• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.41 no.2
• /
• pp.56-64
• /
• 2018

In recent years, imbalanced data is one of the most important and frequent issue for quality control in industrial field. As an example, defect rate has been drastically reduced thanks to highly developed technology and quality management, so that only few defective data can be obtained from production process. Therefore, quality classification should be performed under the condition that one class (defective dataset) is even smaller than the other class (good dataset). However, traditional multi-class classification methods are not appropriate to deal with such an imbalanced dataset, since they classify data from the difference between one class and the others that can hardly be found in imbalanced datasets. Thus, one-class classification that thoroughly learns patterns of target class is more suitable for imbalanced dataset since it only focuses on data in a target class. So far, several one-class classification methods such as one-class support vector machine, neural network and decision tree there have been suggested. One-class support vector machine and neural network can guarantee good classification rate, and decision tree can provide a set of rules that can be clearly interpreted. However, the classifiers obtained from the former two methods consist of complex mathematical functions and cannot be easily understood by users. In case of decision tree, the criterion for rule generation is ambiguous. Therefore, as an alternative, a new one-class classifier using hyper-rectangles was proposed, which performs precise classification compared to other methods and generates rules clearly understood by users as well. In this paper, we suggest an approach for improving the limitations of those previous one-class classification algorithms. Specifically, the suggested approach produces more improved one-class classifier using hyper-rectangles generated by using Gaussian function. The performance of the suggested algorithm is verified by a numerical experiment, which uses several datasets in UCI machine learning repository.

• Journal of Information Technology Applications and Management
• /
• v.15 no.2
• /
• pp.33-49
• /
• 2008

The purpose of this study is to find student factors associated with the students' evaluation of university teaching, and to provide a meaningful reference to policy-making of teaching evaluation. Based on the surveyed questionnaires from 232 students who participated in e-learning course. in combination with applying descriptive statistics, this study analyzes (1) the influences of students' pre-motivation level of student based-on ARCS theory on class satisfaction, (2) the influences of students' grade, attendance, and class involvement on class satisfaction, and (3) the influences of e-learning experience on students' pre-motivation level and class satisfaction. Result of this study shows that: (1) students' pre-motivation level including Attention(A), Relevance(R), Confidence(C), and Satisfaction(S) is positively correlated to students' evaluation for teaching, (2) students' grade, attendance. and class involvement are positively correlated to students' evaluation for teaching, and (3) students' e-learning experience is not related with students' pre-motivation level, whereas it is related with class satisfaction.

• The Journal of Information Systems
• /
• v.31 no.3
• /
• pp.19-45
• /
• 2022

Purpose Since COVID-19, the demand for online class platforms has increased. However, those platforms have not been clearly defined, and related research is also limited. In the context of the knowledge marketplace (KMs), this study examined the effects of class information and trust in creators on class preferences from the perspective of consumption value theory. Design/methodology/approach By establishing a web crawler through Python, this study collected 1,174 class data in Korea's leading knowledge marketplace, Class 101, focusing on diverse class-related information and the number of Instagram followers for individual class creators. Based on class information, this research analyzed the effects of consumers' utilitarian value, social value, and hedonic value on class preference. In addition, this study examined whether consumers' trust in creators moderates the relationship between class information and class preference. Findings According to analysis results, it was found that the higher the consumers' consumption value for each class on KMs, the more positive their preference for the class. Also, it was confirmed that consumers' trust in creators moderates the relationship between class information and class preference.

• The Transactions of the Korean Institute of Electrical Engineers B
• /
• v.52 no.3
• /
• pp.133-139
• /
• 2003

In this paper, a new control scheme of Class-E inverter for Induction Heating (IH) Jar applications with clamped voltage characteristics using Pulse-Frequency-Modulation (PFM) is introduced. To reduce the voltage stress of switch, the proposed PFM control scheme doesn't need any auxiliary circuit in comparison to a family of Active Clamped Class-E (ACCE) inverter. It can decrease voltage stress of switch through modulation of switching frequency. The Class-E inverter using the proposed control scheme has the advantage of not only the same output power when it is compared with a Hybrid-Active Clamped Class-E (Hybrid-ACCE) inverter but also Zero-Voltage-Switching (ZVS), which are characteristics of conventional Class-E and ACCE inverter. The control principles and analysis of proposed method are explained in detail and its validity is verified through simulation and experimental results.

• The Korean Journal of Malacology
• /
• v.18 no.2
• /
• pp.93-217
• /
• 2002

Molluscan specimens collected in the whole country of Korea during the from 1994 through 2002, were identified and classified. A total of 1,560 molluscan species were listed in this study; one species in class Aplacophora, 18 species in class Polyplacophora, 1,038 species in class Gastropoda, 467 species in class Bivalvia, 14 species in class Scaphopoda and 22 species in class Cephalopoda. Among the identified species, 360 species (three species in class Polyplacophora, 266 species in class Gastropoda, 85 species in class Bivalvia, 6 species in class Scaphopoda) are newly recorded to Korean fauna.

