• Journal of applied mathematics & informatics
• /
• v.26 no.1_2
• /
• pp.263-273
• /
• 2008

For the discrete version of an artificial neural network of two neurons with piecewise constant argument, we obtain some sufficient conditions that every solution is either periodic or convergent.

• Journal of Korea Society of Industrial Information Systems
• /
• v.20 no.4
• /
• pp.25-37
• /
• 2015

The early studies indicated that the firm with monopoly power is likely to engage in X-inefficiency such as a managerial slack. The reflection of the X-inefficiency theory has led to the issue that the public sector may be more inefficient than the private sector. In Korea like other many countries the electricity gas water service which can be considered as natural monopoly have been provided mostly by the public sector. In order to provide the empirical evidence to the argument that the public sector may be more inefficient than the private sector this paper estimated the four types of Solow residual which is called the total factor productivity in the electricity gas water service industry with the associated empirical model and compared its productivity with one in the manufacturing industry. The empirical results do not support the argument that the public sector may be more inefficient or less productive than the private sector.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.354-359
• /
• 2022

에세이 자동 평가는 주어진 에세이를 읽고 자동으로 평가하는 작업이다. 본 논문에서는 효과적인 에세이 자동 평가 모델을 위해 Argument Mining 작업을 사용하여 에세이의 논증 구조가 반영된 에세이 표현을 만들고, 에세이의 평가 항목별 표현을 학습하는 방법을 제안한다. 실험을 통해 제안하는 에세이 표현이 사전 학습 언어 모델로 얻은 표현보다 우수함을 입증했으며, 에세이 평가를 위해 평가 항목별로 다른 표현을 학습하는 것이 보다 효과적임을 보였다. 최종 제안 모델의 성능은 QWK 기준으로 0.543에서 0.627까지 향상되어 사람의 평가와 상당히 일치한다.

• The Journal of Information Systems
• /
• v.26 no.4
• /
• pp.107-135
• /
• 2017

Purpose Along with the growth of smart phones and social networking service (SNS), social commerce continues to expand. Although online reviews have become an important source of the information that consumers use to make purchasing decisions, theoretical development and empirical testing in this area are still limited. Thus, there is a need to develop further understanding about the influence of electronic word-of-mouth (eWOM). Drawing upon the heuristic - systematic model (HSM) which is one of the dual-process theories, this study develops a research model that explains key factors influencing consumers' eWOM credibility. Furthermore, this study verifies that consumer's eWOM credibility is a key determinant of eWOM and purchase intentions. Design/methodology/approach The proposed model is empirically tested with 493 users who have experience in WeChat-based social commerce. The structural equation model (SEM) analysis is used to evaluate the research model and hypotheses. Findings The major findings are as follows. First, argument quality of eWOM (a systematic factor) has a positive effect on eWOM credibility. Second, source credibility and recommendation consistency of eWOM (heuristic factors) are positively associated with eWOM credibility. Finally, purchase and eWOM intentions greatly depend on eWOM credibility. These results confirm the effectiveness of HSM in explaining eWOM mechanisms in SNS-based social commerce. The details of findings and implications are presented.

• Vacuum Magazine
• /
• v.1 no.1
• /
• pp.11-16
• /
• 2014

A variety of metal vacuum systems displays the celebrated 1/t pressure, namely, power-law dependence on time t, with the exponent close to unity, as to the origin of which there has been long-standing controversy. Here we propose a chemisorption model for water adsorbates, based on the argument for 2D fermion behavior of water adsorbed on a metal surface, and obtain analytically the power-law behavior of pressure with an exponent unity. Further, the model predicts that the pressure should depend on the temperature T according to $T^{1.5}$, which is indeed confirmed by our experiment.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.38.4-38
• /
• 2001

In this paper, a novel model following sliding mode control is proposed by using a novel sliding mode with virtual state. This sliding surface has nominal dynamics of an original system and makes it possible that the Sliding Mode Control(SMC) technique is combined with the optimal controller. Its design is based on the argument system whose dynamics have one higher order than that of the original system. The reaching phase is eliminated by using an initial virtual state that makes the initial sliding function equal to zero.

