• Journal of Korean Institute of Industrial Engineers
• /
• v.40 no.4
• /
• pp.404-414
• /
• 2014

Yield prediction is important to manage semiconductor quality. Many researches with machine learning algorithms such as SVM (support vector machine) are conducted to predict yield precisely. However, yield prediction using SVM is hard because extremely imbalanced and big data are generated by final test procedure in semiconductor manufacturing process. Using SVM algorithm with imbalanced data sometimes cause unnecessary support vectors from major class because of unselected support vectors from minor class. So, decision boundary at target class can be overwhelmed by effect of observations in major class. For this reason, we propose a under-sampling method with minor class based SVM (MCSVM) which overcomes the limitations of ordinary SVM algorithm. MCSVM constructs the model that fixes some of data from minor class as support vectors, and they can be good samples representing the nature of target class. Several experimental studies with using the data sets from UCI and real manufacturing process represent that our proposed method performs better than existing sampling methods.

• Journal of The Korean Association of Information Education
• /
• v.6 no.1
• /
• pp.75-85
• /
• 2002

The change of following information society, now a point of time is that we need a cyberspace as Class Management Support System for effective classroom Management of web base and realized it and then tried to find the Internet use direction correctively. It guided more effective class management. Class Management Support System is applied that are realized information possibly in Web according to class management's work. we will incite more the Internet use resource is made ability to use in learning and stimulated their hobbies and specialities. That can be resulted fully the support role of usual calss management escape from time and space restriction. So we adapted this system when we leaded students participated themselves actively about student's activity, prospected the chance which is accomplished higher advanced education.

• Journal of the Korea Society for Simulation
• /
• v.14 no.1
• /
• pp.33-42
• /
• 2005

Service Level Agreement (SLA) is a service providing scheme by a service class agreement between a user and a service provider SLA allows that a user can select an expected service class in various service classes provided from a service provider. Recently, SLA management is adapted to support the end-to-end Qos for service users in Beyond 3 Generation (B3G) networks. In B3G networks, SLA provides multiple service classes on access networks so that service classes should be managed to assure the service satisfaction for users. In this paper, we propose a dynamic Qos management scheme by IP traffic class controlling based on SLA in B3G networks. Also, to manage dynamic traffic service, we consider Differentiated services (Diffserv) mechanism for the resource management by SLA. An If service traffic class on SLA can be dynamically changed by Diffserv traffic management to support dynamic end-to-end Qos. Hence, in this paper, we consider the buffer threshold scheme for controlling traffic loads and the traffic level control scheme for implementing the dynamic traffic management by the SLA.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.7B
• /
• pp.641-649
• /
• 2004

We propose a dynamic traffic management model which uses adaptive priority reassignment algorithm to deliver service differentiation in home networks, and implement adaptive priority reassignment algorithm using FPGA. The proposed architecture is designed to handle home network traffic without the need for signaling protocol. We categorize home network traffic into three kinds of traffic class: control data traffic class, the Internet data and non-real-time data traffic class, and multimedia data traffic class (include non-real-time and real-time multimedia data traffic). To support differential service about these kinds of traffic class, we designed and implemented a traffic management framework that dynamically change each traffic class priority depending on bandwidth utilization of each traffic class.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.960-964
• /
• 2004

During system resource improvement process that based on Object-Oriented technology could be affect to the continuous system performance if lack appropriate management and control objects mechanism. This paper proposes a methodology to support continuous system performance and its stability. The adoption is based on Java Container Framework and Collections Framework for object collection. Also includes Software Engineering, Object Migration and Multiple Class Loaders mechanism accommodate to construct Continuous Migration Container (CMC). CMC is a runtime environment provides interfaces for management and control to support upgrading object process. Upgrade object methodology of CMC can be divided into two phase are object equivalence checking and object migration process. Object equivalence checking include object behavior verification and functional conformance verification before object migration process. In addition, CMC use Multiple Class Loaders mechanism to support reload effected classes instead of state transfer in migration process while upgrading object. These operations are crucial for system stability and enhancement efficiency.

