• Korean Journal of Occupational Health Nursing
• /
• v.4
• /
• pp.58-83
• /
• 1995

The small scale industries which have less than 30 employees occupy 86.5% of total number of industries in Korea. And though they have higher accident rate and lower environmental condition than big industries, it has been not mandatory to appointing health care manager at factory. So, from 1993, government subsidizes to the health care management of small industries. The purpose of this study is to identify the real feature of health care status in small industries, and to evaluate the level of health care management, before and after the subsidiary program. 65 small plating industries which have been managed by the same health care management support institution in 1993 were selected for study. Of the 65 industries, 3 which have not taken both environmental evaluation and health screening in 1994, and 9 which have closed were excluded from study sample. And the remaining 53 were analyzed by using the results of environmental evaluation and health screening, reported to the Ministry of Labor, before and after the subsidiary program, the analysis was done by the comparison of the two year paired data of the same industry. Over-permissible-limit rate, health screening implementation rate, above grade C rate were calculated and compared. The status of health care management ; 1. Of the sample industries, 96.9% provide protective equipment and 80.0% set up ventilating system. Protective gloves (89.2%) and protective clothing (80.0%) are widely provided, but ear plugs (4.6%) are rarely provided. 21.5% of the protective equipment are well put on, and 40.4% of the ventilating systems function well. 2. In 1993, 35 industries, 53.8% of the sample, checked working environment twice. Over-permissible-limit rates of heavy metal (12.2%), suspended particle (11.1%), noise (5.5%) were high. To put on protective equipment and to set up local ventilating system were pointed out by the examiners. 3. General health screening was done at 63.1% of the sample industries and 35.3% of total workers were examined. Specific health screening was done at 93.8% of the sample industries and 75.4% of workers were examined. 15.5% of workers was provided to be above grade C and to have digestive system disease (43.3%), circulatory disease (18.9%), and hematopoietic disease (14.2%), etc. 4. In 1993, the subsidiary program of health care management was provided in forms of health education, health counseling, and rounding check of working field. And 61.5%, 83.0%, 55.4% of sample industries respectively received it. The average visit per industry was 1.8. Comparisons of the level of occupational health before and after the subsidiary program ; 1. Over-permissible-limit rates of hazardous factors of 1993 and that of 1994 were compared. The rates of suspended particle, noise, organic solvent of 1994 (37.5%, 13.4%, 24.2% respectively) were higher than that of 1993 (25.0%, 6.0%, 6.3% respectively). In the case of acid, there was no difference between the rate of 1993 and that of 1994. Only the rate of heavy metal decreased from 12.9% in 1993 to 3.0% in 1994. 2. General health screening was done at 38.7% of the sample industries in 1993 and at 44.6% in 1994. But the implementation rate of specific health screening decreased from 72.4% in 1993 to 64.6% in 1994. 3. The implementation rate of specific health screening was analyzed by some health factors. The rate of suspended particle increased from 61.8% in 1993 to 91.2% in 1994. But the rates of the others-noise, organic solvent, heavy metal, specific chemical substances-decreased. 4. Above grade C rate in health screening increased from 27.8% in 1993 to 35.5% in 1994. But that of endocrine disorders and pulmonary disease decreased.

• Journal of Intelligence and Information Systems
• /
• v.16 no.4
• /
• pp.99-112
• /
• 2010

