• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.49 no.4
• /
• pp.311-320
• /
• 2021

A study is performed for the real time fault diagnosis during operation and health estimation relating to performance deterioration in a turbojet engine used for an unmanned air vehicle. For this study the real time dynamic model is derived from the transient thermodynamic gas path analysis. For real fault conditions which are manipulated for the simulation, the detection techniques are applied such as Kalman filter and probabilistic decision-making approach based on statistical hypothesis test. Thereby the effectiveness is verified by showing good fault detection and isolation performances. For the health estimation with measurement parameters, it shows using an assumed performance degradation that the method by adaptive Kalman filter is feasible in practice for a condition based diagnosis and maintenance.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.45 no.2
• /
• pp.65-76
• /
• 2022

Recently, not only traditional statistical techniques but also machine learning algorithms have been used to make more accurate bankruptcy predictions. But the insolvency rate of companies dealing with financial institutions is very low, resulting in a data imbalance problem. In particular, since data imbalance negatively affects the performance of artificial intelligence models, it is necessary to first perform the data imbalance process. In additional, as artificial intelligence algorithms are advanced for precise decision-making, regulatory pressure related to securing transparency of Artificial Intelligence models is gradually increasing, such as mandating the installation of explanation functions for Artificial Intelligence models. Therefore, this study aims to present guidelines for eXplainable Artificial Intelligence-based corporate bankruptcy prediction methodology applying SMOTE techniques and LIME algorithms to solve a data imbalance problem and model transparency problem in predicting corporate bankruptcy. The implications of this study are as follows. First, it was confirmed that SMOTE can effectively solve the data imbalance issue, a problem that can be easily overlooked in predicting corporate bankruptcy. Second, through the LIME algorithm, the basis for predicting bankruptcy of the machine learning model was visualized, and derive improvement priorities of financial variables that increase the possibility of bankruptcy of companies. Third, the scope of application of the algorithm in future research was expanded by confirming the possibility of using SMOTE and LIME through case application.

• Journal of the Korea Convergence Society
• /
• v.12 no.11
• /
• pp.27-33
• /
• 2021

With increasing interest in key areas of the 4th industrial revolution such as artificial intelligence, deep learning and big data, scientific approaches have developed in order to overcome the limitations of traditional decision-making methodologies. These scientific techniques are mainly used to predict the direction of financial products. In this study, the factors of apartment prices, which are of high social interest, were analyzed through SOM. For this analysis, we extracted the real prices of the apartments and selected a total of 16 input variables that would affect these prices. The data period was set from 1986 to 2021. As a result of examining the characteristics of the variables during the rising and faltering periods of the apartment prices, it was found that the statistical tendencies of the input variables of the rising and the faltering periods were clearly distinguishable. I hope this study will help us analyze the status of the real estate market and study future predictions through image learning.

• Clinics in Shoulder and Elbow
• /
• v.26 no.2
• /
• pp.156-161
• /
• 2023

Background: We analyzed association between viewing two-dimensional computed tomography (2D CT) images in addition to radiographs with radial head treatment recommendations after accounting for patient and surgeon factors in a survey-based experiment. Methods: One hundred and fifty-four surgeons reviewed 15 patient scenarios with terrible triad fracture dislocations of the elbow. Surgeons were randomized to view either radiographs only or radiographs and 2D CT images. The scenarios randomized patient age, hand dominance, and occupation. For each scenario, surgeons were asked if they would recommend fixation or arthroplasty of the radial head. Multi-level logistic regression analysis identified variables associated with radial head treatment recommendations. Results: Reviewing 2D CT images in addition to radiographs had no statistical association with treatment recommendations. A higher likelihood of recommending prosthetic arthroplasty was associated with older patient age, patient occupation not requiring manual labor, surgeon practice location in the United States, practicing for five years or less, and the subspecialties "trauma" and "shoulder and elbow." Conclusions: The results of this study suggest that in terrible triad injuries, the imaging appearance of radial head fractures has no measurable influence on treatment recommendations. Personal surgeon factors and patient demographic characteristics may have a larger role in surgical decision making. Level of evidence: Level III, therapeutic case-control study.

• Korean Journal of Construction Engineering and Management
• /
• v.24 no.5
• /
• pp.35-43
• /
• 2023

Along with the increase in Multi-unit housing defect disputes, the importance of defect management is also increased. However, previous studies have mostly focused on the Multi-unit housing's 'common part'. In addition, there is a lack of research on the system for the 'management office', which is a part of the subject of defect management. These resulted in the lack of defect management capability of the management office and the deterioration of management quality. Therefore, this paper proposes a machine learning-based defect data management system for management offices. The goal is to solve the inconvenience of management by using Optical Character Recognition (OCR) and Natural Language Processing (NLP) modules. This system converts handwritten defect information into online text via OCR. By using the language model, the defect information is regenerated along with the form specified by the user. Eventually, the generated text is stored in a database and statistical analysis is performed. Through this chain of system, management office is expected to improve its defect management capabilities and support decision-making.

• Nuclear Engineering and Technology
• /
• v.55 no.9
• /
• pp.3423-3440
• /
• 2023

In this work, a multivariate time-series machine learning meta-model is developed to predict the transient response of a typical nuclear power plant (NPP) undergoing a steam generator tube rupture (SGTR). The model employs Recurrent Neural Networks (RNNs), including the Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), and a hybrid CNN-LSTM model. To address the uncertainty inherent in such predictions, a Bayesian Neural Network (BNN) was implemented. The models were trained using a database generated by the Best Estimate Plus Uncertainty (BEPU) methodology; coupling the thermal hydraulics code, RELAP5/SCDAP/MOD3.4 to the statistical tool, DAKOTA, to predict the variation in system response under various operational and phenomenological uncertainties. The RNN models successfully captures the underlying characteristics of the data with reasonable accuracy, and the BNN-LSTM approach offers an additional layer of insight into the level of uncertainty associated with the predictions. The results demonstrate that LSTM outperforms GRU, while the hybrid CNN-LSTM model is computationally the most efficient. This study aims to gain a better understanding of the capabilities and limitations of machine learning models in the context of nuclear safety. By expanding the application of ML models to more severe accident scenarios, where operators are under extreme stress and prone to errors, ML models can provide valuable support and act as expert systems to assist in decision-making while minimizing the chances of human error.

