• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.14 no.3
• /
• pp.201-207
• /
• 2021

Deep learning, a type of machine learning, performs learning while changing the weights as it progresses through each learning process. Tensor Flow and Keras provide the results of the end of the learning in graph form. Thus, If an error occurs, the result must be discarded. Consequently, existing technologies provide a function to roll back learning results, but the rollback function is limited to results up to five times. Moreover, they applied the concept of MLOps to track the deep learning process, but no rollback capability is provided. In this paper, we construct a system that manages the intermediate value of the learning process by blockchain to record the intermediate learning process and can rollback in the event of an error. To perform the functions of blockchain, the deep learning process and the rollback of learning results are designed to work by writing Smart Contracts. Performance evaluation shows that, when evaluating the rollback function of the existing deep learning method, the proposed method has a 100% recovery rate, compared to the existing technique, which reduces the recovery rate after 6 times, down to 10% when 50 times. In addition, when using Smart Contract in Ethereum blockchain, it is confirmed that 1.57 million won is continuously consumed per block creation.

• The Journal of Industrial Distribution & Business
• /
• v.13 no.6
• /
• pp.39-49
• /
• 2022

Purpose: As part of these efforts, in-depth research is needed on efficient and practical utilization of underground spaces and underground shopping malls. In addition, efforts are being made to find effective alternatives to various problems currently occurring in underground shopping malls, but it is not easy. In addition, the development entity and the maintenance entity are different from each other, and the management is not unified, making it difficult to maintain underground shopping malls. From this point of view, it can be said that it is time to actively and specifically discuss ways to revitalize underground shopping malls. Data and methodology: In the domestic distribution environment, traditional markets and shops are stagnating due to rapid changes in consumption patterns, such as the spread of large companies with advanced distribution techniques such as hypermarkets, shopping malls, and SCM, the rapid increase in Internet and home shopping, and the importance of convenience for young consumers. In order to revitalize underground shopping malls, it is necessary to strengthen the organization and self-rescue efforts of merchants' associations, change consciousness through merchant education, change to specialized markets, find nuclear stores and representative restaurants, and support the hardware sector. Results: The connection of underground shopping malls in each region of the country, where commercial districts are separated from each other, will also play an important role in reviving the function of the city in the future. To do this, it is first necessary to connect underground shopping malls that have been cut off. In other words, connection between connectable underground shopping malls should be promoted. Of course, long-term projects should be promoted step by step, and many consultations should be made on how to connect with the ground for each local government. Conclusion: This is because in the future, the underground space cannot just be a walking place, but another space of the Korean Wave where you can experience satisfying the five senses. K-shopping Hallyu content can be created by creating a characteristic story for each underground shopping mall in the city, permanently this story-oriented event, and creating a safe and elegant environment. If there is a story, so-called "Senomi Shopping" will be possible. A new Korean Wave will be created that can satisfy "the fun of writing, the fun of seeing, and the fun of feeling" at the same time.

• Korean Journal of Construction Engineering and Management
• /
• v.23 no.4
• /
• pp.3-14
• /
• 2022

Autonomous equipment requires a large amount of data from various environments. However, it takes a lot of time and cost for an experiment in a real construction sites, which are difficulties in data collection and processing. Therefore, this study aims to develop a virtual environment for autonomous tower cranes technology development and validation. The authors defined automation functions and operation conditions of tower cranes with three performance criteria: operational design domain, object and event detection and response, and minimum functional conditions. Afterward, this study developed a virtual environment for learning and validation for autonomous functions such as recognition, decision making, and control using the Unity game engine. Validation was conducted by construction industry experts with a fidelity which is the representative matrix for virtual environment assessment. Through the virtual environment platform developed in this study, it will be possible to reduce the cost and time for data collection and technology development. Also, it is also expected to contribute to autonomous driving for not only tower cranes but also other construction equipment.

• Journal of Digital Convergence
• /
• v.20 no.5
• /
• pp.227-232
• /
• 2022

This study is an exploratory study on the sustainable management of MICE events, and was conducted using the FGI method. Twelve experts in the MICE field, industry, and academia were selected as the subjects of the study. The reason is that understanding and specialty of those targets provides research reliability and validity. The study period was 3 months from June to August 2021. As a result, it is very important to prepare a sustainable development strategy in the MICE industry, and in particular, the need for guidelines to be practiced during event operation from the organizer's point of view was derived. In addition, to derive items that can be used in practice based on the theoretical basis, and it was necessary to derive research results using internationally recognized Sustainable Development Goals (UNSDGs) and sustainable MICE research developed in the fields of tourism and MICE. Therefore, through the verification of such experts in the field of MICE, theoretical and practical guidelines from the perspective of the organizer that can be used when holding MICE were developed. The results of the study will provide implications for establishing more effective strategies for hosting sustainable MICE events in the future.

• Korean Journal of Construction Engineering and Management
• /
• v.23 no.5
• /
• pp.3-14
• /
• 2022

Construction projects have high potentials of claims among a variety of stakeholders. Claims on their own are not disputes but they have high potentials leading to disputes if agreements are not made between parties due to conflicting opinions. In the event of the construction disputes between clients and contractors, it could give negative impacts to both parties and, to minimize or pro-actively manage construction disputes, the role of clients is more significant. The objective of the study is to analyze a level of associations between the types of disputes and causes of construction projects based on the association rule analysis, and to identify and discuss key characteristics and implications from client's perspectives. The study analyzes associations between the types of disputes and causes, and also identifies those with a high level of associations. It also presents the outcomes of more systematic analysis compared to descriptive statistics just based on frequencies. Through the analysis of the data cases, the study proposes the directions to resolve the causes of disputes from client's perspectives. It can assist to improve understandings of the relationships between the types of disputes and causes and to pro-actively manage the disputes of construction projects.