Ensemble learning is a method for improving the performance of classification and prediction algorithms. It is a method for finding a highly accurateclassifier on the training set by constructing and combining an ensemble of weak classifiers, each of which needs only to be moderately accurate on the training set. Ensemble learning has received considerable attention from machine learning and artificial intelligence fields because of its remarkable performance improvement and flexible integration with the traditional learning algorithms such as decision tree (DT), neural networks (NN), and SVM, etc. In those researches, all of DT ensemble studies have demonstrated impressive improvements in the generalization behavior of DT, while NN and SVM ensemble studies have not shown remarkable performance as shown in DT ensembles. Recently, several works have reported that the performance of ensemble can be degraded where multiple classifiers of an ensemble are highly correlated with, and thereby result in multicollinearity problem, which leads to performance degradation of the ensemble. They have also proposed the differentiated learning strategies to cope with performance degradation problem. Hansen and Salamon (1990) insisted that it is necessary and sufficient for the performance enhancement of an ensemble that the ensemble should contain diverse classifiers. Breiman (1996) explored that ensemble learning can increase the performance of unstable learning algorithms, but does not show remarkable performance improvement on stable learning algorithms. Unstable learning algorithms such as decision tree learners are sensitive to the change of the training data, and thus small changes in the training data can yield large changes in the generated classifiers. Therefore, ensemble with unstable learning algorithms can guarantee some diversity among the classifiers. To the contrary, stable learning algorithms such as NN and SVM generate similar classifiers in spite of small changes of the training data, and thus the correlation among the resulting classifiers is very high. This high correlation results in multicollinearity problem, which leads to performance degradation of the ensemble. Kim,s work (2009) showedthe performance comparison in bankruptcy prediction on Korea firms using tradition prediction algorithms such as NN, DT, and SVM. It reports that stable learning algorithms such as NN and SVM have higher predictability than the unstable DT. Meanwhile, with respect to their ensemble learning, DT ensemble shows the more improved performance than NN and SVM ensemble. Further analysis with variance inflation factor (VIF) analysis empirically proves that performance degradation of ensemble is due to multicollinearity problem. It also proposes that optimization of ensemble is needed to cope with such a problem. This paper proposes a hybrid system for coverage optimization of NN ensemble (CO-NN) in order to improve the performance of NN ensemble. Coverage optimization is a technique of choosing a sub-ensemble from an original ensemble to guarantee the diversity of classifiers in coverage optimization process. CO-NN uses GA which has been widely used for various optimization problems to deal with the coverage optimization problem. The GA chromosomes for the coverage optimization are encoded into binary strings, each bit of which indicates individual classifier. The fitness function is defined as maximization of error reduction and a constraint of variance inflation factor (VIF), which is one of the generally used methods to measure multicollinearity, is added to insure the diversity of classifiers by removing high correlation among the classifiers. We use Microsoft Excel and the GAs software package called Evolver. Experiments on company failure prediction have shown that CO-NN is effectively applied in the stable performance enhancement of NNensembles through the choice of classifiers by considering the correlations of the ensemble. The classifiers which have the potential multicollinearity problem are removed by the coverage optimization process of CO-NN and thereby CO-NN has shown higher performance than a single NN classifier and NN ensemble at 1% significance level, and DT ensemble at 5% significance level. However, there remain further research issues. First, decision optimization process to find optimal combination function should be considered in further research. Secondly, various learning strategies to deal with data noise should be introduced in more advanced further researches in the future.

• Journal of the Korean Society for Library and Information Science
• /
• v.22
• /
• pp.109-141
• /
• 1992

University library must have a substantial collection development policy, by which the short and long term plannings are set up to meet the goals of university education and to support university function. Only when university library organizes collection building based on the well-planned policy, can it establish the systematic collection building, which can satisfy users, support continuity of work, use budget efficiently, and solve the space problem by with-drawing the materials. But most Korean university libraries, contrary to many foreign cases, don't have proper systematic collection development policy or guideline that controls overall library work from planning the collection of materials to evaluating and preserving them. Questionaire on whether university library has collection development policy or guideline was answered positively by only 6 cases out of 57 cases, which is just 10.5 percent. Even in cases that they have a guideline, the content of the copied guideline they sent was confined to acquisition, which is just a part of collection development. Collection development is a statement needed to set up long and short term plannings in consideration of the aims of the library and the needs of users and to shape and manage the library collection systematically. With the above conception and definition, we try to make a proper collection development policy and guideline for the Korean university libraries. Here we must define the object of the university. Object of university is to have good curriculum, good courses, research activities on the part of the faculty, enlargement of graduate school, and establishment of various institutes, etc. And in guideline, selector, method of selection, level of selection and arrangement of budget, etc. must be described concretely in its contents. Since collection development policy and guideline of Korean university libraries should concern their situation, we must confer with the result of the survey and analysis on the matter of collection determination. Here the contents include the priority of materials to be collected, method of collecting materials, arrangement of budget, and others. The purpose of this study is to develop a tenative collection development guideline in reference to the Jeonju University Library by means of analyzing the guiding principle, contents of the guideline, and present conditions of the Korean university libraries. The systematic collection development based on the Guideline for Collection Development of the Korean University Library, can offer a satisfactory service to the users of the university community, and also contribute to the development of the university itself as well as the university library.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.28 no.4
• /
• pp.409-415
• /
• 2015

