• Geomechanics and Engineering
• /
• 제31권6호
• /
• pp.545-556
• /
• 2022

Evaluation and optimization of tunnel wall convergence (TWC) plays a vital role in preventing potential problems during tunnel construction and utilization stage. When convergence occurs at a high rate, it can lead to significant problems such as reducing the advance rate and safety, which in turn increases operating costs. In order to design an effective solution, it is important to accurately predict the degree of TWC; this can reduce the level of concern and have a positive effect on the design. With the development of soft computing methods, the use of deep learning algorithms and neural networks in tunnel construction has expanded in recent years. The current study aims to employ the long-short-term memory (LSTM) deep neural network predictor model to predict the TWC, based on 550 data points of observed parameters developed by collecting required data from different tunnelling projects. Among the data collected during the pre-construction and construction phases of the project, 80% is randomly used to train the model and the rest is used to test the model. Several loss functions including root mean square error (RMSE) and coefficient of determination (R²) were used to assess the performance and precision of the applied method. The results of the proposed models indicate an acceptable and reliable accuracy. In fact, the results show that the predicted values are in good agreement with the observed actual data. The proposed model can be considered for use in similar ground and tunneling conditions. It is important to note that this work has the potential to reduce the tunneling uncertainties significantly and make deep learning a valuable tool for planning tunnels.

• Clinical Endoscopy
• /
• 제57권4호
• /
• pp.476-485
• /
• 2024

Background/Aims: Sedation has become a standard practice for patients undergoing gastrointestinal (GI) endoscopy. However, considering the serious cardiopulmonary adverse events associated with sedatives, it is important to identify patients at high risk. Machine learning can generate reasonable prediction for a wide range of medical conditions. This study aimed to evaluate the risk factors associated with sedation during GI endoscopy and develop a predictive model for hypoxia during endoscopy under sedation. Methods: This prospective observational study enrolled 446 patients who underwent sedative endoscopy at the Korea University Ansan Hospital. Clinical data were used as predictor variables to construct predictive models using the random forest method that is a machine learning algorithm. Results: Seventy-two of the 446 patients (16.1%) experienced life-threatening hypoxia requiring immediate medical intervention. Patients who developed hypoxia had higher body weight, body mass index (BMI), neck circumference, and Mallampati scores. Propofol alone and higher initial and total dose of propofol were significantly associated with hypoxia during sedative endoscopy. Among these variables, high BMI, neck circumference, and Mallampati score were independent risk factors for hypoxia. The area under the receiver operating characteristic curve for the random forest-based predictive model for hypoxia during sedative endoscopy was 0.82 (95% confidence interval, 0.79-0.86) and displayed a moderate discriminatory power. Conclusions: High BMI, neck circumference, and Mallampati score were independently associated with hypoxia during sedative endoscopy. We constructed a model with acceptable performance for predicting hypoxia during sedative endoscopy.

• Investigative Magnetic Resonance Imaging
• /
• 제18권2호
• /
• pp.120-132
• /
• 2014

목적: 원발성 뇌종양환자에서 방사선 치료 후 추적 자기공명영상에서 새로 생긴 조영증강 뇌병변에 대해 종양재발과 지연성 방사선치료연관변화의 감별에 있어서 확산강조영상 (DWI), 역동적조영관류영상 (DSC PWI), 자화율강조영상 (SWI)의 진단적 가치를 서로 비교하고자 한다. 대상과 방법: 원발성 뇌종양으로 이전에 방사선치료를 받았던 환자 중, 방사선치료 종료 최소 1년 이후에 추적 자기공명영상에서 새롭게 조영증강 되는 병변을 가진 24명의 환자를 대상으로 연구하였다. 새롭게 조영증강 되는 병변은 14명의 종양재발과 10명의 방사선치료연관변화로 확인되었다. 종양재발과 방사선치료연관변화 두 환자 군의 여러변수들은 비대응표본 t 검정을 실시하여 비교 분석하였다. 다중변수 로지스틱 회귀 분석을 이용하여 DWI, DSC PWI, SWI 각 영상의 정량 분석을 통해 얻은 apparent diffusion coefficient (ADC), normalized cerebral blood volume (nCBV), proportion of dark signal intensity (proSWI) 값 중 두 군을 감별해 내는 최상의 예측 변수 (best predictor)를 정하였다. 이후 수신자 조작 특성 (Receiver operating characteristics, ROC) 분석을 통하여 best predictor의 정확도, 민감도, 특이도를 평가하였다. 결과: 방사선치료연관변화 군과 비교하여 종양재발 군에서 평균 nCBV 값이 유의하게 높았고 (P=.004), 평균 proSWI 값은 유의하게 낮았다 (P<.001). 반면, 평균 ADC 값은 두 군간에 유의한 차이를 보이지 않았다. 다중변수 로지스틱 회귀 분석 결과 proSWI 값만이 통계적으로 유의한, 감별 가능한 독립변수였으며, 민감도, 특이도, 정확도는 각각 78.6% (11 of 14), 100% (10 of 10), 87.5% (21 of 24) 였다. 결론: 뇌종양 환자에서 방사선치료 종료 최소 1년 이후에 새로 보이는 조영증강 병변의 감별에 있어 proSWI 값이 가장 중요한 변수인 것으로 나타났다.