• Proceeding of EDISON Challenge
• /
• 2017.03a
• /
• pp.82-94
• /
• 2017

본 연구에서는 고분자의 해밀토니안을 Q-state Potts-Model로 근사하여, 2, 3차원 형태의 선형 고분자 시스템을 모사하고자 한다. 고분자 외부에서 힘이 작용하고 있는 경우와 아닌 경우에 대하여 전산모사 결과를 바탕으로 고분자의 크기의 경향성을 확인하였다. 그 결과, 흔히 사용되는 Flory's Theory와 달리 고분자의 대표적인 특성인 지수(exponent)에 대하여 더 유동적인 관계식을 얻어낼 수가 있었다. 이를 바탕으로, 고분자에서 흔히 사용되고 있는 scaling argument가 잘 적용되는지를 확인해보았으며, 그 결과 일부의 경우에서만 성립함을 확인할 수 있었다.

• Journal of The Korean Association For Science Education
• /
• v.38 no.2
• /
• pp.219-234
• /
• 2018

In this study, we explored the possibility of automating the process of analyzing elements of scientific argument in the context of a Korean classroom. To gather training data, we collected 990 sentences from science education journals that illustrate the results of coding elements of argumentation according to Toulmin's argumentation structure framework. We extracted 483 sentences as a test data set from the transcription of students' discourse in scientific argumentation activities. The words and morphemes of each argument were analyzed using the Python 'KoNLPy' package and the 'Kkma' module for Korean Natural Language Processing. After constructing the 'argument-morpheme:class' matrix for 1,473 sentences, five machine learning techniques were applied to generate predictive models relating each sentences to the element of argument with which it corresponded. The accuracy of the predictive models was investigated by comparing them with the results of pre-coding by researchers and confirming the degree of agreement. The predictive model generated by the k-nearest neighbor algorithm (KNN) demonstrated the highest degree of agreement [54.04% (${\kappa}=0.22$)] when machine learning was performed with the consideration of morpheme of each sentence. The predictive model generated by the KNN exhibited higher agreement [55.07% (${\kappa}=0.24$)] when the coding results of the previous sentence were added to the prediction process. In addition, the results indicated importance of considering context of discourse by reflecting the codes of previous sentences to the analysis. The results have significance in that, it showed the possibility of automating the analysis of students' argumentation activities in Korean language by applying machine learning.

• Journal of The Korean Association For Science Education
• /
• v.33 no.2
• /
• pp.486-500
• /
• 2013

This study is conducted to examine how epistemic reasoning and argument structures of students vary according to data sources used in the process of argumentation implemented in the context of inquiry. To this end, three argument tasks using first-hand data and three argument tasks using second-hand data were developed and applied to the unit on 'Nutrition of Plants' for first year middle school students. According to the results of this study, epistemic reasoning of students manifested during the process of argumentation and varied according to data sources. While most students composed explanations with phenomenon-based or relation-based reasoning in argumentation using first-hand data, all the small groups composed explanations that included model-based reasoning in argumentation using second-hand data. In the case of arguments including phenomenon-based or relation-based reasoning, students described only observable characteristics, with warrants omitted from arguments in many cases. On the other hand, in the case of arguments that included model-based reasoning, explanations were composed by combining the results of observations with theoretical knowledge, with warrants more apparent in their arguments.

• Journal of Astronomy and Space Sciences
• /
• v.31 no.1
• /
• pp.41-50
• /
• 2014

The Earth is not perfectly spherical and its rotational axis is not fixed in space, and these geophysical and kinematic irregularities work as dominant perturbations in satellite orbit propagation. The International Earth Rotation Service (IERS) provides the Conventions as guidelines for using the Earth's model and the reference time and coordinate systems defined by the International Astronomical Union (IAU). These guidelines are directly applied to model orbital dynamics of Earth satellites. In the present work, the effects of the latest conventions released in 2010 on orbit propagation are investigated by comparison with cases of applying the previous guidelines, IERS Conventions (2003). All seven major updates are tested, i.e., for the models of the precession/nutation, the geopotential, the ocean tides, the ocean pole tides, the free core nutation, the polar motion, and the solar system ephemeris. The resultant position differences for one week of orbit propagation range from tens of meters for the geopotential model change from EGM96 to EGM2008 to a few mm for the precession/nutation model change from IAU2000 to IAU2006. The along-track differences vary secularly while the cross-track components show periodic variation. However, the radial-track position differences are very small compared with the other components in all cases. These phenomena reflect the variation of the ascending node and the argument of latitude. The reason is that the changed models tested in the current study can be regarded as small fluctuations of the geopotential model from the point of view of orbital dynamics. The ascending node and the argument of latitude are more sensitive to the geopotential than the other elements. This study contributes to understanding of the relation between the Earth's geophysical properties and orbital motion of satellites as well as satellite-based observations.