• Geomechanics and Engineering
• /
• v.31 no.5
• /
• pp.453-460
• /
• 2022

It is essential for geotechnical engineers to conduct studies and make predictions about the stability of slopes, since collapse of a slope may result in catastrophic events. The Gaussian process regression (GPR) approach was carried out for the purpose of predicting the factor of safety (FOS) of the slopes in the study that was presented here. The model makes use of a total of 327 slope cases from Iran, each of which has a unique combination of geometric and shear strength parameters that were analyzed by PLAXIS software in order to determine their FOS. The K-fold (K = 5) technique of cross-validation (CV) was used in order to conduct an analysis of the accuracy of the models' predictions. In conclusion, the GPR model showed excellent ability in the prediction of FOS of slope stability, with an R² value of 0.8355, RMSE value of 0.1372, and MAPE value of 6.6389%, respectively. According to the results of the sensitivity analysis, the characteristics (friction angle) and (unit weight) are, in descending order, the most effective, the next most effective, and the least effective parameters for determining slope stability.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.5
• /
• pp.236-242
• /
• 2024

Software Defect Prediction (SDP) enhances the efficiency of software development by proactively identifying modules likely to contain errors. A major challenge in SDP is improving prediction performance. Recent research has applied deep learning techniques to the field of SDP, with the SAINT model particularly gaining attention for its outstanding performance in analyzing structured data. This study compares the SAINT model with other leading models (XGBoost, Random Forest, CatBoost) and investigates the latest deep learning techniques applicable to SDP. SAINT consistently demonstrated superior performance, proving effective in improving defect prediction accuracy. These findings highlight the potential of the SAINT model to advance defect prediction methodologies in practical software development scenarios, and were achieved through a rigorous methodology including cross-validation, feature scaling, and comparative analysis.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.479-487
• /
• 2021

In this paper, we propose an Sound Event Detection (SED) model using self-training based on a noisy student model. The proposed SED model consists of two stages. In the first stage, a mean-teacher model based on an Residual Convolutional Recurrent Neural Network (RCRNN) is constructed to provide target labels regarding weakly labeled or unlabeled data. In the second stage, a self-training-based noisy student model is constructed by applying different noise types. That is, feature noises, such as time-frequency shift, mixup, SpecAugment, and dropout-based model noise are used here. In addition, a semi-supervised loss function is applied to train the noisy student model, which acts as label noise injection. The performance of the proposed SED model is evaluated on the validation set of the Detection and Classification of Acoustic Scenes and Events (DCASE) 2020 Challenge Task 4. The experiments show that the single model and ensemble model of the proposed SED based on the noisy student model improve F1-score by 4.6 % and 3.4 % compared to the top-ranked model in DCASE 2020 challenge Task 4, respectively.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.24-24
• /
• 2019

This study investigates variations in the Western Pacific Subtropical High (WPSH) and its impact on South Korean precipitation in late summer during the period between 1958 and 2017. Composite analysis reveals that precipitation occurrence is directly linked to the displacement of the WPSH western ridge, a single, large-scale feature of the atmosphere in the Pacific Ocean. When WPSH ridging is located northwest (NW) of its climatological mean position, excessive precipitation is expected in late summer due to enhanced moisture transport. On the other hand, a precipitation deficit is frequently observed when the western ridge is located in the southeast (SE). Different phases of the WPSH are associated with lagged patterns of Pacific and Atlantic atmospheric and oceanic variability, introducing the potential to predict variability in the WPSH western ridge and its climate over northern East Asia by one month. Based on the identified SST patterns, a simple statistical model is developed and improvement in the ability to predict is confirmed through a cross-validation framework. Finally, the potential for further improvements in WPSH-based predictions is addressed.

• Annual Conference of KIPS
• /
• 2010.11a
• /
• pp.276-279
• /
• 2010

소프트웨어 제품 라인 개발 패러다임은 관련 제품들 사이의 공통점과 차이점을 이용해 보다 전략적인 재사용을 가능하게 함으로써 소프트웨어 개발 생산성을 높여 주는 개발 방법론이다. 공통점과 차이점을 분석하고 모델링하기 위해 가장 중요한 모델이 특성 모델이다. 특성 모델은 규모가 커짐에 따라 오류를 포함할 가능성이 커지며 이를 검증하기 위한 자동화된 도구가 필요하다. 본 논문에서는 온톨로지를 자바 언어로 구현 가능하게 해주는 Protege API, OWL기반의 시맨틱 웹 규칙 언어인 SWRL, 규칙 추론 엔진인 Pellet Reasoner 등의 기술을 이용한 특성 모델 검증 도구를 제안한다.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.4
• /
• pp.511-523
• /
• 2023

Research on the integration of unmanned aerial vehicles and deep learning for reinforced concrete damage detection is actively underway. Convolutional neural networks have a high impact on the performance of image classification, detection, and segmentation as backbones. The MobileNet, a pre-trained convolutional neural network, is efficient as a backbone for an unmanned aerial vehicle-based damage detection model because it can achieve sufficient accuracy with low computational complexity. Analyzing vanilla convolutional neural networks and MobileNet under various conditions, MobileNet was evaluated to have a verification accuracy 6.0~9.0% higher than vanilla convolutional neural networks with 15.9~22.9% lower computational complexity. MobileNetV2, MobileNetV3Large and MobileNetV3Small showed almost identical maximum verification accuracy, and the optimal conditions for MobileNet's reinforced concrete damage image feature extraction were analyzed to be the optimizer RMSprop, no dropout, and average pooling. The maximum validation accuracy of 75.49% for 7 types of damage detection based on MobilenetV2 derived in this study can be improved by image accumulation and continuous learning.

