• 한국시뮬레이션학회논문지
• /
• 제12권2호
• /
• pp.55-62
• /
• 2003

The explosive Increase of wireless networks and the advancement of mobile devices lead to the expansion of mobile environment. In accordance with the development of mobile environment, the need to use mobile database is increased sharply, and also it accompanies the related problems. The current mobile database system is based on the centralized method from which a synchronized server manages multiple mobile database management system to synchronize. From this mobile system architecture, several kinds of problems can be detected such as the management of synchronization issues between mobile databases and the transaction management issues. Furthermore, the current mobile database management system does not consider any solution on the fault tolerance. To solve those problems, this paper proposes the mobile agent-based mobile database management system. The proposed system provide high confidence and efficiency by enhancing the network efficiency and fault tolerance through the mobile grouping.

• 산업공학
• /
• 제19권1호
• /
• pp.1-8
• /
• 2006

The mobile communication network has to offer good quality of services (QoS), high capacity, and more coverage at a lower cost. However, with the increase of cellular user, the shortage of capacity due to unbalanced call distribution and lack of QoS are common. This paper deals with dynamic sectorization for efficient resource management to solve load unbalancing among microcells in CDMA (Code Division Multiple Access) microcellular system. Dynamic load balancing can be effected by grouping micro-cells properly and grouping can be developed through a routing mechanism. Therefore, we use ants and their routes to choose the optimum grouping of micro-cells into sectors using Multiple Ant Colony System (MACS)in this paper.

• Biotechnology and Bioprocess Engineering:BBE
• /
• 제10권5호
• /
• pp.462-469
• /
• 2005

A top-down approach is known to be a useful and effective technique for the design and analysis of metabolic systems. In this Study, we have constructed a grouped metabolic network for Lactococcus lactis under aerobic conditions using grouped enzyme kinetics. To test the usefulness of grouping work, a non-grouped system and grouped systems were compared quantitatively with each other. Here, grouped Systems were designed as two groups according to the extent of grouping. The overall simulated flux values in grouped and non-grouped models had pretty similar distribution trends, but the details on flux ratio at the pyruvate branch point showed a little difference. This result indicates that our grouping technique can be used as a good model for complicated metabolic networks, however, for detailed analysis of metabolic network, a more robust mechanism Should be considered. In addition to the data for the pyruvate branch point analysis, Some major flux control coefficients were obtained in this research.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제23권12호
• /
• pp.677-683
• /
• 2017

Apache Storm이란 대표적인 실시간 분산 처리 시스템으로써, 분산 서버를 통해 실시간 데이터를 빠르게 처리하는 특징을 갖는다. 기존에 Storm은 다수의 서버에 트래픽을 분배할 때, 셔플(Shuffle) 그룹핑은 처리 지연 문제가 발생하며 이를 개선한 로컬(Local-or-Shuffle) 그룹핑은 트래픽이 특정 노드에 편중되는 문제가 발생할 수 있다. 본 논문은 이러한 기존 Storm 그룹핑에서 발생할 수 있는 문제를 해결하기 위한 지역성 고려(Locality-aware) 그룹핑을 제안한다. 실험에서는 제안하는 지역성 고려 그룹핑이 기존의 셔플 그룹핑 및 로컬 그룹핑에 비해 우수함을 확인하였다. 본 논문은 기존의 Storm의 한계점인 지역성과 로드 밸런싱을 동시에 고려한 우수한 결과라 사료된다.

• 한국멀티미디어학회논문지
• /
• 제8권4호
• /
• pp.544-553
• /
• 2005

본 논문에서는 사회적 상호작용을 극대화 하고, 중도 탈락률을 낮추며, 학습자의 고립감을 해소하기 위해 바람직한 온라인 학습공동체 형성을 위해 취향검사 항목을 개발 하였으며, 개발한 취향검사 항목을 동질성과 이질성으로 구분하여 항목을 개발하고, 이를 바탕으로 온라인상에서 지능적 에이전트 기법에 의한 학습공동체를 자동으로 그룹핑하는 e-learning 시스템(IIGS)을 개발하였다. 실제 1,000명을 대상으로 실험한 결과, 151개 그룹이 자동으로 생성되었으며 이중 34% 정도가 매우 높은 학습만족을 보여 그룹을 계속적으로 유지하고자 하였다.

