pp.101-113
• /
• 2024

Federated learning has garnered attention as an efficient method for training machine learning models in a distributed environment while maintaining data privacy and security. This study proposes a novel FedRFBagging algorithm to optimize the performance of random forest models in such federated learning environments. By dynamically adjusting the trees of local random forest models based on client-specific data characteristics, the proposed approach reduces communication costs and achieves high prediction accuracy even in environments with numerous clients. This method adapts to various data conditions, significantly enhancing model stability and training speed. While random forest models consist of multiple decision trees, transmitting all trees to the server in a federated learning environment results in exponentially increasing communication overhead, making their use impractical. Additionally, differences in data distribution among clients can lead to quality imbalances in the trees. To address this, the FedRFBagging algorithm selects only the highest-performing trees from each client for transmission to the server, which then reselects trees based on impurity values to construct the optimal global model. This reduces communication overhead and maintains high prediction performance across diverse data distributions. Although the global model reflects data from various clients, the data characteristics of each client may differ. To compensate for this, clients further train additional trees on the global model to perform local optimizations tailored to their data. This improves the overall model's prediction accuracy and adapts to changing data distributions. Our study demonstrates that the FedRFBagging algorithm effectively addresses the communication cost and performance issues associated with random forest models in federated learning environments, suggesting its applicability in such settings.

• Journal of the Korean Geosynthetics Society
• /
• v.11 no.4
• /
• pp.27-35
• /
• 2012

Facings of mechanically stabilized earth retaining walls have function to fix the reinforcement and prevent backfill loss, but the walls are lack of structural rigidity capable of resisting applied loads. The reinforced subgrade with rigid wall was developed to have the structural functions under train loading. Though it has lots of advantages such as small deformation after construction, its negative side effects of economics and difficult construction were mainly mentioned and not practically used. To apply it for railroad subgrade, this study focus on the construction cost down and the enhancement of constructability without functional loss. To do so, the behaviors of reinforced subgrade with rigid wall were evaluated with the change of the vertical spacing and length of reinforcement. Small scale model tests (1/10 scale) and 3 m full scale tests were performed to evaluate deformation characteristics of reinforced subgrade under simulated train loading. Even though it uses short reinforcement, it showed small horizontal displacement of wall and plastic settlement of subgrade. Also, it was verified that not only 30 cm but also 40 cm of vertical spacing of reinforcement had good performance in serviceability aspects.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.17 no.11
• /
• pp.537-544
• /
• 2016

National Competency Standards (NCS) need to be introduced to train newly hired staff and to gradually improve employees' work performance in the information security industry. In particular, the introduction of a new NCS curriculum for new hires is important in order to retain and efficiently manage professionals in the information security field. However, the legacy NCS is not clearly designed for the information security field. So a formal curriculum has been suggested for institutions training the information security workforce. Therefore, this study establishes a competency unit based on the types of personnel, their duties, and required knowledge. To select the competency unit, this study reviewed prior research to understand the required skills and work knowledge, and reviewed recruitment-based NCS that public agencies and public and private companies have carried out, including them in the study. The selected competency unit was classified into a required competency unit and an elective competency unit based on the importance of the duties and the demands of training. Through a verification process for the new, licensed career path model in the NCS information and communications field, this study suggests updated NCS competency units and required courses to provide an appropriate NCS curriculum for newly hired employees in the information security industry.

• Journal of Korean Academy of Nursing
• /
• v.26 no.1
• /
• pp.194-222
• /
• 1996

Various accidents and injuries are currently occurring in Korea at increasingly high rates. Good quality emergency care service is urgently needed to cope with these various forms of accidents and injuries. In order to develop a sound emergency care system, there need to be a plan to educate and train professionals specifically in emergency care. One solution for the on going problem would be to educate and train emergency clinical nurse specialists. This study on a strategy for curriculum development for emergency clinical nurse specialist was based on the following five content areas, developed from literature related to the curriculum of emergency nursing and emergency care situation : 1. Nurses working in the emergency rooms of three university hospitals were analyzed for six days to identify categories of nursing activities. 2. Two hundreds and eleven nurses working in the emergency rooms of 12 university hospitals were surveyed to identify needs for educational content that should be included in a curriculum for the clinical nurse specialist. 3. Examination of the environment in which emergency management was provided. 4. Identification of characteristics of patients in the emergency room. 5. The role of emergency clinical nurse specialist was identified through literature, recent data, and research materials. The following curriculum was formulated using the above mentioned process. 1. The philosophy of education for emergency clinical nurse specialist was established through a realistic philosophical framework. In this frame, client, environment, health, nursing, and learning have been defined. 2. The purpose of education is framed on individual development, social structure, nursing process and responsibility along with the role and function of the emergency clinical nurse specialist. 3. The central theme was based on human, environment, health and nursing. 4. The elements of structure in the curriculum content were divided to include two major threads, I, e., vertical and horizontal : The vertical thread to consist of the client, life cycle, education, research, leadership and consultation, and the horizontal thread to consist of level of nursing (prevention to rehabilitation), and health to illness based on the health care system developed by Betty Neuman system model. 5. Behavioral objectives for education were structured according to the emergency clinical nurse specialist role and function as a master degree prepared in various emergency settings. 6. The content of the curriculum consisted of three core courses(9 credits), five major courses(15 credits), six elective courses(12 credits) and six prerequisite courses (12 credits). Thus 48 credits are required. Recommendations : 1. To promote tile quality of the emergency care system, the number of emergency professionals, has to be expanded. Further the role and function of the emergency clinical nurse specialist needs to be specified in both the medical law and the Nursing Practice Act. 2. In order to upgrade the qualification of emergency clinical nurse specialists, the course should be given as part of the graduate Program. 3. Certification should be issued through the Korean Nurses Association.