• Journal of Communications and Networks
• /
• v.4 no.2
• /
• pp.136-147
• /
• 2002

In this paper, the maximal capacity of the data network link has attempted to be exploited by using the dynamic allocation strategy. We propose a new methodology based on the economic models for competing traffic classes (classes of sessions) in packet networks. As the demand for network services accelerates, users' satisfaction to the service level might decrease due to the congestion at the network nodes. To prevent this, efficient allocation of a networks resources, such as available bandwidth and switch capacity, is needed. By using the so-called user profile as well as the utility (e.g., data rate) functions, it is possible to allocate data rates and other utilities using the arbitrary number of QoS classes, say $0.01,...,$10.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.585-587
• /
• 2023

Long-tailed datasets have an imbalanced distribution because they consist of a different number of data samples for each class. However, there are problems of the performance degradation in tail-classes and class-accuracy imbalance for all classes. To address these problems, this paper suggests a learning method for training of long-tailed dataset. The proposed method uses and combines two methods; one is a resampling method to generate a uniform mini-batch to prevent the performance degradation in tail-classes, and the other is a reweighting method to address the accuracy imbalance problem. The purpose of our proposed method is to train the learning models to have uniform accuracy for each class in a long-tailed dataset.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.2
• /
• pp.45.3-46
• /
• 2021

Supernovae classes have been defined phenomenologically, based on spectral features and time series data, since the specific details of the physics of the different explosions remain unrevealed. However, the number of these classes is increasing as objects with new features are observed, and the next generation of large-surveys will only bring more variety to our attention. We apply the machine learning technique of multi-label classification to the spectra of supernovae. By measuring the probabilities of specific features or 'tags' in the supernova spectra, we can compress the information from a specific object down to that suitable for a human or database scan, without the need to directly assign to a reductive 'class'. We use logistic regression to assign tag probabilities, and then a feed-forward neural network to filter the objects into the standard set of classes, based solely on the tag probabilities. We present STag, a software package that can compute these tag probabilities and make spectral classifications.

• Korean Journal of Computational Design and Engineering
• /
• v.13 no.2
• /
• pp.98-108
• /
• 2008

Plant, civil engineering and construction equipment data of overseas construction are obtained and then analyzed, classified and integrated by experts. With those refined data set, we built classification system and defined property information with reference to international standard (ISO 15926, IRDL). If class in ISO 15926 is predefined for equipment of interest, we used the class as is. If not, we created and defined new classes on the basis of ISO 15926 classes. If there is similar class for equipment of interest, extension or inheritance methods were used. As a result, classification system of five levels and 637 classes were built and construction equipment information were expressed in open structure of XML such as tree structure of classification system and detailed information with number equipments for each specific equipment. We also developed the electronic catalog system which is basically equipment information management system providing various product search functions.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.24 no.1
• /
• pp.39-48
• /
• 1999

An economic procedure of selective assembly is proposed when a product is composed of two mating components. The major qualify characteristic of the product is the clearance between the two components. The components are divided into several classes prior to assembly. The component characteristics are assumed to be independently and normally distributed with equal variance. The procedure is designed so that the proportions of both components in their corresponding classes are the same. A cost model is developed based on a quadratic loss function and methods of obtaining the optimal class limits as well as the optimal number of classes are provided. Formulas for obtaining the proportion of rejection and the unavailability of mating components are also provided. The proposed model is compared with the equal width and the equal area partitioning methods using a numerical example.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.16 no.5
• /
• pp.583-589
• /
• 2013

In this paper, we proposed a system that efficiently provides support maps which includes the grid based terrain analysis information. To do this, we use the FDB which is defined as a GIS database that contains features with attributes attached to the features. The FDB is composed of a number of features and feature classes. In order to create support maps, it is necessary to classify feature classes that are associated with each support map and to search them in a grid map. The proposed system use a ontology model to classify semantically feature classes and the quad-tree data structure to find them in a grid map quickly. Therefore, our system is expected to be utilized for global path planning of UGV. In this paper, we show the possibility through an experimental implementation.

