• Journal of the Korea Safety Management & Science
• /
• v.8 no.3
• /
• pp.77-86
• /
• 2006

After conducting an investigation of a specific workplace, 372 labor workers, who are employed at a certain domestic petrochemical company valve operating and maintenance processing section, were personally questioned about specific work-related ailments. According to the results, 57% of workers who were working in the petrochemical industry have suffered from pain-related musculoskeletal disorders in, at least, more than 1 body part. Specifically, there are more than 28% of workers suffering from pain in more than two body parts. Among the total work population, 18.8 % of workers are assumed to have a high probability of being diagnosed with musculoskeletal disorder. Also, according to the RULA checklist which evaluates environmental improvement conditions of the workplace, 76.7% of workplaces among the inspected areas have high hazardous work factors that are related to musculoskeletal disorder and needs to be improved upon and observed right away or in the foreseeable future. The prior factor has a high correlation and pain rate which is due to an improper work posture. Originally, this improper work posture is caused mostly by ergonomically incorrect facility design. Furthermore, the structure of the workplace does not consider the worker's individual build.

• The Korean Journal of Applied Statistics
• /
• v.13 no.1
• /
• pp.115-131
• /
• 2000

이 논문에서는 평균-이동모형(mean-shift model)을 이상점을 위한 대립모형으로 사용하여 변량모형(random effect model)에서의 이상점 검출을 위한 베이즈인자(Bayes factor)를 제시한다. 그러나 가능한 사전 정보가 없어서 무정보사전분포(noninformative prior distribution)가 사용되어야만 할 때, 대부분의 무정보사전분포는 부적절분포(improper distribution)이기 때문에 베이즌 인자에는 사전분포로부터 나온 미지의 상수가 포함되어 잇다. 이 문제를 해결하기 위해 이 논문에서는 Berger와 Pericchi (1996)가 제시한 내재베이즈인자(the intrinsic Bayes factor;IBF)를 사용한다. 또한 이 베이즈인자를 계산상 어려움을 해결하기 위해 Verdinellidh Wasserman(1995)의 일반화 세비디지키 밀도비를 이용하여 수정하고 이것을 이용하여 이상점을 검출하는 방법을 제시한다. 마지막으로 인위적으로 이상점을 포함하고 있는 데이터를 만들고 제시된 방법으로 가상실험을 하고 또한 실제 데이터에서 제시한 방법으로 이상점을 찾아보았다.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2006.11a
• /
• pp.169-172
• /
• 2006

Most studies using state-space models have been conducted under the assumption of independently distributed noises in measurement and state equation without adequate verification of the assumption. To avoid the improper use of state-space model, testing the assumption prior to the parameter estimation of state-space model is very important. The purpose of this paper is to investigate the general relationship between parameters of state-space models and those of ARIMA processes. Under the assumption, we derive restricted parameter spaces of ARIMA(p,0,p-1) models with mutually different AR roots where $p\;{\le}\;5$. In addition, the results of ARIMA(p,0,p-1) case can be expanded to more general ARIMA models, such as ARIMA(p-1,0,p-1), ARIMA(p-1,1,p-1), ARIMA(p,0,p-2) and ARIMA(p-1,1,p-2).

• Journal of Korean Society for Quality Management
• /
• v.30 no.2
• /
• pp.118-129
• /
• 2002

A multiple test of a mean parameter, λ, in the Poisson model is considered using the Bayes factor. Under noninformative improper priors, the intrinsic Bayes factor(IBF) of Berger and Pericchi(1996) and the fractional Bayes factor(FBF) of O'Hagan(1995) called as the default or automatic Bayes factors are used to select one among three models, M$_1$: λ< $λ_0, M$_2$: λ= $λ_0, M$_3$: λ> $λ_0. Posterior probability of each competitive model is computed using the default Bayes factors. Finally, theoretical results are applied to simulated data and real data.

• Journal of the Korean Statistical Society
• /
• v.28 no.1
• /
• pp.73-92
• /
• 1999

In Bayesian model selection or testing problems, one cannot utilize standard or default noninformative priors, since these priors are typically improper and are defined only up to arbitrary constants. The resulting Bayes factors are not well defined. A recently proposed model selection criterion, the intrinsic Bayes factor overcomes such problems by using a part of the sample as a training sample to get a proper posterior and then use the posterior as the prior for the remaining observations to compute the Bayes factor. Surprisingly, such Bayes factor can also be computed directly from the full sample by some proper priors, namely intrinsic priors. The present paper explains how to derive intrinsic priors for simple tree ordered exponential means. Some numerical results are also provided to support theoretical results and compare with classical methods.

• Journal of the Korean Statistical Society
• /
• v.31 no.1
• /
• pp.63-75
• /
• 2002

Most attempts at Bayesian analysis of neural networks involve hierarchical modeling. We believe that similar results can be obtained with simpler models that require less computational effort, as long as appropriate restrictions are placed on parameters in order to ensure propriety of posterior distributions. In particular, we adopt a model first introduced by Lee (1999) that utilizes an improper prior for all parameters. Straightforward Gibbs sampling is possible, with the exception of the bias parameters, which are embedded in nonlinear sigmoidal functions. In addition to the problems posed by nonlinearity, direct sampling from the posterior distributions of the bias parameters is compounded due to the duplication of hidden nodes, which is a source of multimodality. In this regard, we focus on sampling from the marginal posterior distribution of the bias parameters with Markov chain Monte Carlo methods that combine traditional Metropolis sampling with a slice sampler described by Neal (1997, 2001). The methods are illustrated with data examples that are largely confined to the analysis of nonparametric regression models.