• The Journal of Society for e-Business Studies
• /
• v.9 no.3
• /
• pp.145-154
• /
• 2004

Class extraction of object or component based software development methodology is the major factor for software quality. Each method has the class definition and extraction method however there are some troubles when the beginners try. Especially, the conceptual gap results to make the class extraction hard. This research suggests three diagrams to support it. They are Information diagram, Behavio diagram and Management diagram. They specify which services a stakeholder wants, which information to support the service, which actions to solve the service and who has the responsibility for those. Any analysis process which takes class extraction can utilize these diagrams.

• Journal of Korean Elementary Science Education
• /
• v.38 no.1
• /
• pp.55-72
• /
• 2019

The purpose of this study is to find out what kind of difficulties teachers face in the management of the elementary gifted institutions in the rural area of Chungbuk province. In this study, rural area was defined that the area where gifted education is difficult to access due to geographical accessibility or where gifted education service is restricted. The participants of the this study were three elementary gifted teachers who manage elementary gifted education institutions and is teaching science to disadvantaged gifted students. We collected data about the difficulties and support needs for the management of the gifted institutions in the rural area through the questionnaire and the interview. The results of this study are as follows. First, the common difficulty that teachers expressed was the lack of parental attention and awareness of gifted education. Second, the teachers who participated in this study perceived different difficulties according to their area, experience, and environment. Teacher A is the lack of awareness of the managers, teacher B is the difficulty of student selection, teacher C is the most difficult factor in teacher quality management. This difference in perceptions also led to differences in the support improvements required by gifted class teachers in the rural area. Third, the three teachers commonly referred to difficulties to access due to geographical accessibility, and demanded the integrated management of the gifted class in the rural area and the support for the expansion of the class.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.15 no.4
• /
• pp.121-133
• /
• 2020

This study aims to measure the result of R&D support projects on the beneficiary companies of the World Class 300 project, a representative domestic middle market enterprise incubation project. In particular, beneficiary companies were classified into industry groups(materials, components, equipment, modules, consumer goods) and the empirical study was conducted on how the result of R&D support project was differentiated by each industry group's characteristics. The analysis was performed on 272 companies selected as WC300 project during 2011 to 2017. The result of Panel regression Analysis confirmed that R&D support project had divergent influences on the performance (net profit) of companies depending on which industry group they belong. The samples of materials, components, equipment companies turned out not to have statistical significance. The samples of modules companies showed negative correlations, while those of consumer goods was observed to have positive correlations. This study demonstrates that it is essential to plan and implement policies suited to the characteristics of each industry groups to maximize the effect of R&D support projects henceforward.

• Asia pacific journal of information systems
• /
• v.21 no.2
• /
• pp.43-58
• /
• 2011

Support vector machines (SVMs), a machine learning technique, has been applied to not only binary classification problems such as bankruptcy prediction but also multi-class problems such as corporate credit ratings. However, in general, the performance of SVMs can be easily worse than the best alternative model to SVMs according to the selection of predictors, even though SVMs has the distinguishing feature of successfully classifying and predicting in a lot of dichotomous or multi-class problems. For overcoming the weakness of SVMs, this study has proposed an approach for selecting features for multi-class SVMs that utilize the impurity measures of classification trees. For the selection of the input features, we employed the C4.5 and CART algorithms, including the stepwise method of discriminant analysis, which is a well-known method for selecting features. We have built a multi-class SVMs model for credit rating using the above method and presented experimental results with data regarding S&P 500 companies.

• Journal of The Korean Association of Information Education
• /
• v.12 no.2
• /
• pp.131-139
• /
• 2008

Recently the wireless internet technologies have been widely used in education environments. The purpose of this paper is to design and implement mobile class management support system. The proposed system is based on the existing CSCL theory and modified in accordance with mobile environments. The proposed system has the following characteristics. First, various types of cooperation are possible in this system. In this system, students can participate actively through tools such as BBS. Second, wired and wireless web sites are linked systematically so that users can access via either site. Two sites are complementary each other. Third, parents are encouraged to participate in class activities. Some basic information such as curriculum and daily announcement is open for parents. The proposed system has been applied to elementary school students. Extensive survey shows that the proposed system has good effect on participation, interaction, and satisfaction in class management, respectively.