• Structural Engineering and Mechanics
• /
• 제84권5호
• /
• pp.605-618
• /
• 2022

Glass fiber reinforced polymer (GFRP) elastic gridshells consist of long continuous GFRP tubes that form elastic deformations. In this paper, a method for the form-finding of gridshell structures is presented based on the interpretable machine learning (ML) approaches. A comparative study is conducted on several ML algorithms, including support vector regression (SVR), K-nearest neighbors (KNN), decision tree (DT), random forest (RF), AdaBoost, XGBoost, category boosting (CatBoost), and light gradient boosting machine (LightGBM). A numerical example is presented using a standard double-hump gridshell considering two characteristics of deformation as objective functions. The combination of the grid search approach and k-fold cross-validation (CV) is implemented for fine-tuning the parameters of ML models. The results of the comparative study indicate that the LightGBM model presents the highest prediction accuracy. Finally, interpretable ML approaches, including Shapely additive explanations (SHAP), partial dependence plot (PDP), and accumulated local effects (ALE), are applied to explain the predictions of the ML model since it is essential to understand the effect of various values of input parameters on objective functions. As a result of interpretability approaches, an optimum gridshell structure is obtained and new opportunities are verified for form-finding investigation of GFRP elastic gridshells during lifting construction.

• 한국생활과학회지
• /
• 제21권1호
• /
• pp.59-70
• /
• 2012

The purposes of this study was to investigate the relationship and interaction between infant temperament, development, mother's parenting stress and social support on mother's self-efficacy. Participants in this study included 1610 infants (825 boys, 785 girls) and their mothers. The major findings of this study were as follows: First, infant temperament demonstrated a direct relationship to mother's self-efficacy. Lower levels of infant temperament indicated lower levels of self-efficacy, and higher levels of infant temperament indicated higher levels of self-efficacy. Second, infant development demonstrated a direct relationship to mother's self-efficacy. Greater communicative and social interaction between mother and child demonstrated a higher level of maternal self-efficacy. Third, mothers' parenting stress demonstrated a direct relationship to mother's self-efficacy. Higher levels of parenting stress demonstrated lower levels of maternal self-efficacy. Forth, Social support demonstrated a direct relationship to mother's self-efficacy. Greater levels of social support demonstrated lower level of maternal self-efficacy. Fifth, the greatest single relationship effecting mother's self-efficacy was mother's parenting stress. This research suggests the need for development of diverse social policies and programs to help mothers reduce maternal parenting stress and support the development of positive parenting skills with the goal of boosting mother's self-efficacy.

• 한국IT서비스학회지
• /
• 제18권3호
• /
• pp.95-115
• /
• 2019

In recent years, social media that has emerged with the development of network technology has changed the channels of information dissemination. The social Q&A community is a platform for knowledge-sharing activities in a question-and-answer manner based on Web 2.0. In knowledge-sharing activities, valuable new knowledge continues to be produced and will surely bring great benefits to individuals or businesses. In the social Q&A community, the user's subjective factors play a crucial role in influencing the user's continued use and participation in knowledge-sharing activities. In order for users to actively participate in knowledge-sharing activities in the community, it needs to grasp their subjective ideas. This study explores the issue of sharing knowledge by users of the social Q&A community "Zhihu", or how to drive community revitalization efforts from these. The three factors self-efficacy, self-development motivation, and social comparison tendencies were derived, and identify their relationship with knowledge-sharing activities and community-boosting efforts through empirical analysis. In addition, the influence of knowledge acquisition on knowledge provision was investigated through sense of reciprocity. Implications of the study findings and the future research directions were also discussed.

• 대한치위생과학회지
• /
• 제2권2호
• /
• pp.1-11
• /
• 2019

Objectives: The purpose of this study was to examine the relationship between the self-efficacy of dental hygiene students and their satisfaction with clinical practice to provide information for developing programs aimed at instilling self-efficacy and boosting satisfaction with clinical practice. Methods: A self-reported survey was conducted with dental hygiene students in the metropolitan area. Of those surveys distributed, 243 questionnaires were analyzed. Results: The results of this study are as follows: 1. Regarding self-efficacy, the subjects had a mean result of 2.96(out of five points), and 3.09 points in satisfaction with clinical practice. 2. As for the subfactors of self-efficacy, the students who got higher grades, whose awareness of dental hygienists was better and whose awareness of dental hygienists after on-site clinical practice was better scored higher in terms of confidence and self-regulation. In task difficulty preference, the students who were more satisfied with majoring in dental hygiene and whose awareness of dental hygienists was better scored higher. In terms of motivation for choosing the dental hygiene department, the students scored higher when the department was their preferred option. 3. In satisfaction with clinical practice, the students who were aware of dental hygienists at the time of college entrance was better. Those whose awareness of dental hygienists after experiencing clinical practice was better and who were more satisfied with majoring in dental hygiene expressed more satisfaction with clinical practice. 4. Satisfaction with clinical practice was higher when self-efficacy was better. Conclusions: Based on the above results, it is thought that developing programs that can enhance self-efficacy and include on-site clinical practice would be beneficial as higher self-efficacy levels were related to higher clinical practice satisfaction.