• Journal of Korea Water Resources Association
• /
• v.54 no.7
• /
• pp.453-462
• /
• 2021

As the frequency of localized heavy rainfall has increased during recent years, the importance of high-resolution radar data has also increased. This study aims to correct the bias of Dual Polarization radar that still has a spatial and temporal bias. In many studies, various statistical techniques have been attempted to correct the bias of radar rainfall. In this study, the bias correction of the S-band Dual Polarization radar used in flood forecasting of ME was implemented by a Convolutional Autoencoder (CAE) algorithm, which is a type of Convolutional Neural Network (CNN). The CAE model was trained based on radar data sets that have a 10-min temporal resolution for the July 2017 flood event in Cheongju. The results showed that the newly developed CAE model provided improved simulation results in time and space by reducing the bias of raw radar rainfall. Therefore, the CAE model, which learns the spatial relationship between each adjacent grid, can be used for real-time updates of grid-based climate data generated by radar and satellites.

• Journal of Korea Entertainment Industry Association
• /
• v.15 no.5
• /
• pp.115-128
• /
• 2021

The purpose of this study is to examine the importance and satisfaction perceived in the purchase of goods and the use of a court in badminton club members. The results will be used for basic data to increase club members and present the methods to activate badminton. The survey on goods, price, programs, facilities, staff, and publicity was conducted. The IPA matrix was applied for data processing. The following conclusions were drawn. First, as a result of analyzing the ranking of importance and satisfaction, the first place of importance was coach's professionalism of staff factors, followed by safety of facility factors and program contents and effects of program factors. The first place of satisfaction was cleanliness and management of facility factors, followed by coach's professionalism of staff factors and staff's kindness of staff factors. Second, as a result of the IPA matrix of importance and satisfaction, Quadrant I included appropriateness of training time and program contents and effect of program factors, parking size and cleanliness and management of facility factors, coach's professionalism and staff's service attitude of staff factors, and customer service and complaint resolution of publicity factors. Quadrant II showed appropriateness of price, value for money, and discount policy of price factors and materials and design of goods factors. Quadrant III included excellent customer service of goods of goods factors, various program construction of program factors, court location and accessibility, and various convenient facilities of facility factors, and various publicity and event programs, website construction, and various publicity strategies of publicity factors. Quadrant IV showed brand value of goods, awareness, and brand specialty of goods of goods factors.

• Journal of Korean Society of Disaster and Security
• /
• v.14 no.4
• /
• pp.17-27
• /
• 2021

Streamflow prediction is a very vital disaster mitigation approach for effective flood management and water resources planning. Lately, torrential rainfall caused by climate change has been reported to have increased globally, thereby causing enormous infrastructural loss, properties and lives. This study evaluates the contribution of rainfall to streamflow prediction in normal and peak rainfall scenarios, typical of the recent flood at Piney Resort in Vernon, Hickman County, Tennessee, United States. Daily streamflow, water level, and rainfall data for 20 years (2000-2019) from two USGS gage stations (03602500 upstream and 03599500 downstream) of the Piney River watershed were obtained, preprocesssed and fitted with Long short term memory (LSTM) model. Tensorflow and Keras machine learning frameworks were used with Python to predict streamflow values with a sequence size of 14 days, to determine whether the model could have predicted the flooding event in August 21, 2021. Model skill analysis showed that LSTM model with full data (water level, streamflow and rainfall) performed better than the Naive Model except some rainfall models, indicating that only rainfall is insufficient for streamflow prediction. The final LSTM model recorded optimal NSE and RMSE values of 0.68 and 13.84 m³/s and predicted peak flow with the lowest prediction error of 11.6%, indicating that the final model could have predicted the flood on August 24, 2021 given a peak rainfall scenario. Adequate knowledge of rainfall patterns will guide hydrologists and disaster prevention managers in designing efficient early warning systems and policies aimed at mitigating flood risks.

• Korean Association of Arts Management
• /
• no.56
• /
• pp.305-327
• /
• 2020

This study started with the question of whether the museum oral history recording projects, which have been increasing steadily in recent years, are being used in various forms, especially in exhibitions. This paper is emphasized on the need for the oral history-related projects to lead to various museum activities including exhibitions and educations and so on. As a practical example of this, to explore the future directions and tasks for oral history projects in museums 'A Research Project of Producing Oral History Video Clips for the Exhibition of IMF Financial Crisis Situations' of National Museum of Korean Contemporary History is analyzed. This research argues that oral history functions as an exhibition representation device that more actively reveal the reality of a specific historical event. Therefore, this study suggests that the museum can be developed as a venue for various discourses in which citizens participate actively using oral history.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.2
• /
• pp.119-129
• /
• 2023

Sleep apnea causes various secondary disease such as hypertension, stroke, myocardial infarction, depression and cognitive impairment. Early detection and continuous management of sleep apnea are urgently needed since it causes cardio-cerebrovascular diseases. In this study, wearable device for monitoring respiration during sleep using PVDF film was developed to detect vibration through trachea caused by breathing, which determines normal breathing and sleep apnea. Variables such as respiration rate and apnea were extracted based on the detected breathing sound data, and a noise reduction algorithm was established to minimize the effect even when there is a noise signal. In addition, it was confirmed that irregular breathing patterns can be analyzed by establishing a moving threshold algorithm. The results show that the accuracy of the respiratory rate from the developed device was 98.7% comparing with the polysomnogrphy result. Accuracy of detection for sleep apnea event was 92.6% and that of the sleep apnea duration was 94.0%. The results of this study will be of great help to the management of sleep disorders and confirmation of treatment by commercialization of wearable devices that can monitor sleep information easily and accurately at home during daily life and confirm the progress of treatment.