• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.6
• /
• pp.592-599
• /
• 2015

In big data environment, we need new approach for big data analysis, because the characteristics of big data, such as volume, variety, and velocity, can analyze entire data for inferring population. But traditional methods of statistics were focused on small data called random sample extracted from population. So, the classical analyses based on statistics are not suitable to big data analysis. To solve this problem, we propose an approach to efficient big data analysis. In this paper, we consider a big data analysis using principal component analysis, which is popular method in multivariate statistics. To verify the performance of our research, we carry out diverse simulation studies.

• Journal of the Ergonomics Society of Korea
• /
• v.31 no.6
• /
• pp.749-756
• /
• 2012

Objective: This study was performed to investigate the prevalence of musculoskeletal disorders of traditional Korean instrument player using a variety of traditional classical instruments: Gayageum, Geomungo, Ajaeng, Haegeum, Daegeum, Piri and Samul instruments. Background: A large percentage of instrument players have suffered from the musculoskeletal pain of each body parts. However, there is no research on the prevalence of musculoskeletal disorders of traditional Korean musical instrument players. Method: Through the focus group interview, a questionnaire to investigate musculoskeletal disorders was developed. The questionnaire consisted of four parts: demographic factors, performance factors, musculoskeletal disorders symptoms, musculoskeletal disorder experience. For the survey, 118 expert players participated. The data from the survey were analyzed by correlation analysis and chi-square analysis. Results: The symptoms of musculoskeletal disorders and the severe pain from musculoskeletal disorders were observed at neck, shoulder, back and knee. The musculoskeletal experience was statistically related to the factor of body height in Gayageum and Geomungo. In addition, the musculoskeletal experience in Geomungo was related to age and career. However, the musculoskeletal experience in Ajaeng and Haegeum was only related to the factor of hobby. The musculoskeletal experience in Daegeum and Piri was related to stretching. In addition, there was a statistical significance between the musculoskeletal experience and sex in Daegum. In Samul instruments, the statistical significance was observed at age, BMI, career and stretching. Conclusion: The symptoms of playing-related musculoskeletal disorders of traditional Korean musical players were prevalently observed at neck, shoulder, back and knee. In addition, these symptoms were related to the various demographic factors such as age, body height, BMI, career, sex, hobby and stretching. Application: The results of this study can be used as the preliminary data for preventing the musculoskeletal injuries of traditional Korean musical instrument players.

• Journal of the Korean Data and Information Science Society
• /
• v.12 no.2
• /
• pp.1-10
• /
• 2001

Neural networks we increasingly being seen as an addition to the statistics toolkit which should be considered alongside both classical and modern statistical methods. Neural networks are usually useful for classification and function estimation. In this paper we concentrate on function estimation using neural networks with weight decay factor The use of weight decay seems both to help the optimization process and to avoid overfitting. In this type of neural networks, the problem to decide the number of hidden nodes, weight decay parameter and iteration number of learning is very important. It is called the optimization of weight decay neural networks. In this paper we propose a automatic optimization based on genetic algorithms. Moreover, we compare the weight decay neural network automatically learned according to automatic optimization with ordinary neural network, projection pursuit regression and support vector machines.

• The Journal of Korean Medicine
• /
• v.42 no.4
• /
• pp.120-132
• /
• 2021

Objectives: The prevalence of Parkinson's disease is on an upward trend along with an increase in the aging population but there is no available treatment that halts the progression of neurodegeneration. This study reports a numerical analysis on Donguibogam and suggests novel herbal drugs, which have never been researched before but found to be deemed effective in this study. Methods: Referring to 71 Korean medicine symptom terms that represent the symptoms of Parkinson's disease, 4170 prescriptions described in Donguibogam were classified into two groups based on whether their main effects were effective for Parkinson's disease or not. Comparing the two groups, the chi-square test was performed to select statistically significant herbs, while the t-test, Wilcoxon test, and descriptive statistics were performed to determine the appropriate dose. Results: One hundred and twenty-seven prescriptions effective for Parkinson's disease were identified. The chi-square test determined 17 herbs that are effective for symptomatic treatment. Among the medicinal herbs, the authors suggest Osterici seu Notopterygii Radix et Rhizoma, Ephedrae Herba, Aconiti Tuber, Myrrha, Sinomeni Caulis et Rhizoma, and Aconiti Kusnezoffii Tuber as herbal candidates that have never been studied for Parkinson's disease. Through the statistical tests, it was judged that the mean value of the dose of the entire prescription was the appropriate dose for each herb. Conclusions: Seventeen herbs were selected for Parkinson's disease and the appropriate daily dose were calculated. Furthermore, this study presented a new process that applies a statistical method to traditional medical literature and preselecting herbs deemed effective for specific diseases.