For the design of Concrete-Filled Steel Tube(CFST) columns, the outside diameter D to the steel tube thickness t ratio(D/t ratio) is limited to prevent the local buckling of steel tubes. Each design code proposes the respective model to compute the maximum D/t ratio using the yield strength of steel $f_y$ or $f_y$ and the elastic modulus of steel E. Considering the uncertainty in $f_y$ and E, the reliability index ${beta}$ for the local buckling of a CFST section can be calculated by formulating the limit state function including the maximum D/t models. The resulted ${beta}$ depends on the maximum D/t model used for the reliability analysis. This variability in reliability analysis is due to ambiguity in choosing computational models and it is called as "modelling uncertainty." This uncertainty can be considered as "non-specificity" of an epistemic uncertainty and modelled by constructing possibility distribution functions. In this study, three different computation models for the maximum D/t ratio are used to conduct reliability analyses for the local buckling of a CFST section and the reliability index ${beta}$ will be computed respectively. The "non-specific ${beta}s$" will be modelled by possibility distribution function and a metric, degree of confirmation, is measured from the possibility distribution function. It is shown that the degree of confirmation increases when ${beta}$ decreases. Conclusively, a new set of reliability indices associated with a degree of confirmation is determined and it is allowed to decide reliability index for the local buckling of a CFST section with an acceptable confirmation level.

• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.241-254
• /
• 2011

Financial time-series forecasting is one of the most important issues because it is essential for the risk management of financial institutions. Therefore, researchers have tried to forecast financial time-series using various data mining techniques such as regression, artificial neural networks, decision trees, k-nearest neighbor etc. Recently, support vector machines (SVMs) are popularly applied to this research area because they have advantages that they don't require huge training data and have low possibility of overfitting. However, a user must determine several design factors by heuristics in order to use SVM. For example, the selection of appropriate kernel function and its parameters and proper feature subset selection are major design factors of SVM. Other than these factors, the proper selection of instance subset may also improve the forecasting performance of SVM by eliminating irrelevant and distorting training instances. Nonetheless, there have been few studies that have applied instance selection to SVM, especially in the domain of stock market prediction. Instance selection tries to choose proper instance subsets from original training data. It may be considered as a method of knowledge refinement and it maintains the instance-base. This study proposes the novel instance selection algorithm for SVMs. The proposed technique in this study uses genetic algorithm (GA) to optimize instance selection process with parameter optimization simultaneously. We call the model as ISVM (SVM with Instance selection) in this study. Experiments on stock market data are implemented using ISVM. In this study, the GA searches for optimal or near-optimal values of kernel parameters and relevant instances for SVMs. This study needs two sets of parameters in chromosomes in GA setting : The codes for kernel parameters and for instance selection. For the controlling parameters of the GA search, the population size is set at 50 organisms and the value of the crossover rate is set at 0.7 while the mutation rate is 0.1. As the stopping condition, 50 generations are permitted. The application data used in this study consists of technical indicators and the direction of change in the daily Korea stock price index (KOSPI). The total number of samples is 2218 trading days. We separate the whole data into three subsets as training, test, hold-out data set. The number of data in each subset is 1056, 581, 581 respectively. This study compares ISVM to several comparative models including logistic regression (logit), backpropagation neural networks (ANN), nearest neighbor (1-NN), conventional SVM (SVM) and SVM with the optimized parameters (PSVM). In especial, PSVM uses optimized kernel parameters by the genetic algorithm. The experimental results show that ISVM outperforms 1-NN by 15.32%, ANN by 6.89%, Logit and SVM by 5.34%, and PSVM by 4.82% for the holdout data. For ISVM, only 556 data from 1056 original training data are used to produce the result. In addition, the two-sample test for proportions is used to examine whether ISVM significantly outperforms other comparative models. The results indicate that ISVM outperforms ANN and 1-NN at the 1% statistical significance level. In addition, ISVM performs better than Logit, SVM and PSVM at the 5% statistical significance level.