• 한국수자원학회논문집
• /
• 제42권3호
• /
• pp.213-225
• /
• 2009

기존의 정상성 Markov Chain 모형은 자료 자체의 Markov 특성만을 고려하여 모의하는 기법으로서 수자원 설계에서 여러 가지 목적으로 이용되어 지고 있다. 그러나 일강수량의 천이확률 및 매개변수 등이 과거와 일정하다는 정상성을 기본 가정으로 하기 때문에 평균의 변동성 등과 같은 외부충격을 모형에 적용할 수 없다. 이러한 관점에서 본 연구의 가장 큰 목적은 기존일강수량 모형을 외부인자를 받아들일 수 있는 모형으로 개발하는 것이다. 즉, Markov Chain 모형의 매개변수인 천이확률과 확률분포형의 매개변수 등을 연결함수(link function)를 통해 외부인자와 연동하도록 하였으며 정준상관분석을 통해 매개변수를 추정하였다. 개발된 모형을 서울지방 1961-2006년까지의 일강수량 자료를 대상으로 검증하는 절차를 가졌다. 추정된 결과를 보면 계절강수량의 특성뿐만 아니라 일강수량의 특성 또한 적절하게 모의되는 것을 확인할 수 있다. 따라서 본 연구에서 개발된 모형은 GCM 예측결과를 입력자료로 활용한다면 일강수계열의 장단기 모의를 위한 downscaling 기법으로 사용될 수 있다. 또한, 기후변화 시나리오가 입력자료로 이용된다면 기후변화에 따른 수자원 영향 평가를 위한 downscaling 기법으로 활용이 가능할 것으로 판단된다.

• 한국수자원학회논문집
• /
• 제47권9호
• /
• pp.825-838
• /
• 2014

정상성 마코프 연쇄 모형은 일강우모의 모형으로 광범위하게 이용되고 있다. 하지만 정상성 마코프 연쇄 모형의 기본가정은 통계학적 특성이 시간에 따라 변화하지 않는 것으로, 일강우모의 시에 평균 또는 분산의 경향적 변화를 효과적으로 반영할 수 없다. 이러한 문제점을 인지하여 본 연구에서는 연주기 및 계절변화에 대하여 우수한 모의 능력을 나타내는 GCM의 모의결과를 입력자료로 이용하여 일강우량을 모의하기 위한 통계학적 상세화(downscaling) 기법인 비정상성 은닉 마코프 모형을 개발하였다. 개발된 모형을 낙동강 유역에 존재하는 영주지점, 문경지점 및 구미지점의 관측강우량에 적용한 결과, 일단위 및 계절단위의 강우량의 통계적 특성을 기존 모형에 비하여 개선된 결과를 도출할 수 있었으며, 또한 개발된 모형은 극치강수량 복원에 있어서도 관측값과 보다 유사한 결과를 보여 주었다. 이러한 점에서 정확성이 확보된 GCM 계절예측자료가 입력자료로 NHMM 모형에 활용된다면 예측기반의 일강수 상세화 모형으로 활용될 수 있을 것으로 판단된다. 이와 더불어, 기후변화 시나리오 입력자료가 사용된다면 기후변화 상세화 모형으로서도 적용될 수 있을 것으로 사료된다.