• Journal of Technologic Dentistry
• /
• v.44 no.3
• /
• pp.97-103
• /
• 2022

Purpose: This study sought to identify the factors influencing learning motivation and demotivation in online dental technology students. Methods: A survey was conducted from October 1 to 30, 2021, on 188 dental technology students. The collected data were processed using the IBM SPSS IBM SPSS Statistics ver. 22.0 statistical program (IBM), and frequency, factor, and one-way ANOVA analyses were performed, for which the significance was set at 0.05. Results: It was found that the main online learning motivation factors were the usefulness of the learning content, interest, and confidence in the activities, the relationships with the teachers and friends, the feedback, and learning satisfaction. The factors that reduced the students' online learning motivation were interaction difficulties, maladaptation to the self-directed learning environment, the inadequate number of learning activities, and activity difficulty. Conclusion: Based on the identified online class motivation and demotivation factors, better systematic management and increased research are needed to improve the quality of non-face-to-face classes.

• 한국정보교육학회:학술대회논문집
• /
• 2021.08a
• /
• pp.45-50
• /
• 2021

In order for SW education to be established in the era of non-face-to-face caused by COVID-19, research on the efficiency of SW education according to face-to-face and non-face classes is needed. Therefore, this study classified the operation status of the curriculum of 30 SW gifted classes nationwide in 2020 according to the class method(face-to-face, non-face, and blended). Subsequently, the results of class time and production per person were compared and analyzed through quantitative analysis. According to the study, the type of classes that performed the most classes compared to the planned number of hours was non-face-to-face(90.9%), followed by face-to-face(84.2%) and the least was blended(80.5%). The average number of products per student was the highest in the face-to-face class(0.504), while the blended class(0.421) and non-face-to-face class(0.42). Based on the results of this study, the non-face-to-face approach is advantageous in securing the number of hours, but various measures should be prepared to solve this problem because teachers and students find it difficult to guide the output.

• Journal of applied mathematics & informatics
• /
• v.13 no.1_2
• /
• pp.413-424
• /
• 2003

L. H Encinas, A. J. Menezes and J. M. Masque in [3] proposed a classification of isomorphism classes of hyperelliptic curve of genus 2 over finite fields with characteristic different from 2 and 5. Y. Choie and D. Yun in [2] obtained the number of isomorphic classes of hyperelliptic curves of genus 2 over $F_{2-}$ using direct counting method. We have obtained isomorphism classes of hyperelliptic curves of genus 2 over $F_{2n}$ for odd n, represented by an equation of the form $y^2$ + $a_{5}$ y = $x^{5}$ + $a_{8}$ x + $a_{10}$ ( $a_{5}$ $\neq$0) [1]. In this paper we characterize hyperelliptic curves of genus 2 over $F_{2n}$ for even n, represented by an equation of the form $y^2$ + $a_{5}$ y = $x^{5}$ + $a_{5}$ x + $a_{10}$ ( $a_{5}$ $\neq$0).>0).

• Journal of the Korean Institute of Rural Architecture
• /
• v.16 no.1
• /
• pp.43-52
• /
• 2014

Education is about sharing knowledge and using techniques and skills that can person's ability. the goal is to educate and develop people as a whole. In order to achieve this educational purpose efficiently, our society has many educational courses. The educational facility which can be used in a variety of educational methods should be made flexible. It should provide an environment in which ever student can grow. Being creative, using a variety of methods should help to develop character in people. In order to achieve this, the school's facilities should be future-oriented. It is necessary to research systemically about its changes and feed back according to the social demand, the educational purpose, and the changes of educational courses. This study is intended to present basic material which will be need from now on for the facility of the special classes in jeju city. Elementary schools should consider having a reasonable scale and arrangement pattern. The facility for special classes has to find a way to operate flexibly according to the changes of the class number.