• Journal of the Earthquake Engineering Society of Korea
• /
• v.8 no.1
• /
• pp.59-65
• /
• 2004

Recently, as steel structures become higher and more long-spanned, application of high strength steel. SM570, is considered, For accurate seismic design, seismic analysis of steel structures needs a constitutive equation describing the characteristic of steel materials under non-proportional cyclic loading, While the use of SM570 material is much increased these days, research for description and generalization of cyclic plasticity behavior are insufficient, In this study, a cyclic plasticity model is proposed by results of material tests, i.e, monotonic and low cycle tests, Proposed cyclic plasticity model is applied to 3-Dimensional FE program and we carried out seismic analysis of pipe-section steel pier using SM570, Comparison between experiment and analysis results shows that the proposed constitutive equation is able to describe exactly the complicated plastic behavior of steel structure using SM570.

• Journal of the Korean earth science society
• /
• v.26 no.1
• /
• pp.9-29
• /
• 2005

By analyzing and integrating established geological field study instances, this study offered a new geological field excursion teaching model with several steps: unifying question raising. excursion generalization. intensive field direction, searching, primary conclusion, re-searching, group discussion. adjustment, and excursion summarizing. Then by Qualitatively assaying the responses which students showed after applying this teaching model, a concrete teaching plan was sought for earth science teachers who were planning to begin geological field excursion classes. Students evaluated very highly on the unifying question and excursion generalization because these items provided a sense of direction and an overall theme for geological excursion in advance. Also. since the students had little to none geological knowledge and field excursion experience, the intensive field direction gave them a lot of help with their field excursion activities. Students thought that coming up with a primary conclusion based on the summary of what they had observed in their activities was original. and highly valued the process of sharing different opinions in group discussions and drawing out a final conclusion. Teachers should help students develop a friendly atmosphere, by organizing group activities and continuously feedlng them with uniting questions and excursion generalization within the groups. Also they should prepare enough contents for intensive field direction and ways to get their points across. In the process, they should arrange beforehand detailed instructions for every outcome, with the intention of solving the question. Furthermore. teachers should follow carefully how conclusions are drawn. instruct students not to reach conclusions based on mere assumptions, and be aware of misconceptions students have toward geological phenomenon in advance, so that the discussion can be lead in the right direction.

• Electronics and Telecommunications Trends
• /
• v.35 no.3
• /
• pp.9-19
• /
• 2020

Recently, a technique for applying a deep learning language model pretrained from a large corpus to fine-tuning for each application task has been widely used as a language processing technology. The pretrained language model shows higher performance and satisfactory generalization performance than existing methods. This paper introduces the major research trends related to deep learning pretrained language models in the field of language processing. We describe in detail the motivations, models, learning methods, and results of the BERT language model that had significant influence on subsequent studies. Subsequently, we introduce the results of language model studies after BERT, focusing on SpanBERT, RoBERTa, ALBERT, BART, and ELECTRA. Finally, we introduce the KorBERT pretrained language model, which shows satisfactory performance in Korean language. In addition, we introduce techniques on how to apply the pretrained language model to Korean (agglutinative) language, which consists of a combination of content and functional morphemes, unlike English (refractive) language whose endings change depending on the application.

• The KSFM Journal of Fluid Machinery
• /
• v.3 no.3 s.8
• /
• pp.19-24
• /
• 2000

It is well known that the finite element analysis often has an inaccuracy when it is in conflict with test results. Model updating is concerned with the correction of analytical model by processing records of response from test results. The famous one of the model updating methods is FRF sensitivity method. However, it has demerit that the solution is not unique. So, the neural network is recommended when an unique and exact solution is desired. The generalization ability of radial basis function neural network is used in model updating. As an application model, a cantilever and a rotor system are used. Specially the machined clearance($C_p$) of a journal bearing is updated.