• 디지털융복합연구
• /
• 제14권11호
• /
• pp.417-425
• /
• 2016

본 연구는 학령기 아동을 대상으로 아동이 지각한 사회적지지, 자기효능감이 스트레스 대처행동에 미치는 영향을 파악하고자 연구되었다. 자료수집기간은 2014년 5월 12일부터 5월 23일까지 였으며 D시에 있는 초등학생 312명을 대상으로 하였다. 수집된 자료는 SPSS를 이용하여 실수와 백분율, 평균과 표준편차, t-test, ANOVA, 상관관계, 다중회귀분석을 하였다. 연구분석결과 아동이 지각하는 사회적지지, 자기효능감은 긍정적 스트레스 대처행동과 유의한 정적 상관관계를 나타내었다. 또한 아동의 긍정적인 스트레스 대처행동에 영향을 미치는 예측요인으로는 사회적지지, 자기효능감이 유의한 영향력을 나타내었다. 이러한 결과를 토대로 아동의 긍정적 스트레스 대처행동에 사회적지지, 자기효능감이 많은 영향을 주고 있음을 확인하였으며, 추후 학령기 아동의 스트레스 감소에 영향을 줄 수 있는 사회적지지와 자기효능감 향상 프로그램의 개발과 중재가 필요하다.

• 한국학교보건학회지
• /
• 제19권2호
• /
• pp.13-24
• /
• 2006

Purpose : The purpose of this study was to develop an obese elementary students education program available for blended learning environments, which refered to a combination of ofline and online learning. And it's also intended to verify its efect to bost the eficiency of obese schol children's learning about obesity.Methods : The subjects were 52(experimental group=21, control group=31 ) fourth, fifth and sixth graders who OVA. Results : The obese elementary students education program tailored to blended learning setting wasn't more conducive to lowering the obesity level of the obese students t han the typical obese elementary students education program. But the obese elementary students education was more efective than the typical one in improving the knowle dge of the obese students about obesity, and in enhancing the academic self-eficacy of the children. Conclusion : As the obese elementary students education program tailored to blended learning setting was effective in boosting the relevant knowledge and academic self-efficacy of the children, sustained research efforts should be put into developing obese elementary students education programs available for both online and offline to help students to access information on obesity and manage it better.

• Computers and Concrete
• /
• 제32권4호
• /
• pp.373-381
• /
• 2023

Fly ash, granulated blast furnace slag, marble waste powder, etc. are just some of the by-products of other sectors that the construction industry is looking to include into the many types of concrete they produce. This research seeks to use surrogate machine learning methods to forecast the compressive strength of self-compacting concrete. The surrogate models were developed using Gradient Boosting Machine (GBM), Support Vector Machine (SVM), Random Forest (RF), and Gaussian Process Regression (GPR) techniques. Compressive strength is used as the output variable, with nano silica content, cement content, coarse aggregate content, fine aggregate content, superplasticizer, curing duration, and water-binder ratio as input variables. Of the four models, GBM had the highest accuracy in determining the compressive strength of SCC. The concrete's compressive strength is worst predicted by GPR. Compressive strength of SCC with nano silica is found to be most affected by curing time and least by fine aggregate.

• Journal of Power Electronics
• /
• 제18권4호
• /
• pp.1051-1066
• /
• 2018

This study presents a novel step-up switched capacitor multilevel inverter (SCMLI) structure. The proposed structure comprises 2 unequal DC voltage sources, 4 capacitors, and 14 unidirectional power switches. It can synthesize 21 output voltage levels. The important features of the proposed topology are its self-voltage boosting and inherent capacitor voltage balancing capabilities. Furthermore, a cascaded structure of the proposed SCMLI with an asymmetric DC voltage source configuration is presented. The proposed topology and its cascaded structure are compared with conventional and other recently developed topologies in terms of different aspects, such as the required components to produce a specific number of output voltage levels, the total standing voltage (TSV) and peak inverse voltage of the structure, and the maximum number of switches in the conducting path. Furthermore, a cost function is developed to verify the cost-effectiveness of the proposed topology with respect to other topologies. The TSV of the proposed topology is significantly lower than those of other topologies. Moreover, the developed topology is cost-effective compared with other topologies. A detailed operating principle, power loss analysis, and selection procedure for switched capacitors are presented for the proposed SCMLI structure. Extensive simulation and experimental studies of a 21-level inverter structure prove the effectiveness and merits of the proposed SCMLI.

• 한국컴퓨터정보학회논문지
• /
• 제27권10호
• /
• pp.19-28
• /
• 2022

본 논문에서는 실시간 고성능 다중 객체 추적을 수행하기 위해 최적의 TBD (Tracking-by-detection) 메커니즘을 결정할 수 있는 Detector Scheduler를 제안한다. Detector Scheduler는 서로 다른 프레임 간의 특징량 차이를 측정하는 것으로 검출기 실행 여부를 결정하여 전체 추적 속도를 향상한다. 하지만, Detector Scheduler의 학습에 필요한 GT (Ground Truth) 생성이 어렵기 때문에 Detector Scheduler를 추적 결과만을 통해 학습 가능한 자가 학습 방법을 제안한다. 제안된 자가 학습 방법은 프레임 간의 객체 카디널리티와 객체 외형 특징량의 비유사도가 커질 때 검출기를 실행할 수 있도록 의사 레이블을 생성하고 제안된 손실함수를 통해 Detector Scheduler를 학습한다.

• 스마트미디어저널
• /
• 제13권4호
• /
• pp.9-15
• /
• 2024

Tree-based algorithms have been the dominant methods used build a prediction model for tabular data. This also includes personal credit data. However, they are limited to compatibility with categorical and numerical data only, and also do not capture information of the relationship between other features. In this work, we proposed an ensemble model using the Transformer architecture that includes text features and harness the self-attention mechanism to tackle the feature relationships limitation. We describe a text formatter module, that converts the original tabular data into sentence data that is fed into FinBERT along with other text features. Furthermore, we employed FT-Transformer that train with the original tabular data. We evaluate this multi-modal approach with two popular tree-based algorithms known as, Random Forest and Extreme Gradient Boosting, XGBoost and TabTransformer. Our proposed method shows superior Default Recall, F1 score and AUC results across two public data sets. Our results are significant for financial institutions to reduce the risk of financial loss regarding defaulters.