• Journal of the Korean Chemical Society
• /
• v.31 no.6
• /
• pp.567-575
• /
• 1987

Fourty five ancient Korean glass pieces have been determined for 19 elements such as Ag, As, Br, Ce, Co, Cr, Eu, Fe, Hf, K, La, Lu, Na, Ru, Sb, Sc, Sm, Th and Zn, and for one such as Pb by instrumental neutron activation analysis and by atomic absorption spectrometry, respectively. The multivariate data have been analyzed for the relation among elemental contents through the variance-covariance matrix. The data have been further analyzed by a principal component mapping method. As the results training set of 5 class have been chosen, based on the spread of sample points in an eigen vector plot and archaeological data. The 5 training set consisting of 36 species and a test set consisting of 9 species bave finally been analyzed for the assignment to certain classes or outliers through the statistical isolinear multiple component analysis (SIMCA). The results have showed the whole species for 5 training set and 3 species in the test set are assigned appropriately and these are in accord with the results by principal component mapping.

• Journal of Korea Game Society
• /
• v.3 no.2
• /
• pp.35-41
• /
• 2003

There arose problems of double investment about the game engine part when a lot of game software similar to the property and procedure processed in the game engine develop new game without the reference or reuse in the other games. In particular, using various software development processes is one of main problems of double investment when the enterprises for the game software development develop games now Accordingly, because it does not make standardization of process about the game engine, it does not understand and reuse products created in process of the other software development process in development now. Accordingly, the newly analyzed and designed software was big problems with the present game software about the game engine process similar to the other game software when the enterprises for any game software develop a special game. For solving these problems, this study is to suggest the process improvement about the game engine development, analysis of structure and relation, classification and combination method by the class and module, implementation of storage, and processor model in order to apply the development method based on the component.

• The Korean Journal of Applied Statistics
• /
• v.29 no.2
• /
• pp.309-319
• /
• 2016

For binary classification models, we consider a risk score that is a function of linear scores and estimate the coefficients of the linear scores. There are two estimation methods: one is to obtain MLEs using logistic models and the other is to estimate by maximizing AUC. AUC approach estimates are better than MLEs when using logistic models under a general situation which does not support logistic assumptions. This paper considers imbalanced data that contains a smaller number of observations in the default class than those in the non-default for credit assessment models; consequently, the AUC approach is applied to imbalanced data. Various logit link functions are used as a link function to generate imbalanced data. It is found that predicted coefficients obtained by the AUC approach are equivalent to (or better) than those from logistic models for low default probability - imbalanced data.

• Journal of the Korean Society of School Health
• /
• v.8 no.1
• /
• pp.155-166
• /
• 1995

The aim of this study was to determine the extent to which sociodemographic and health related life-style variables explain body weight distribution and to understand weight contol behavior. To study this study 298 students were selected, it was consisted of obesity group(101) and control group(197). The average age of subjects was 14.2 and the prevalence of obesity was 2-3 per class as 5.6% among 1,793. 71% among same subject was showed higher weight pattern than last one year, ovesity group which was obesity both in 93 and 94 was 34%. Correlation between body weight(under weight/obesity) and independent variables including sociodemographic factor and health- related life style tested through Multiple Classification Analysis was very significant, explained 36% of the total variance. Sociodemografic and hereditary factors such as education level, age of father and physical features of parents, life style factors as exercise preference and perceived health status showed highly contribution to body weight. Concretely, there were showed a higher obesity prevalence tendency when education level and age of father was high, physical features of parents was obesity. In otherwise, there were showed a higher underweight prevalence tendency when education level and age of father was low. Experience rates of weight control was 53% generally, 84% in obesity group, and 11% in underweight group. There were utilized weight control behaviors through diet method mainly in obesity group, diet and exercise methods in underweight group. There were showed that underweight group are prefer exercise to obesity group. Conclusionally, These findings suggest that education, age, physical features of parents, exercise preference and perceived health status is important factors related to body weight among middle school girls. Therefore, there will be considered as valuable factors when we practice health education and consultation related to body weight. Furthermore it is necessary to provide of various informations about weight control and to develop systematic weight control program.

• The Journal of Information Systems
• /
• v.31 no.1
• /
• pp.59-90
• /
• 2022

Purpose This paper presents a new method of classifying online platforms. It also explains how to apply the framework using case studies and generate new insight about platform strategies and policy development. Design/methodology/approach This paper focuses on the relationship between platforms, especially the hierarchy and power relations, and broadly classifies platforms as follows: content/services, meta information, app stores, operating systems, and cloud. Both the content/service platform and the meta information platform have matching as their main function. However, most content/services tend to collect and access information through meta-information platforms, so meta-information platforms are closer to infrastructure than content/service platforms. App store, operating system, and cloud can be said to be platforms of platforms. A small number of companies in the US and China dominate platforms of platforms, and become the recent development and regulatory targets of their respective governments. Findings We should be wary of the attempts to regulate domestic platforms by importing foreign regulations that ignore the hierarchical structure that our framework highlights. We believe that Korea's strategy to become a true platform powerhouse is clear. As one of the few countries with significant companies in the area of meta information platforms, it will be necessary to fully utilize the position and advance into the strategically important area of platforms of platforms. Furthermore, it is necessary to encourage world-class companies to appear in Korea in the app store, operating system, and cloud. To do so, the government needs to introduce promotion policies to strategically nurture such platforms rather than to regulate them.