• 국제학술발표논문집
• /
• The 4th International Conference on Construction Engineering and Project Management Organized by the University of New South Wales
• /
• pp.432-438
• /
• 2011

The purpose of the present thesis is to develop bankruptcy prediction models capable of being applied to the Korean construction industry and to deduce an optimal model through comparative evaluation of final developed models. A study population was selected as general contractors in the Korean construction industry. In order to ease the sample securing and reliability of data, it was limited to general contractors receiving external audit from the government. The study samples are divided into a bankrupt company group and a non-bankrupt company group. The bankruptcy, insolvency, declaration of insolvency, workout and corporate reorganization were used as selection criteria of a bankrupt company. A company that is not included in the selection criteria of the bankrupt company group was selected as a non-bankrupt company. Accordingly, the study sample is composed of a total of 112 samples and is composed of 48 bankrupt companies and 64 non-bankrupt companies. A financial ratio was used as early predictors for development of an estimation model. A total of 90 financial ratios were used and were divided into growth, profitability, productivity and added value. The MDA (Multivariate Discriminant Analysis) model and BLRA (Binary Logistic Regression Analysis) model were used for development of bankruptcy prediction models. The MDA model is an analysis method often used in the past bankruptcy prediction literature, and the BLRA is an analysis method capable of avoiding equal variance assumption. The stepwise (MDA) and forward stepwise method (BLRA) were used for selection of predictor variables in case of model construction. Twenty two variables were finally used in MDA and BLRA models according to timing of bankruptcy. The ROC-Curve Analysis and Classification Analysis were used for analysis of prediction performance of estimation models. The correct classification rate of an individual bankruptcy prediction model is as follows: 1) one year ago before the event of bankruptcy (MDA: 83.04%, BLRA: 93.75%); 2) two years ago before the event of bankruptcy (MDA: 77.68%, BLRA: 78.57%); 3) 3 years ago before the event of bankruptcy (MDA: 84.82%, BLRA: 91.96%). The AUC (Area Under Curve) of an individual bankruptcy prediction model is as follows. : 1) one year ago before the event of bankruptcy (MDA: 0.933, BLRA: 0.978); 2) two years ago before the event of bankruptcy (MDA: 0.852, BLRA: 0.875); 3) 3 years ago before the event of bankruptcy (MDA: 0.938, BLRA: 0.975). As a result of the present research, accuracy of the BLRA model is higher than the MDA model and its prediction performance is improved.

• Ocean Systems Engineering
• /
• 제10권2호
• /
• pp.201-226
• /
• 2020

A systematic numerical comparative study of the performance of semicircular and rectangular submerged breakwaters interacting with solitary waves is the basis of this paper. To accomplish this task, Nwogu's extended Boussinesq model equations are employed to simulate the interaction of the wave with breakwaters. The finite difference technique has been used to discretize the spatial terms while a fourth-order predictor-corrector method is employed for time discretization in our numerical model. The proposed computational scheme uses a staggered-grid system where the first-order spatial derivatives have been discretized with fourth-order accuracy. For validation purposes, five test cases are considered and numerical results have been successfully compared with the existing analytical and experimental results. The performances of the rectangular and semicircular breakwaters have been examined in terms of the wave reflection, transmission, and dissipation coefficients (RTD coefficients) denoted by K_R, K_T, K_D. The latter coefficient K_D emerges due to the non-energy conserving K_R and K_T. Our computational results and graphical illustrations show that the rectangular breakwater has higher reflection coefficients than semicircular breakwater for a fixed crest height, but as the wave height increases, the two reflection coefficients approach each other. un the other hand, the rectangular breakwater has larger dissipation coefficients compared to that of the semicircular breakwater and the difference between them increases as the height of the crest increases. However, the transmission coefficient for the semicircular breakwater is greater than that of the rectangular breakwater and the difference in their transmission coefficients increases with the crest height. Quantitatively, for rectangular breakwaters the reflection coefficients K_R are 5-15% higher while the diffusion coefficients K_D are 3-23% higher than that for the semicircular breakwaters, respectively. The transmission coefficients K_T for rectangular breakwater shows the better performance up to 2.47% than that for the semicircular breakwaters. Based on our computational results, one may conclude that the rectangular breakwater has a better overall performance than the semicircular breakwater. Although the model equations are non-dissipative, the non-energy conserving transmission and reflection coefficients due to wave-breakwater interactions lead to dissipation type contribution.

