• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권4호
• /
• pp.1263-1274
• /
• 2021

In online competitive social networks, each user can be influenced by different competing influencers and consequently chooses different products. But their interest may change over time and may have swings between different products. The existing influence spreading models seldom take into account the time-related shifts. This paper proposes a minimum cost influence maximization algorithm based on the competitive transition probability. In the model, we set a one-dimensional vector for each node to record the probability that the node chooses each different competing influencer. In the process of propagation, the influence maximization on Competitive Linear Threshold (IMCLT) spreading model is proposed. This model does not determine by which competing influencer the node is activated, but sets different weights for all competing influencers. In the process of spreading, we select the seed nodes according to the cost function of each node, and evaluate the final influence based on the competitive transition probability. Experiments on different datasets show that the proposed minimum cost competitive influence maximization algorithm based on IMCLT spreading model has excellent performance compared with other methods, and the computational performance of the method is also reasonable.

• 한국정보시스템학회지:정보시스템연구
• /
• 제30권4호
• /
• pp.203-226
• /
• 2021

Purpose The purpose of this study is to enhance the accuracy score of personal credit scoring using the convolutional neural networks and secure the transparency of the deep learning model using eXplainalbe Artifical Inteligence(XAI) technique. Design/methodology/approach This study built a classification model by using the convolutional neural networks(CNN) and applied a methodology that is transformation of numerical data to imaged data to apply CNN on personal credit data. Then layer-wise relevance propagation(LRP) was applied to model we constructed to find what variables are more influenced to the output value. Findings According to the empirical analysis result, this study confirmed that accuracy score by model using CNN is highest among other models using logistic regression, neural networks, and support vector machines. In addition, With the LRP that is one of the technique of XAI, variables that have a great influence on calculating the output value for each observation could be found.

• Nuclear Engineering and Technology
• /
• 제53권2호
• /
• pp.532-548
• /
• 2021

Nuclear power plants contain several monitoring systems that can identify the in-vessel phenomena of a severe accident (SA). Though a lot of analysis and research is carried out on SA, right from the development of the nuclear industry, not all the possible circumstances are taken into consideration. Therefore, to improve the efficacy of the safety of nuclear power plants, additional analytical studies are needed that can directly monitor severe accident phenomena. This paper presents an interacting multiple model (IMM) based fault detection and diagnosis (FDD) approach for the identification of in-vessel phenomena to provide the accident propagation information using reactor vessel (RV) out-wall temperature distribution during severe accidents in a nuclear power plant. The estimation of wall temperature is treated as a state estimation problem where the time-varying wall temperature is estimated using IMM employing three multiple models for temperature evolution. From the estimated RV out-wall temperature and rate of temperature, the in-vessel phenomena are identified such as core meltdown, corium relocation, reactor vessel damage, reflooding, etc. We tested the proposed method with five different types of SA scenarios and the results show that the proposed method has estimated the outer wall temperature with good accuracy.

• 한국가시화정보학회지
• /
• 제20권3호
• /
• pp.49-57
• /
• 2022

The interaction of planar shock wave with rectangular water column is investigated numerically. The flow phenomenon like reflection, transmission, cavitation, recirculation of shock wave, and large negative pressure due to expansion waves was discussed qualitatively and quantitatively. The numerical simulation was performed in a shock tube with a water column, and planar shock was initiated with a pressure ratio of 10. Three cases of the water column with different thicknesses, namely 0.5D, 1D, and 2D, were installed and studied. Water naturally has a higher acoustic impedance than air and mitigates the shock wave considerably. The numerical simulations were modelled using Eulerian and Volume of fluids multiphase models. The Eulerian model assumes the water as a finite structure and can visualize the shockwave propagation inside the water column. Through the volume of fluids model, the stages of breakup of the water column and mitigation effects of water were addressed. The numerical model was validated against the experimental results. The computational results show that the installation of a water column significantly impacts the mitigation of shock wave.

• 대한토목학회논문집
• /
• 제28권6B호
• /
• pp.691-696
• /
• 2008

본 연구에서는 자유수면 흐름에 적용할 수 있는 연직방향에 대해 좌표변환된 3차원 동수압 모형을 제시하였다. 제시한 모형은 자유수면과 동수압의 해석을 위하여, 2중 예측-수정(double predictor-corrector)방법을 적용하였다. 본 연구에서는 정확한 동역학적 경계조건(자유수면에서의 압력은 0인 조건)을 적용하는 방법을 검토하였고, 이 경계조건은 기존에 개발된 모형에 미소한 수정을 통하여 적용 가능함을 보여주었다. 본 연구에서 제시한 모형과 기존 모형의 계산결과를 비교하였을 때 동역학적 경계조건의 정확한 적용이 매우 중요함을 알 수 있다.

• Structural Engineering and Mechanics
• /
• 제90권6호
• /
• pp.591-600
• /
• 2024

This study investigated the acceleration response and shape change characteristics of a gravity quay wall according to the magnitude of the applied acceleration. The quay wall was defined as a port facility damaged by the Kobe earthquake. Four experimental scenarios were established based on the inclination condition grades, considered to be a significant defect factor in the quay wall. Then, the shaking table test was conducted using scaled-down quay wall models constructed per each scenario. The ground acceleration was gradually increased from the peak ground acceleration (PGA) of 0.1 g to 0.7 g. After each ground acceleration test, acceleration installed on the wall and backfill ground and inclination on the top of the wall were measured to assess the amplification of peak response acceleration and maximum response amplitude and the change in the inclination of the quay wall. This study also analyzed the separation of the quay wall from the backfill and the crack pattern of the backfill ground according to PGA values and inclination condition grades. The result of this study shows that response acceleration could provide a reasonable prediction for the changes in the inclination of the quay wall and the crack generation and propagation on the backfill from a current inclination condition grade.