• The Korean Journal of Applied Statistics
• /
• v.35 no.2
• /
• pp.217-227
• /
• 2022

Quantile regression is widely used in many fields based on the advantage of providing an efficient tool for examining complex information latent in variables. However, modern large-scale and high-dimensional data makes it very difficult to estimate the quantile regression model due to limitations in terms of computation time and storage space. Divide-and-conquer is a technique that divide the entire data into several sub-datasets that are easy to calculate and then reconstruct the estimates of the entire data using only the summary statistics in each sub-datasets. In this paper, we studied on a variable selection method using Bayes information criteria by applying the divide-and-conquer technique to the penalized quantile regression. When the number of sub-datasets is properly selected, the proposed method is efficient in terms of computational speed, providing consistent results in terms of variable selection as long as classical quantile regression estimates calculated with the entire data. The advantages of the proposed method were confirmed through simulation data and real data analysis.

• The Korean Journal of Applied Statistics
• /
• v.32 no.4
• /
• pp.607-618
• /
• 2019

A new and interesting task in statistics is to effectively analyze functional data that frequently comes from advances in modern science and technology in areas such as meteorology and biomedical sciences. Functional linear regression with scalar response is a popular functional data analysis technique and it is often a common problem to determine a functional association if a functional predictor variable affects the scalar response in the models. Recently, Kong et al. (Journal of Nonparametric Statistics, 28, 813-838, 2016) established classical testing methods for this based on functional principal component analysis (of the functional predictor), that is, the resulting eigenfunctions (as a basis). However, the eigenbasis functions are not generally suitable for regression purpose because they are only concerned with the variability of the functional predictor, not the functional association of interest in testing problems. Additionally, eigenfunctions are to be estimated from data so that estimation errors might be involved in the performance of testing procedures. To circumvent these issues, we propose a testing method based on fixed basis such as B-splines and show that it works well via simulations. It is also illustrated via simulated and real data examples that the proposed testing method provides more effective and intuitive results due to the localization properties of B-splines.

• The Journal of Advanced Prosthodontics
• /
• v.6 no.6
• /
• pp.427-433
• /
• 2014

PURPOSE. Surface color is one of the main criteria to obtain an ideal esthetic. Many factors such as the type of the material, surface specifications, number of firings, firing temperature and thickness of the porcelain are all important to provide an unchanged surface color in dental ceramics. The aim of this study was to evaluate the color changes in dental ceramics according to the material type and glazing methods, during the multiple firings. MATERIALS AND METHODS. Three different types of dental ceramics (IPS Classical metal ceramic, Empress Esthetic and Empress 2 ceramics) were used in the study. Porcelains were evaluated under five main groups according to glaze and natural glaze methods. Color changes (${\Delta}E$) and changes in color parameters (${\Delta}L$, ${\Delta}a$, ${\Delta}b$) were determined using colorimeter during the control, the first, third, fifth, and seventh firings. The statistical analysis of the results was performed using ANOVA and Tukey test. RESULTS. The color changes which occurred upon material-method-firing interaction were statistically significant (P<.05). ${\Delta}E$, ${\Delta}L$, ${\Delta}a$ and ${\Delta}b$ values also demonstrated a negative trend. The MC-G group was less affected in terms of color changes compared to other groups. In all-ceramic specimens, the surface color was significantly affected by multiple firings. CONCLUSION. Firing detrimentally affected the structure of the porcelain surface and hence caused fading of the color and prominence of yellow and red characters. Compressible all-ceramics were remarkably affected by repeated firings due to their crystalline structure.