• Journal of Korean Academy of Fundamentals of Nursing
• /
• v.8 no.3
• /
• pp.357-365
• /
• 2001

Purpose: The purpose of this study was to identify the death orientation of college student nurses and to evaluate the effectiveness in death orientation of a class on 'hospice care'. this was done by investigating the difference between the death orientation before and after the class. Method: The instrument used in this Investigation was the death orientation scale developed by Thorson and Powell (1988). The subjects of the study were the 47 college student nurses in a 'hospice care' class. The research was carried out between March 2, 2001, when the class began, and June 21, 2001 when the class ended The collected data were examined using arithmetic mean, percentage, t-test, ANOVA. Result: The results of the study are as follows : 1. The average death orientation score of the subjects before the class was 2.80. 2. There were a significant difference in the death orientation scores before and after the class (p<0.001). Before the class, 'hospice care', the students had an average score of 2,80. Their average score decreased to 2.55 after the class. Of 25 death orientation items on the scale, 14 items decreased significantly after the class as compared to before the class. 3. According to personal attributes (religion. the existence of religion, the experience of a death in recently) of subjects, there were no significant statistical differences in death orientation (p>0.05). Conclusion: This study showed that a class on hospice care was effective in changing the death orientation of student nurses.

• The korean journal of orthodontics
• /
• v.52 no.6
• /
• pp.432-438
• /
• 2022

Objective: To compare crown-root angulations of the permanent maxillary anterior teeth in skeletal Class I, Class II, and Class III Korean malocclusion patients using cone-bean computed tomography (CBCT) images. Methods: Sixty CBCT images were collected from orthodontic patients archive based on skeletal Class I (0˚< A point-nasion-B point angle [ANB] < 4˚), Class II (ANB ≥ 4˚), and Class III (ANB ≤ 0˚) to have 20 samples in each group. Mesiodistal crown-root angulation (MDCRA) and labiolingual crown-root angulation (LLCRA) were evaluated after orientation of images. Crown-root angulations were compared among Class I, Class II, and Class III groups and among the maxillary anterior teeth in each group. Results: LLCRAs of the maxillary central incisor and the lateral incisor were significantly lower in Class III group than those in Class I group. However, those of the canine showed no significant differences among groups. MDCRAs of the maxillary anterior teeth did not significantly differ among groups either. Conclusions: Our results suggest that skeletal Class III malocclusion might affect LLCRA of the maxillary incisors, especially the central incisor.

• Journal of Technologic Dentistry
• /
• v.39 no.4
• /
• pp.263-273
• /
• 2017

Purpose: This study aimed to explore the relations between class distracting factors and class satisfaction of the dental technology students and then provide a primary data to help further related studies and develop educational programs with which instructors can efficiently manage their classroom. Methods: For this study we have conducted a survey started from the beginning of May 2017 to the end of June. The subjects of the survey were Dental Technology students of D-city, K-city, W-city, selected by random sampling method. The questionnaire was self-administrated and 437 valid results were chosen for our analysis among 450 distributed questionnaires. Results: The results of the research was as follows. Firstly, The overall average point of class distracting factors was 2.5 point. The environmental factors were the highest point as 2.59 and as for the subcategories tiredness and drowsiness was the highest point as 2.76. Secondly, The overall average point of class satisfaction turned out 3,88 point and compliance with class and attitude factors gained the highest point as 4.06. Of the subcategories strict roll checking was the highest point as 4.17. Thirdly, As for class distracting factors from general characteristics a statistical significance was shown as follows; 'instructor factor'(p<.01), 'learner factor'(p<.05), 'total class distracting factor'(p<.05) in the area of gender, 'environmental factor'(p<.001), 'total class distracting factor'(p<.01), 'learner factor'(p<.05), 'instructor factor'(p<.05) in the area of gender 'learner factor'(p<.001), 'instructor factor'(p<.001), 'environmental factor'(p<.001), 'total class distracting factor'(p<.01) in the area of class grade, 'environmental factor'(p<.05) in GPA. Fourthly, A statistical significance, a negative correlation (p<.01) were shown between class distracting factors and class satisfaction. Class distracting factor that especially affects the class satisfaction was instructor factor(p<.001) and the explanatory power of the model turned out 14.7%, which was statistically meaningful (p<.001). Conclusion : Results of this study reveal that instructor factor is the key to class satisfaction of the students. So it is crucial that the instructor faithfully prepare for the class to reinforce the students' learning. Additionally further studies should be followed with more subjects and newer perspectives to develop innovative teaching methodology.

• Journal of the Chungcheong Mathematical Society
• /
• v.22 no.4
• /
• pp.799-814
• /
• 2009

A class invariant is the value of a modular function that generates a ring class field of an imaginary quadratic number field such as the singular moduli of level 1. In this paper, using Shimura Reciprocity Law, we compute the Galois actions of some class invariants from the generalized Weber functions $\mathfrak{g}_0,\mathfrak{g}_1,\mathfrak{g}_2$ and $\mathfrak{g}_3$ over quadratic number fields with discriminant $D{\equiv}1$ (mod 12).

• Journal of the Korean Mathematical Society
• /
• v.40 no.2
• /
• pp.273-286
• /
• 2003

Let $F^{\alpha}$G be a twisted group algebra. A subalgebra of $F^{\alpha}$G generated by all class sums of partition P of G is called a projective class algebra in $F^{alpha}$G associated with partition P. In this paper we study various partitions of G determined by actions of certain operator groups on G and construct projective class algebras depending on the actions. With regard to projective class algebras, we investigate structures of associated skew group algebras and fixed group algebras.