• Computers and Concrete
• /
• 제34권1호
• /
• pp.15-31
• /
• 2024

This study explores the failure mechanisms of 'I' shaped non-persistent cracks under uniaxial loads through a combination of experimental tests and numerical simulations. Concrete specimens measuring 200 mm×200 mm×50 mm were manufactured, featuring 'I' shaped non-persistent joints. The number of these joints varied from one to three, with angles set at 0, 30, 60, and 90 degrees. Twelve configurations, differing in the placement of pre-existing joints, were considered, where larger joints measured 80 mm in length and smaller cracks persisted for 20 mm with a 1 mm crack opening. Numerical models were developed for the 12 specimens, and loading in Y-axis direction was 0.05 mm/min, considering a concrete tensile strength of 5 MPa. Results reveal that crack starting was primarily influenced by the slope of joint that lacks persistence in relation to the loading direction and the number of joints. The compressive strength of the samples exhibited variations based on joint layout and failure mode. The study reveals a correlation between the failure behavior of joints and the number of induced tensile fracture, which increased with higher joint angles. Specimen strength increased with decreasing joint angles and numbers. The strength and failure processes exhibited similarities in both laboratory testing and numerical modeling methods.

• Structural Engineering and Mechanics
• /
• 제91권3호
• /
• pp.251-262
• /
• 2024

Numerical modeling using the finite element method (FEM) offers crucial insights into the mechanical behavior of prostheses, including stress and strain distribution, load transfer, and stress intensity factors. Analyzing cracking in PMMA surgical cement (polymethylmethacrylate) for total hip prostheses (THP) is essential for understanding the loosening phenomenon, as the rupture of orthopedic cement is a primary cause. By understanding various failure mechanisms, significant advancements in cemented total prostheses can be achieved. This study performed a numerical analysis using a 3D FEM model to evaluate stress levels in different THP models, aiming to model damage in the orthopedic cement used in total hip arthroplasty. Utilizing ABAQUS software, FEM, and XFEM, the damage in three types of THPs-Charnley (CMK3), Osteal (BM3), and THOMPSON was modeled under stumbling loading conditions. XFEM allowed for the consideration of crack propagation between the cement and bone, while the GEARING criterion employed a user-defined field subroutine to model damage parameters. The study's findings can contribute to improving implant fixation techniques and preventing postoperative complications in orthopedic surgery.

• Computers and Concrete
• /
• 제34권4호
• /
• pp.427-446
• /
• 2024

The I-shaped steel reinforced recycled aggregate concrete (SRRC) composite structure has the advantages of high bearing capacity and environmental protection, and the interfacial bond strength is an important theory. To this end, the I-shaped SRRC bond strength and its calculation based on artificial neural network (ANN) will be studied. Firstly, 39 push out tests of I-shaped SRRC were conducted, the load-slip curve has obvious regularity, which is divided into 4 segments by 3 regular points. Three bond strengths were defined based on these three rule points, and the approximate ranges of their values and the laws of influence of each factor on them were found. Secondly, the Elman ANN model used for the prediction of bond strength was established, and the parameters of Elman ANN predicting I-shaped SRRC bond strength were studied, and the effects of detailed parameters on the prediction results were revealed. Finally, the bond strength of SRRC was predicted using Elman and BP (back propagation) neural network models, both of which showed good prediction results. This study is a theoretical basis for the design and fine simulation of I-shaped SRRC composite structures.

• 한국정보전자통신기술학회논문지
• /
• 제12권6호
• /
• pp.629-635
• /
• 2019

여러 분야에서 사용되는 이미지 분류를 위한 딥러닝(Deep Learning) 모델은 오류 역전파 방법을 통해 미분을 구현하고 미분 값을 통해 예측 상의 오류를 학습한다. 엄청난 계산량을 향상된 계산 능력으로 해결하여, 복잡하게 설계된 모델에서도 파라미터의 전역 (혹은 국소) 최적점을 찾을 수 있다는 것이 장점이다. 하지만 정교하게 계산된 데이터를 만들어내면 이 딥러닝 모델을 '속여' 모델의 예측 정확도와 같은 성능을 저하시킬 수 있다. 이렇게 생성된 적대적 사례는 딥러닝을 저해할 수 있을 뿐 아니라, 사람의 눈으로는 쉽게 발견할 수 없도록 정교하게 계산되어 있다. 본 연구에서는 임의의 잡음 신호를 추가하는 방법을 통해 적대적으로 생성된 이미지 데이터셋을 탐지하는 방안을 제안한다. 임의의 잡음 신호를 추가하였을 때 일반적인 데이터셋은 예측 정확도가 거의 변하지 않는 반면, 적대적 데이터셋의 예측 정확도는 크게 변한다는 특성을 이용한다. 실험은 공격 기법(FGSM, Saliency Map)과 잡음 신호의 세기 수준(픽셀 최댓값 255 기준 0-19) 두 가지 변수를 독립 변수로 설정하고 임의의 잡음 신호를 추가하였을 때의 예측 정확도 차이를 종속 변수로 설정하여 시뮬레이션을 진행하였다. 각 변수별로 일반적 데이터셋과 적대적 데이터셋을 구분하는 탐지 역치를 도출하였으며, 이 탐지 역치를 통해 적대적 데이터셋을 탐지할 수 있었다.