• Journal of Oriental Neuropsychiatry
• /
• v.26 no.1
• /
• pp.63-78
• /
• 2015

Objectives: Pinelliae Rhizoma has traditionally been used as an anti-depressant in oriental medicine. This study is to investigate the effect of Pinelliae Rhizoma extract (PRe) on psychological stress in genome wild expression of mice. Methods: After giving physical stress to mice, PRe was orally administered with 100 mg/kg/day for five days. After extracting whole brain tissue from the mice, their genome changes were observed by micorarray analysis method. The genome changes were analyzed by IMAGENE 4.0, TREEVIEW, FatiGo algorithems, BOND database, cytoscape program, etc. Results: 1. PRe administered group were remained at normal level; 60% of increase was shown in expressed genes by physical stress, and 65% of decrease was shown in expressed genes by psychological stress. 2. Genes with increased expression in control group that remained at a normal state in PRe administered group were involved with the gene of a cellular metabolic process on biological process, protein binding on molecular function, and cell part on cell composition. The pathway was found to be cytokin-cytokin receptor interaction. 3. Genes with decreased expression in control group that remained at a normal state in PRe administered group were involved with the gene of a cellular metabolic process on biologiacl detail and coupled ATPaes activity on molecular function. This gene related path was Ubiquintin mediated proteolysis etc. 4. Core node genes analyzed by protein interaction network were Vinculin, Cell sdivision cycle 42 homolog (S. cerevisiae) etc. They played an important role in maintaining cytoskeleton and controlling cell cycle. Conclusions: Several genes were up-regulated and down-regulated in response to psychological stress. The expression of most of the genes that were altered in response to psychological stress was restored to normal levels in PRe treated mice. When the interaction network information was analyzed, the recovery of the core node genes in PRe treated mice indicates that this final set of genes may be the effective target of PRe.

• Journal of Korean Neurosurgical Society
• /
• v.61 no.3
• /
• pp.363-375
• /
• 2018

Intraoperative monitoring (IOM) utilizes electrophysiological techniques as a surrogate test and evaluation of nervous function while a patient is under general anesthesia. They are increasingly used for procedures, both surgical and endovascular, to avoid injury during an operation, examine neurological tissue to guide the surgery, or to test electrophysiological function to allow for more complete resection or corrections. The application of IOM during pediatric brain tumor resections encompasses a unique set of technical issues. First, obtaining stable and reliable responses in children of different ages requires detailed understanding of normal age-adjusted brain-spine development. Neurophysiology, anatomy, and anthropometry of children are different from those of adults. Second, monitoring of the brain may include risk to eloquent functions and cranial nerve functions that are difficult with the usual neurophysiological techniques. Third, interpretation of signal change requires unique sets of normative values specific for children of that age. Fourth, tumor resection involves multiple considerations including defining tumor type, size, location, pathophysiology that might require maximal removal of lesion or minimal intervention. IOM techniques can be divided into monitoring and mapping. Mapping involves identification of specific neural structures to avoid or minimize injury. Monitoring is continuous acquisition of neural signals to determine the integrity of the full longitudinal path of the neural system of interest. Motor evoked potentials and somatosensory evoked potentials are representative methodologies for monitoring. Free-running electromyography is also used to monitor irritation or damage to the motor nerves in the lower motor neuron level : cranial nerves, roots, and peripheral nerves. For the surgery of infratentorial tumors, in addition to free-running electromyography of the bulbar muscles, brainstem auditory evoked potentials or corticobulbar motor evoked potentials could be combined to prevent injury of the cranial nerves or nucleus. IOM for cerebral tumors can adopt direct cortical stimulation or direct subcortical stimulation to map the corticospinal pathways in the vicinity of lesion. IOM is a diagnostic as well as interventional tool for neurosurgery. To prove clinical evidence of it is not simple. Randomized controlled prospective studies may not be possible due to ethical reasons. However, prospective longitudinal studies confirming prognostic value of IOM are available. Furthermore, oncological outcome has also been shown to be superior in some brain tumors, with IOM. New methodologies of IOM are being developed and clinically applied. This review establishes a composite view of techniques used today, noting differences between adult and pediatric monitoring.