• Journal of Korea Foundry Society
• /
• v.42 no.6
• /
• pp.369-376
• /
• 2022

The die-casting process is an important process for various industries, but there are limitations in the profitability and productivity of related companies due to the high defect rate. In order to overcome this, this study has developed die-casting fault detection modules based on industrial AI technologies. The developed module is constructed from three-stage models depending on the characteristics of the dataset. The first-stage model conducts fault detection based on supervised learning from the dataset without labels. The second-stage model realizes one-class classification based on semi-supervised learning, where the dataset only has production success labels. The third-stage model corresponds to fault detection based on supervised learning, where the dataset includes a small amount of production failure cases. The developed fault detection module exhibited outstanding performance with roughly 96% accuracy for actual process data.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.1D
• /
• pp.133-140
• /
• 2006

This paper suggests to reclassify railroad accident types and to standardize them as the standardized code for the railroad safety management system. The existing railroad accident types in both domestic and foreign cases have been carefully analyzed in the beginning. Based on the case studies, the new railroad accident types are classified into 9 classes which are not overlapped one another and 9 classes have been subdivided into 40 different accident patterns. All these patterns are linked with 9 different accident objects and 6 accident locations. Therefore, this study suggested the combination of 4 distinct code factors: accident class, accident pattern, accident object, and accident location to standardize them. In addition, inter-operation between the proposed codes and the existing accident types is suggested. This code will play a major role in the railroad safety management system composed of accident prevention, accident preparedness, accident response, and accident recovery.

• Journal of Sasang Constitutional Medicine
• /
• v.11 no.1
• /
• pp.137-157
• /
• 1999

The purpose of this study is to help classifying Sasang Constitution through correlation with sound characteristic. This study was done it under the suppose that Sasang Constitution has correlation with sound spectrogram. The following result were obtained about correlation between sound spectrogram and Sasang Constitution by comparison and analysis 1. Soeumin answered his voice low tone, smooth and quiet in the survey. Soyangin answered his voice high, clear, fast and speaking random. Taeumin answered his voice low, thick and muddy. 2. Taeyangin was significantly slow compared with the others in the time of reading composition. Taeyangin was significantly slow compared with the others in Formant frequency 1. Taeyangin was significantly discriminated from Soeumin in Formant frequency 5. Taeyangin was significantly low compared with the others in Bandwidth 2. Soeumln was significantly low compared with Taeyangin in Pitch Maximum and Pitch Maximum-Pitch Minimum. Taeyangin was significantly high compared with the others in Energy mean. 3. In list of specification, the discrimination rate was higher than that by lists of 13 in the results of Multi-dimensional 4-class minimum-distance. The discrimination rate of three disposition except Soyangin was higher than that of four disposition in the results of One way ANOVA and Analysis of dis crimination in SPSS/PC+. In CART, the estimate rate of Sasang Constitution discrimination was higher than any other method. It is considered that there is a correlation between sound spectrogram and Sasang constitution according to the results. And method of Sasang constitution classification through sound spectrogram analysis can be one method as assistant for the objectification of Sasang constitution classification.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.40 no.1
• /
• pp.19-29
• /
• 2003

This paper presents a novel hybrid classifier for face recognition with artificially generated virtual training samples. We utilize both the nearest neighbor approach in feature angle space and a connectionist model to obtain a synergy effect by combining the results of two heterogeneous classifiers. First, a classifier called the nearest feature angle (NFA), based on angular information, finds the most similar feature to the query from a given training set. Second, a classifier has been developed based on the recall of stored frontal projection of the query feature. It uses a frontal recall network (FRN) that finds the most similar frontal one among the stored frontal feature set. For FRN, we used an ensemble neural network consisting of multiple multiplayer perceptrons (MLPs), each of which is trained independently to enhance generalization capability. Further, both classifiers used the virtual training set generated adaptively, according to the spatial distribution of each person's training samples. Finally, the results of the two classifiers are combined to comprise the best matching class, and a corresponding similarit measure is used to make the final decision. The proposed classifier achieved an average classification rate of 96.33% against a large group of different test sets of images, and its average error rate is 61.5% that of the nearest feature line (NFL) method, and achieves a more robust classification performance.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.133-140
• /
• 2022

We investigate the performance of deep learning-based Korean language models on a task of predicting the score range of Korean essays written by foreign students. We construct a data set containing a total of 304 essays, which include essays discussing the criteria for choosing a job ('job'), conditions of a happy life ('happ'), relationship between money and happiness ('econ'), and definition of success ('succ'). These essays were labeled according to four letter grades (A, B, C, and D), and a total of eleven essay score range prediction experiments were conducted (i.e., five for predicting the score range of 'job' essays, five for predicting the score range of 'happiness' essays, and one for predicting the score range of mixed topic essays). Three deep learning-based Korean language models, KoBERT, KcBERT, and KR-BERT, were fine-tuned using various training data. Moreover, two traditional probabilistic machine learning classifiers, naive Bayes and logistic regression, were also evaluated. Experiment results show that deep learning-based Korean language models performed better than the two traditional classifiers, with KR-BERT performing the best with 55.83% overall average prediction accuracy. A close second was KcBERT (55.77%) followed by KoBERT (54.91%). The performances of naive Bayes and logistic regression classifiers were 52.52% and 50.28% respectively. Due to the scarcity of training data and the imbalance in class distribution, the overall prediction performance was not high for all classifiers. Moreover, the classifiers' vocabulary did not explicitly capture the error features that were helpful in correctly grading the Korean essay. By overcoming these two limitations, we expect the score range prediction performance to improve.