• The Korean Journal of Applied Statistics
• /
• v.24 no.2
• /
• pp.269-278
• /
• 2011

There are many researches that have considered the distribution functions and appropriate covariates corresponding to the scores in order to improve the accuracy of a diagnostic test, including the ROC curve that is represented with the relations of the sensitivity and the specificity. The ROC analysis was used by the regression model including some covariates under the assumptions that its distribution function is known or estimable. In this work, we consider a general situation that both the distribution function and the elects of covariates are unknown. For the ROC analysis, the mixtures of normal distributions are used to estimate the distribution function fitted to the credit evaluation data that is consisted of the score random variable and two sub-populations of parameters. The AUC measure is explored to compare with the nonparametric and empirical ROC curve. We conclude that the method using normal mixtures is fitted to the classical one better than other methods.

• Wind and Structures
• /
• v.26 no.1
• /
• pp.11-24
• /
• 2018

Full-scale measurements of wind action on the open roof structure of the WuXi grand theater, which is composed of eight large-span free-form leaf-shaped space trusses with the largest span of 76.79 m, were conducted during the passage of Typhoons HaiKui and SuLi. The wind pressure field data were continuously and simultaneously monitored using a wind pressure monitoring system installed on the roof structure during the typhoons. A detailed analysis of the field data was performed to investigate the characteristics of the fluctuating wind pressure on the open roof, such as the wind pressure spectrum, spatial correlation coefficients, peak wind pressures and non-Gaussian wind pressure characteristics, under typhoon conditions. Three classical methods were used to calculate the peak factors of the wind pressure on the open roof, and the suggested design method and peak factors were given. The non-Gaussianity of the wind pressure was discussed in terms of the third and fourth statistical moments of the measured wind pressure, and the corresponding indication of the non-Gaussianity on the open roof was proposed. The result shows that there were large pulses in the time-histories of the measured wind pressure on Roof A2 in the field. The spatial correlation of the wind pressures on roof A2 between the upper surface and lower surface is very weak. When the skewness is larger than 0.3 and the kurtosis is larger than 3.7, the wind pressure time series on roof A2 can be taken as a non-Gaussian distribution, and the other series can be taken as a Gaussian distribution.

• Korean Journal of Head & Neck Oncology
• /
• v.8 no.1
• /
• pp.14-20
• /
• 1992

To clarify the safety of both total and near-total thyroidectomy, and to guide a selectionof an adequate type of surgical treatment of thyroid diseases, 192 consecutive total or near-total thyroidectomy cases were reviewed. They were divided into two groups: ont, the total thyroidectomy group(Group T,N=111) and the other, the near-total thyroidectomy group (Group NT, N=81). In both groups, complication rates, associations of complication rates with extents of surgery and stage of lesion were observed. Complication rate was significantly higher in Group T (53.6% vs 12.3%, p<0.05). But the rate of permanent complications such as permanent hypoparathyroidism and recurrent laryngeal nerve injury was remarkably low(4.5% in Group T, 6.0% in Group NT) and shows no significant difference in both groups. There was no permanent complication in cases where any type of neck dissection had not been performed regardless of the type thyroidectomy. But among whom underwent central compartmental neck dissection(CCND) and functional neck dissection(FND), 4(4.4%) and 4(6.4%) cases showed permanent complications. There was no statistical significance in differences between Group I and NT. In cases who underwent concomittant classical radical neck dissection(RND), 3(25.5%) showed permament complications. In this subgroups, complications were significantly higher in Group T(p<0.005). Complications were also directly related to the stage of the lesion. Only one patient showed permanent complication in 74 intracapsular lesions but 9 permanent complications were observed in 118 advanced lesions. We could clarify both total and near-total thyroidectomy were safe operations and the complications were related to accompanying neck dissections and the disease status rather than total or near-total thyroidectomy itself. Thus, we think that for the cases where higher complication rates are expected, such as locally advanced thryoid cancers or the cases which required wider neck dissection, the near-total thyroidectomy would be a preferrable method.