• Advances in aircraft and spacecraft science
• /
• v.3 no.4
• /
• pp.367-378
• /
• 2016

Thrust Vectoring is a dynamic feature that offers many benefits in terms of maneuverability and control effectiveness. Thrust vectoring capabilities make the satisfaction of take-off and landing requirements easier. Moreover, it can be a valuable control effector at low dynamic pressures, where traditional aerodynamic controls are less effective. A numerical investigation of Fluidic Thrust Vectoring (FTV) is completed to evaluate the use of fluidic injection to manipulate flow separation and cause thrust vectoring of the primary jet thrust. The methodology presented is general and can be used to study different techniques of fluidic thrust vectoring like shock-vector control, sonic-plane skewing and counterflow methods. For validation purposes the method will focus on the dual-throat nozzle concept. Internal nozzle performances and thrust vector angles were computed for several range of nozzle pressure ratios and fluidic injection flow rate. The numerical results obtained are compared with the analogues experimental data reported in the scientific literature. The model is integrated using a finite volume discretization of the compressible URANS equations coupled with a Spalart-Allmaras turbulence model. Second order accuracy in space and time is achieved using an ENO scheme.

• Journal of Korea Multimedia Society
• /
• v.13 no.12
• /
• pp.1786-1797
• /
• 2010

Automatic understanding of contents in document image is a very hard problem due to involvement with mathematically challenging problems originated mainly from the over-determined system induced by document segmentation process. In both academic and industrial areas, there have been incessant and various efforts to improve core parts of content retrieval technologies by the means of separating out segmentation related issues using semi-structured document, e.g., invoice,. In this paper we proposed classification models for text lines on invoice document in which text lines were clustered into the five categories in accordance with their contents: purchase order header, invoice header, summary header, surcharge header, purchase items. Our investigation was concentrated on the performance of machine learning based models in aspect of linear-discriminant-analysis (LDA) and non-LDA (logic based). In the group of LDA, na$\"{\i}$ve baysian, k-nearest neighbor, and SVM were used, in the group of non LDA, decision tree, random forest, and boost were used. We described the details of feature vector construction and the selection processes of the model and the parameter including training and validation. We also presented the experimental results of comparison on training/classification error levels for the models employed.

• Smart Structures and Systems
• /
• v.22 no.2
• /
• pp.175-183
• /
• 2018

Time-series data often contain one of the most valuable pieces of information in many fields including manufacturing. Because time-series data are relatively cheap to acquire, they (e.g., vibration signals) have become a crucial part of big data even in manufacturing shop floors. Recently, deep-learning models have shown state-of-art performance for analyzing big data because of their sophisticated structures and considerable computational power. Traditional models for a machinery-monitoring system have highly relied on features selected by human experts. In addition, the representational power of such models fails as the data distribution becomes complicated. On the other hand, deep-learning models automatically select highly abstracted features during the optimization process, and their representational power is better than that of traditional neural network models. However, the applicability of deep-learning models to the field of prognostics and health management (PHM) has not been well investigated yet. This study integrates the "residual fitting" mechanism inherently embedded in the wavelet transform into the convolutional neural network deep-learning structure. As a result, the architecture combines a signal smoother and classification procedures into a single model. Validation results from rotor vibration data demonstrate that our model outperforms all other off-the-shelf feature-based models.

• Journal of The Geomorphological Association of Korea
• /
• v.23 no.3
• /
• pp.67-78
• /
• 2016

Soil carbon(C) is an essential property for characterizing soil quality. Understanding spatial patterns of soil C is particularly limited for mountain areas. This study aims to predict the spatial pattern of soil C using terrain analysis in a steep mountainous area. Specifically, model performances and prediction uncertainties were investigated based on the number of resampling repetitions. Further, important predictors for soil C were also identified. Finally, the spatial distribution of uncertainty was analyzed. A total of 91 soil samples were collected via conditioned latin hypercube sampling and a digital soil C map was developed using support vector regression which is one of the powerful machine learning methods. Results showed that there were no distinct differences of model performances depending on the number of repetitions except for 10-fold cross validation. For soil C, elevation and surface curvature were selected as important predictors by recursive feature elimination. Soil C showed higher values in higher elevation and concave slopes. The spatial pattern of soil C might possibly reflect lateral movement of water and materials along the surface configuration of the study area. The higher values of uncertainty in higher elevation and concave slopes might be related to geomorphological characteristics of the research area and the sampling design. This study is believed to provide a better understanding of the relationship between geomorphology and soil C in the mountainous ecosystem.