• Asia pacific journal of information systems
• /
• v.18 no.3
• /
• pp.45-65
• /
• 2008

The rapid growth of the Internet has increased the amount of transmission of personally identifiable information. At the same time, with new Internet related technologies, organizations are trying to collect and access more personal information than before, which in turn makes individuals concern more about their information privacy. For their successful businesses, organizations have tried to alleviate these concerns in two ways: (1) by offering privacy policies that promise certain level of privacy protection; (2) by offering benefits such as financial gains or convenience. In this paper, we interpret these actions in the context of the information processing theory of motivation. This paper follows Hann et al.(2007)'s methods to analyze Internet users privacy concerns in Korea and tries to compare the findings. Our research objectives are as follows: First, we analyze privacy concern mitigation strategies in the framework of the expectancy theory of motivation. Subsequently, we show how the expectancy theory based framework is linked o the conjoint analysis. We empirically validate the predictions that the means to mitigate privacy concerns are associated with positive valences resulting in an increase in motivational score. In order to accommodate real-life contexts, we investigate these means in trade-off situation, where an organization may only be able to offer partially complete privacy protection and/or promotions and/or convenience, While privacy protection (secondary use, improper access) are associated with positive valences, we also find that financial gains can significantly increase the individuals' motivational score of a website in Korea. One important implication of this empirical analysis is that organizations may possess means to actively manage the privacy concerns of Internet users. Our findings show that privacy policies are valued by users in Korea just as in the US or Singapore. Hence, organizations can capitalize on this, by stating their privacy policy more prominently. Also organizations would better think of strategies or means that may increase online users' willingness to provide personal information. Since financial incentives also significantly increase the individuals' motivational score of website participation, we can quantify the value of website privacy protection in terms of monetary gains. We find that Korean Internet users value the total privacy protection (protection against errors, improper access, and secondary use of personal information) as worthy as KW 25,550, which is about US 28. Having done this conjoint analysis, we next adopt cluster analysis methodology. We identify two distinct segments of Korea's internet users-privacy guardians and information sellers, and convenience seekers. The immediate implication of our study is that firms with online presence must differentiate their services to serve these distinct segments to best meet the needs of segments with differing trade-offs between money and privacy concerns. Information sellers are distinguished from privacy guardians by prior experience of information provision, To the extent that businesses cannot observe an individual's prior experience, they must use indirect methods to induce segmentation by self-selection as suggested in classic economics literature of price discrimination, Businesses could use monetary rewards to attract information sellers to provide personal information. One step forward from the societal trends that emphasize the need of legal protection of information privacy, our study wants to encourage organizations and related authorities to have the viewpoints to consider both importance of privacy protection and the necessity of information trade for the growth of e-commerce.

• The Mathematical Education
• /
• v.61 no.1
• /
• pp.1-28
• /
• 2022

The purpose of this study is to explore how students' fractional knowledge is related to their solving of proportion problems. To this end, 28 clinical interviews with four middle-grade students, each lasting about 30~50 minutes, were carried out from May 2021 to August 2021. The present study focuses on two 7th grade students who exhibited their ability to coordinate three levels of units prior to solving whole number problems. Although the students showed interiorization of three levels of units in solving whole number problems, how they coordinated three levels of units were different in solving proportion problems depending on whether the problems required reasoning with whole numbers or fractions. The students could coordinate three levels of units prior to solving the problems involving whole numbers, they coordinated three levels of units in activity for the problems involving fractions. In particular, the ways the two students employed partitioning operations and how they coordinated quantitative unit structures were different in solving proportion problems involving improper fractions. The study contributes to the field by adding empirical data corroborating the hypotheses that students' ability to transform one three levels of units structure into another one may not only be related to their interiorization of recursive partitioning operations, but it is an important foundation for their construction of splitting operations for composite units.

• Journal of radiological science and technology
• /
• v.9 no.1
• /
• pp.125-129
• /
• 1986

We quantified the relationship of fluorescence depends upon the thickness of each absorber by the result of experiment which was practiced to know the relationship of x-ray absorption and photographic effect in the range of diganostic energy about water, acrylic resin, aluminium and copper. There was no great difference in the photographic effect between water and acryl because the absorption degree of the two was mostly equal. But in case of aluminium and copper, it was opposite from the prior. Therefore when we considered the photographic effect in the range of diagnostic energy, it was thought that water and acryl were proper as a absorbers, but not only copper but also aluminium was improper as a absorber.

• Journal of the Korean Society for information Management
• /
• v.32 no.4
• /
• pp.55-76
• /
• 2015

The prior studies on the practical lawyer's information seeking behaviors, including Leckie et al. (1996) model showed that role of the work and nature of the work in attorney's practices had an significant influence on the attorney's information seeking behaviors. This study now asks if these prior analyses can be applicable to attorneys' practices nowadays. This study performed surveys and interviews with 21 practical attorneys in korea who were grouped by their experience period and the size of law firms. This study concludes that role of the work in Leckie et al. model still affects the attorneys' information seeking behaviors today and moreover, the attorney's experience and the size of law firms as variables also have made an impact on attorney's behaviors so far. By the way, this study further finds that the attorneys prefer the digital information in online database and formal information like statutes or case laws. These results are definitely different from them of the prior studies. In addition, this study suggests that the behavior such as meaningful using of the informal information in difficulties with improper information can be kinds of the attorney's information seeking behaviors.