• Journal of Information Technology Applications and Management
• /
• v.23 no.1
• /
• pp.45-59
• /
• 2016

An ensemble classifier is a method that combines output of multiple classifiers. It has been widely accepted that ensemble classifiers can improve the prediction accuracy. Recently, ensemble techniques have been successfully applied to the bankruptcy prediction. Bagging and random subspace are the most popular ensemble techniques. Bagging and random subspace have proved to be very effective in improving the generalization ability respectively. However, there are few studies which have focused on the integration of bagging and random subspace. In this study, we proposed a new hybrid ensemble model to integrate bagging and random subspace method using genetic algorithm for improving the performance of the model. The proposed model is applied to the bankruptcy prediction for Korean companies and compared with other models in this study. The experimental results showed that the proposed model performs better than the other models such as the single classifier, the original ensemble model and the simple hybrid model.

• The Korean Journal of Health Service Management
• /
• v.11 no.2
• /
• pp.129-142
• /
• 2017

Objectives : The purpose of this study was to identify the main variables of difference in high school students' oral health promotion behaviors among adolescents and to improve their academic and oral health promotion behaviors. Methods : The research subjects consisted of 311 high school students in Jeonju. Results : The adequacy of the hypothetical model accounted for 46.9 % of the oral health promotion behavior. The Redundancy of all variables showed the value of the positive values, indicating that the Goodness of fit was greater than the optimum value of the model, and the model of the PLS was a desirable model. The effects of perceived benefits, self efficacy, and social support on oral health promotion behaviors were found to be higher in oral health promotion behaviors. Conclusions : This study is expected to have a significant impact on the perception of the oral health promotion for adolescents in the future and will contribute to the expansion and generalization of Pender's oral health promotion model.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.24 no.6
• /
• pp.538-544
• /
• 2012

Analysis of thermal performance is required for the economic operation of turbine cycle of power plant. We developed corrective model of main feed water flow which is the most important parameter for the precise analysis of turbine cycle performance. Classification model for the identification of feed water flow measurement status was applied to increase the suitability of the corrective model. We used neural network and support vector machine to develop estimation model of main feed water flow with more generalization capability. The estimation model can be used practically to evaluate corrective performance of turbine cycle plant.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2669-2671
• /
• 2001

In this paper, we propose design a Multi-Fuzzy Inference model structure. In order to determine structure of the proposed Multi-Fuzzy Inference model, HCM clustering method is used. The parameters of membership function of the Multi-Fuzzy are identified by genetic algorithms. A aggregate performance index with a weighting factor is used to achieve a sound balance between approximation and generalization abilities of the model. We use simplified inference and linear inference as inference method of the proposed Multi-Fuzzy model and the standard least square method for estimating consequence parameters of the Multi-Fuzzy. Finally, we use some of numerical data to evaluate the proposed Multi-Fuzzy model and discuss about the usefulness.

• Bulletin of the Korean Chemical Society
• /
• v.14 no.4
• /
• pp.485-487
• /
• 1993

The critical behavior of the classical D-dimensional spin model (D${\ge}$2), which is intermediate model that link up the Ising (D = 1) and the spherical model (D = ${\infty}$), is studied for the case of constant coupling interaction independent of the spin-spin distance (Curie-Weiss model). Analytical results show that the critical behavior of the present model is in quantitative agreement with the prediction of the phenomenological mean-field theory independent of D. Critical temperature is calculated to be T$_c$=k/JD. This gives a quantitative explanation of the relationship between the spin degree of freedom and the critical temperature.

• Proceedings of the KIPE Conference
• /
• 2019.07a
• /
• pp.459-460
• /
• 2019

The performance dynamics of battery is very sensitive to operating conditions (i.e temperature, load current, and state of charge). A model developed based on certain conditions may perform well under the similar conditions but can not accurately predict the performance for changing conditions. Thus, a generalized model is needed which can accurately emulate the battery dynamic behavior under all conditions. In addition, the components of the model should relate to the physicochemical processes that occur inside the battery. Electrochemical impedance curve shows better visible reflection of the processes inside battery as compared to voltage curve. The model trained for parameterization using neural network has better generalization than simple curve fitting. Thus, this study proposes recurrent neural network based parameterization of the Lithium ion battery model followed by impedance based identification.