• Journal of Korean Society of Archives and Records Management
• /
• v.21 no.3
• /
• pp.227-246
• /
• 2021

The study seeks to analyze the procedures and methods of preparing the records management reference table of the electronic human resource management system dataset, the roles of participating organizations, and the contents of each management reference table area from the records manager's perspective to help the person in charge of establishing the management reference table. Improvement plans were suggested based on the problems that appeared during the process of preparing the reference table. As a major improvement plan, a separate selecting policy at the level of the national archives should be designed for the national important dataset records in the government information system, which should be operated such that it preserves the entire dataset rather than a part. It is necessary to set the unit function-data table-unstructured data mapping data as mandatory items, and the selection and management criteria for unstructured data that significantly influence system operation should be additionally prepared. Regarding the setting of the disposition delay period, because there is an aspect of increasing complexity, it is deemed desirable to operate it by integrating related unit functions or setting the retention period longer.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.50 no.4
• /
• pp.259-267
• /
• 2022

The existing method of deriving the fail-safe design requirements for the domestic developed rotary-wing aircraft system may miss the factors that cause critical system function failures, when being applied to the latest integrated avionics system. It is because the existing method analyzes the severity effect of the failures caused by a single item. To solve the issue, we present a systematic analysis procedure for deriving fail-safe design requirements of system architecture by utilizing functional hazard assessment and development assurance level analysis of SAE ARP4754A, international standard for complex system development. To demonstrate that our proposed procedure can be a solution for the aforementioned issue, we set up experimental environments that include common factors that can cause critical function failures of a system, and we conducted a cross-validation with the existing method. As a result, we showed that the proposed procedure can identify the potential critical common factors that the existing method have missed, and that the proposed procedure can derive fail-safe design requirements to control the common factors.

• Journal of Korean Medical Ki-Gong Academy
• /
• v.23 no.1
• /
• pp.10-29
• /
• 2024

Objective : The purpose of this study is to determine whether An's breathing meditation qigong therapy (ABMQT) delivers bioenergy to the frontal lobe, prefrontal lobe, the olfactory tract in the mesolimbic pathway, olfactory bulb, CV22, CV21, olfactory area and vocal-related areas in Parkinson's disease (PD) patients to help improve olfactory disorders (anosmia) and vocal functions. Methods : The subjects of this study were 4 patients with idiopathic PD (3 males/1 female, 65.0±NA/68.7±10.2 years old). ABMQT was applied once a week, 120 minutes per session for 12 weeks in a non-invasive and noncontact manner, and the test before and after ABMQT application included olfactory impairment test the Korean version of Sniffin' stick test (KVSS), voice acoustic test, aerodynamic test, vocal handicap index (VHI-30), and auditory perception scale test tools. The results before and after the experiment were analyzed assuming a normal distribution, and a chi-square test was performed using a continuity correction, and the significance level was set to p<0.05. And the medical diagnosis and findings of the examiner (doctor in charge) before and after the experiment were described. Results : KVSS was significant as 0.2±0.5 and 9.0±0.0 before and after the experiment. There was no significant difference between the voice acoustic test FO and Jitter, the vocal aerodynamic test MPT, SP, AE, the vocal disorder index test, and the auditory perception test. However, the medical diagnosis findings of four study subjects showed that olfactory disorders, voice disorders, and laryngeal function were improved before and after the application of ABMQT. Conclusions : The breathing meditation qigong program showed significant effects on improving the olfactory disorders (anosmia) and speech function of each study subject. However, to produce meaningful results, it is thought that experiments involving a larger number of research participants are necessary, and additional blood and FMRI tests are conducted to verify metabolic activities and the olfactory neuron signal transmission system.