• 한국산림과학회지
• /
• 제87권1호
• /
• pp.40-49
• /
• 1998

본 연구는 보다 정확한 모수(母數) 추정을 통한 생장(生長)모델의 현실성을 향상시키는데 이용되는 생장 측정간격(임목의 측정 초기 연령 $T_1$과 재측정 연령 $T_2$의 기간)의 적합한 조합을 선택하기 위한 계획을 제공하는데 목적이 있다. 다양한 생장식을 데이터에 적용한 후 가장 적합한 것으로 판정된 생장식을 분석에 이용하였다. 여러 생장식을 분석한 결과 최적의 생장식으로 판명된 더미 변수를 포함하는 변형 Schumacher 방정식을 임분 흉고단면적(胸高斷面績) 생장식과 평균수고(平均樹高) 생장식을 얻기 위하여 이용하였다. 그리고 사용된 자료는 뉴질랜드 남섬 전역에서 측정된 업송(業松)(Pseudotsuga menziesii Mirb.Franco)의 생장 측정기간이 변형되지 않은 데이터와 모든 가능한 생장 측정기간을 포함하는 변형된 2종류의 데이터이었다. 단기의 측정기간에서부터 장기의 측정기간의 범위를 포함하는 데이터(모든 가능한 생장 측정기간을 포함하는 데이터)를 사용할 때 흉고단면적 생장식과 임분 평균수고 생장식에서 모수 추정의 정확성이 증가되는 것이 발견되었다.

• 한국의류학회지
• /
• 제37권7호
• /
• pp.1000-1014
• /
• 2013

This study identifies the determinants that influence attitude toward and the intention to use mobile shopping services through fashion applications (apps) based on the technology acceptance model. In addition, gender and age group differences were examined. Data were collected from subjects who have used smartphone fashion related apps; subsequently, a total of 327 data were analyzed. About 46% of respondents were males, with a mean age of 34.4 years that ranged from 20 to 49 years old. Multiple regression models were developed based on the research model. Perceived usefulness, perceived ease of use, perceived enjoyment, perceived risks (security risk and quality risk), fashion involvement, and fashion app attributes (product attributes and service attributes) were employed as predictors of attitudes towards mobile shopping. Attitudes towards mobile shopping and subjective norms with the aforementioned variables measured the intention to use. Attitudes towards mobile shopping were predicted by perceived enjoyment, perceived usefulness, and service attributes. Attitudes toward mobile shopping and subjective norms were the most important predictors of the intention to use. Gender differences were found in that service attributes were significant for attitudes towards mobile shopping only in the male model. Age differences were also found and perceived usefulness was the most important predictor of attitudes toward mobile shopping among those in their 20's; however, perceived enjoyment was the most important among those in their 30's and 40's. Quality risk was only significant to explain intention to use among those in their 40's. The findings of this study are useful to understand the possibility of the adoption of mobile shopping though fashion apps and provide basic insight into market segmentation.

• 한국정보통신학회논문지
• /
• 제21권11호
• /
• pp.2076-2082
• /
• 2017

우리나라 스마트폰 이용자는 경제활동 인구의 99% 이상 대부분이 사용하고 있으며 초기 형성단계를 지나 포화상태에 도달한 것으로 전문가들은 내다보고 있다. 본 연구는 지배적 디자인 속성이 스마트폰 사용자의 지속사용의도에 미치는 영향을 알아보고자 하였다. 예측변수로는 확장 기술수용모델에서 제시된 인지된 유용성, 인지된 사용 용이성을 선택하였고 지배적 디자인 속성을 매개 조절변수로 선택하여 개념모델을 완성하였다. 연구대상은 부산경남과 익산전북지역에 거주하는 스마트폰 사용자 135명이며 설문을 통해 기초 데이터를 수집하였다. 인구통계학적인 분석은 IBM SPSS Statistics 19로 하였고 Smart PLS를 사용하여 확인적 요인분석과 변수 간의 인과관계에 대한 경로분석을 실시하였다. 분석결과 지속사용의도에 이르는 모든 경로가 유의미한 영향을 미치는 것으로 나타났다. 또한 지배적 디자인 속성이 태도를 매개하여 조절할 때 지속사용의도는 76% 설명력을 가지는 것으로 나타났다.