• 제어로봇시스템학회논문지
• /
• 제17권8호
• /
• pp.739-746
• /
• 2011

Recently, recommender systems have been widely applied in E-commerce websites to help their customers find the items what they want. A recommender system should be able to provide users with useful information regarding their interests. The ability to immediately respond to the changes in user's preference is a valuable asset of recommender systems. This paper proposes a novel recommender system which aims to effectively adapt and respond to the immediate changes in user's preference. The proposed system combines IEC (Interactive Evolutionary Computation) with a content-based filtering method and also employs data grouping in order to improve time efficiency. Experiments show that the proposed system makes acceptable recommendations while ensuring quality and speed. From a comparative experimental study with an existing recommender system which uses the content-based filtering, it is revealed that the proposed system produces more reliable recommendations and adaptively responds to the changes in any given condition. It denotes that the proposed approach can be an alternative to resolve limitations (e.g., over-specialization and sparse problems) of the existing methods.

• 교육시설 논문지
• /
• 제18권3호
• /
• pp.3-11
• /
• 2011

This study aims at proposing some reform measures for the middle school grouping system in Gwangju Metropolitan City, which is divided 86 middle schools into 10 clusters and 3 school districts. In doing so, it analyzes the present status of educational environment and student walking distance in each school district such as the number of student per teacher, the student density, the school size and the gender ratio in class. And it conducts a survey of 5,363 middle school students, 3,966 parents and 1,007 teachers, also evaluates their satisfaction levels and needs with the student allocation system. As the result of the survey and data analysis, it finds out some problems in some school districts which are gender imbalance in class, the preference for private middle schools and inconvenience in commuting to school. To solve these problems, the study suggests the better alternatives to replace the current system. Firstly, to set up the basic fundamental principles detailed in 3 action plan, which emphasize the adherence to a close-range allocation, the appropriate size of school and class, and the equalization of educational environment. Secondly, to establish the information system for managing the school district in order to be more objective and transparent. Finally, it gives a concrete proposal which divides the 10th school grouping system into the 11th. The result would be expected to ease the gender imbalance and the concentration of private middle schools, to improve the student walking condition to school.

• Journal of information and communication convergence engineering
• /
• 제10권1호
• /
• pp.40-44
• /
• 2012

Data mining, which is attracting public attention, is a process of drawing out knowledge from a large mass of data. The key technique in data mining is the ability to maximize the similarity in a group and minimize the similarity between groups. Since grouping in data mining deals with a large mass of data, it lessens the amount of time spent with the source data, and grouping techniques that shrink the quantity of the data form to which the algorithm is subjected are actively used. The current grouping algorithm is highly sensitive to static and reacts to local minima. The number of groups has to be stated depending on the initialization value. In this paper we propose a gene algorithm that automatically decides on the number of grouping algorithms. We will try to find the optimal group of the fittest function, and finally apply it to a data mining problem that deals with a large mass of data.

• 산업경영시스템학회지
• /
• 제23권61호
• /
• pp.137-146
• /
• 2000

One of the problems faced in implementing cellular manufacturing systems is machine-part group formation. This paper proposes machine-part grouping algorithms based on Self-Organizing Map(SOM) neural networks and K-Means algorithm in cellular manufacturing systems. Although the SOM spreads out input vectors to output vectors in the order of similarity, it does not always find the optimal solution. We rearrange the input vectors using SOM and determine the number of groups. In order to find the number of groups and grouping efficacy, we iterate K-Means algorithm changing k until we cannot obtain better solution. The results of using the proposed approach are compared to the best solutions reported in literature. The computational results show that the proposed approach provides a powerful means of solving the machine-part grouping problem. The proposed algorithm Is applied by simple calculation, so it can be for designer to change production constraints.

• International Journal of Advanced Culture Technology
• /
• 제8권3호
• /
• pp.280-284
• /
• 2020

In the rapidly changing science and technology environment, an in-depth understanding of users of STEM information is an essential factor in designing a user-centered information system. The purpose of this study is to investigate and analyze in-depth the behaviors and needs of users who use STEM information. In this study, the needs of users for STEM information and STEM information sites are dealt with in depth using the FGI qualitative method. In addition, it includes the analysis results of grouping of similar sites according to various aspects of use of STEM information sites. As a result of grouping similar sites based on awareness and level of use,, they were grouped by domestic-international, paid-free, integrated-specific fields. As a result of grouping similar sites according to the purpose of use, they were grouped by domestic and international papers, research reports, and patents. As a result of grouping similar sites according to usage attributes, they were grouped by diversity, reliability, and specialization. As for the positions of similar sites perceived by users, Science Direct and PubMed showed high specialization and high quality, Google Scholar showed integration and popularity, and RISS showed four attributes evenly. Suggestions for information system design are discussed.