• Atmosphere
• /
• v.32 no.1
• /
• pp.1-15
• /
• 2022

Recently, Japan's Meteorological Research Institute presented the d4PDF database (Database for Policy Decision-Making for Future Climate Change, d4PDF) through large-scale climate ensemble simulations to overcome uncertainty arising from variability when the general circulation model represents extreme-scale precipitation. In this study, the change of precipitation characteristics between the historical and future climate conditions in the Yongdam-dam basin was analyzed using the d4PDF data. The result shows that annual mean precipitation and seasonal mean precipitation increased by more than 10% in future climate conditions. This study also performed an analysis on the change of the return period rainfall. The annual maximum daily rainfall was extracted for each climatic condition, and the rainfall with each return period was estimated. In this process, we represent the extreme-scale rainfall corresponding to a very long return period without any statistical model and method as the d4PDF provides rainfall data during 3,000 years for historical climate conditions and during 5,400 years for future climate conditions. The rainfall with a 50-year return period under future climate conditions exceeded the rainfall with a 100-year return period under historical climate conditions. Consequently, in future climate conditions, the magnitude of rainfall increased at the same return period and, the return period decreased at the same magnitude of rainfall. In this study, by using the d4PDF data, it was possible to analyze the change in extreme magnitude of rainfall.

• International Journal of Advanced Culture Technology
• /
• v.11 no.4
• /
• pp.223-232
• /
• 2023

The purpose of this study is to determine the impact of social network sports community consciousness on loyalty and sports attitude. In order to achieve this research purpose, the population of the study was selected as adult men and women over the age of 20 who are active in the social network sports community in 2022. The sampling method used cluster random sampling to select a total of 300 people, 150 men and 150 women, as research subjects. The survey tool used was the questionnaire method, and the questionnaire whose reliability and validity had been verified in previous studies at home and abroad was used by requoting, modifying, or supplementing it to suit the purpose of this study. It was also structured on a 5-point scale. Frequency analysis, factor analysis, reliability analysis, simple regression analysis, and multiple regression analysis were performed on the collected data using the statistical program SPSS Windows 20.0 Version. The results obtained through this process are as follows. First, social network sports community consciousness was found to have a partial effect on loyalty. Second, social network sports community consciousness was found to have a partial effect on sports attitudes. Third, social network sports community loyalty was found to have a partial effect on sports attitudes. Considering these results, various activities such as decision-making process, relationship formation, and opinion expression of modern people are carried out by the O-line community. In addition, while in the past it was a format that led from offline activities to online activities, currently, there are more and more formats that lead from online activities to offline activities. Therefore, modern people's SNS sports community activities provide many experiences, which creates a sense of community and sports attitudes are formed based on this. This can be said to lead to loyal activities.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.97-109
• /
• 2024

This research aims to investigate how major cities in Korea utilize smart city-related technologies, develop key performance indicators (KPIs) to measure the smartness and efficiency of cities, and propose a methodology for assessing and suggesting smart city policy directions based on Data Envelopment Analysis (DEA). Referring to the CITYkeys Smart City Performance Measurement Framework, 10 key performance indicators (KPIs) were derived. For each KPI, city statistical data were allocated to input and output variables, and 15 cities were assigned as Decision Making Units (DMUs). The DEA methodology was employed to evaluate the operational efficiency and scale profitability of cities, providing insights into the operational efficiency of each city. Finally, the operational efficiency among DMUs was ranked to propose smart city policy directions for each city.

• Journal of Chest Surgery
• /
• v.57 no.1
• /
• pp.36-43
• /
• 2024

Background: Coronavirus disease 2019 (COVID-19) can lead to acute respiratory failure, which frequently necessitates invasive mechanical ventilation and extracorporeal membrane oxygenation (ECMO). However, the limited availability of ECMO resources poses challenges to patient selection and associated decision-making. Consequently, this retrospective single-center study was undertaken to evaluate the characteristics and clinical outcomes of patients with COVID-19 receiving ECMO. Methods: Between March 2020 and July 2022, 65 patients with COVID-19 were treated with ECMO and were subsequently reviewed. Patient demographics, laboratory data, and clinical outcomes were examined, and statistical analyses were performed to identify risk factors associated with mortality. Results: Of the patients studied, 15 (23.1%) survived and were discharged from the hospital, while 50 (76.9%) died during their hospitalization. The survival group had a significantly lower median age, at 52 years (interquartile range [IQR], 47.5-61.5 years), compared to 64 years (IQR, 60.0-68.0 years) among mortality group (p=0.016). However, no significant differences were observed in other underlying conditions or in factors related to intervention timing. Multivariable analysis revealed that the requirement of a change in ECMO mode (odds ratio [OR], 366.77; 95% confidence interval [CI], 1.92-69911.92; p=0.0275) and the initiation of continuous renal replacement therapy (CRRT) (OR, 139.15; 95% CI, 1.95-9,910.14; p=0.0233) were independent predictors of mortality. Conclusion: Changes in ECMO mode and the initiation of CRRT during management were associated with mortality in patients with COVID-19 who were supported by ECMO. Patients exhibiting these factors require careful monitoring due to the potential for adverse outcomes.