• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.4
• /
• pp.21-34
• /
• 2023

Federated Learning is divided into synchronous federated learning and asynchronous federated learning. Asynchronous federated learning has a time advantage over synchronous federated learning, but asynchronous federated learning still has some challenges to obtain better performance. In particular, preventing performance degradation in non-IID training datasets, selecting appropriate clients, and managing stale gradient information are important for improving model performance. In this paper, we deal with K-asynchronous federated learning by using non-IID datasets. In addition, unlike traditional method using static K, we proposed an algorithm that adaptively adjusts K and we can reduce the learning time. Additionally, the we show that model performance is improved by using stale gradient handling method. Finally, we use a method of judging model performance to obtain strong model stability. Experiment results show that overall algorithm can obtain advantages of reducing training time, improving model accuracy, and improving model stability.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.4
• /
• pp.823-831
• /
• 2021

In the Internet age, the digital cultural industry is booming, start-up companies want to win a place in the fiercely competitive red sea market of the Internet, innovative business models have become an important challenge and opportunity for countless companies. Taking the virtual idol "Hatsune Miku" as the example, the analysis is carried out at the levels of technological innovation, value proposition, resource combination, operation process and marketing model. By comparing the differences between the first-generation software Meiko of Crypton and the second-generation software "Hatsune Miku", we will study the components of the business model to explore the characteristics and advantages of this business model, and in-depth operation models and innovations. The analysis is expected to provide a substantial reference for other electronic product and software development companies in their business models.

• Journal of The Korean Association of Information Education
• /
• v.21 no.1
• /
• pp.139-149
• /
• 2017

Artificial intelligence education is very important in the 21st century knowledge information society. Even if it is very important to understand artificial intelligence and practice computer programming in computer education in the fourth industrial revolution, but there is no teaching and learning model to understand artificial intelligence and computer programming education. In this paper, the proposed model consists of problem understanding step, data organizing step, artificial intelligence model setting step, programming step, and report writing step. At the program step, students can choose to copy, transform, create, and challenge steps to their level. In this study, the validity of the model was proved by the Delphi evaluation of elementary school teachers. The results of this study provide a good opportunity for elementary school students to practice artificial intelligence programs.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.1
• /
• pp.287-294
• /
• 2017

BIM technology has been used in the domestic AEC field since the middle 2000s. BIM has proved its worth in cutting-edge buildings, mega-buildings and freeform buildings in particular. Many freeform buildings could not be completed due to the low level of construction technique. However, many successful cases emerged after adopting digital technology, including BIM which encouraged architects to challenge freeform designs. The modeling software that can generate the freeform shape are not usually able to build the efficient BIM data type in the AEC industry. In this study a process model of the parametric freeform construction member generation and conversion to BIM data is shown and the prototype system is demonstrated.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.68-68
• /
• 2021

Sesan강과 Srepok강은 베트남, 캄보디아, 라오스가 공유하는 3S강 유역 (Sesan강, Srepok강, Sekong강)의 일부로 연구 및 관리된다. 3S강 유역은 Mekong강의 중요한 지류이며 Mekong강 유역의 상당 부분을 구성한다(Mekong강 유역 면적의 10%, 연간 총 유출량의 20%). 베트남 측 Sesan강 유역의 면적은 11,255km²이고 Srepok강 유역 면적은 18,162km²이다. Sesan강과 Srepok 강의 상류는 베트남 중부 고원의 긴 산맥에 위치하고 있다. Sesan강과 Srepok강 유역은 기후변화에 따른 홍수, 가뭄, 어업 지속 가능성 감소, 퇴적 등 많은 문제와 도전에 직면 할 것으로 예측되고 있다. 본 연구에서는 World Bank의 "Viet Nam Mekong Integrated Water Resources Management (M-IWRM) Project의 일환으로 베트남 정부 차원에서 처음으로 구축한 수자원관리 의사결정지원 시스템인 "DSS-2S"를 활용하여, Sesan-Srepok강 유역의 강둑 침식 위험성을 분석하였다. DSS-2S는 MIKE Hydro Basin을 기반으로 SWAT모델, 수리모델, 하상변동 모델, 및 수질모델 등과 연계 하여 구축되었다. 2030 년을 목표 연도로 설정하고, 기후 변화 시나리오와 사회 경제적 발전을 기반으로 DSS-2S에 포함되어 있는 유사 이송 및 수리학적 모델을 활용하여 주요 하천 단면에서의 평균 유속과 하상 침식 양을 예측하였다. 유속 및 심부 침식 기준에 근거하여 강둑 침식 위험성을 분석하였다. 모델의 시뮬레이션 결과를 기반으로 강둑 침식 위험이 있는 강 구간은 고(高)유속과 높은 침식의 조합에 의해 결정되었다. 고위험 침식 예상지는 Sesan강 유역의 Dak Bla, Po Ko, 및 Se San강에 총 길이 73.5km에 걸쳐 발생 할 것으로 분석되었으며, 침식 위험이 매우 높은 지역은 Dak Bla 강에 총 길이 2,286m, Po Ko 강에 총 길이 5,096m 정도가 발생 하는 것으로 분석되었다. 강둑 세국을 유발할 수 있는 다양한 인자들을 고찰하였으며, 본 성과는 베트남 중앙 정부의 장기수 자원 종합계획 수립의 기본 자료로 활용 될 예정이다.

• Journal of Technology Innovation
• /
• v.31 no.2
• /
• pp.29-58
• /
• 2023

The knowledge and technological competitiveness produced by universities determines a country's scientific technology and industrial competitiveness. The aim of this study is to find the evolution of Triple Helix Model (THM) by analyzing cases of technological catching-up and nurturing the human capital policies during Korea industrialization era, from 1960 to 2020. To fulfill the aims, this study utilized qualitative methodologies. In this study not only the process of acquiring technologies and core technologies but also the system for nurturing the human capital focused on scientific and engineers was analyzed. In addition characteristics of each ear of Triple Helix Model (THM) were examined. From the results, this study suggests Triple Helix Syncro-Scenario (THSS) which is the compelling innovation model, based on scenario, combined the personal experiences various critical thoughts and problems with enterprises and government for creating the incessment innovations.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2016.02a
• /
• pp.218-218
• /
• 2016

실린더 형태의 유전체 관에 나선형으로 도전체 안테나를 설치하는 타입의 유도 결합 플라즈마원은 간단한 구조로 화학 조성 분석용부터 나노 분말 제조, 반도체용 식각/증착, 표면 처리, 자동차 및 일반 산업 부품용 증착 보조원등으로 널리 사용되고 있다. 고밀도 라디칼/이온의 공급을 위해서 투입 전력을 증가시키는 경우 높은 전력 밀도로 인해서 유전체 관에 인가되는 열응력이 대기압 및 관 고정용 구조물에 의한 구조 응력에 더해져서 파손에 이르는 경우가 발생될 수 있다. 실제 실린더 길이 전체를 안테나 코일로 감는 경우에도 플라즈마 발생 밀도가 높은 지역은 중심 일부 영역에 국한 되는 공정 영역도 있어서 이에 대한 분석이 필요하다. CFD-ACE+를 이용하여 플라즈마의 생성, 냉각수의 열전도, 외부 공냉식 팬의 역할등에 대해서 수치 모델을 작성하여 검토하였다. 나선형 냉각코일의 경우 냉각수량을 일정값 이상으로 증가시키는 경우 유속이 지나치게 빨라져서 열원이 있는 내경쪽 표면에서 열전도가 유속에 비례해서 증가하지 못하는 단점이 발생할 수 있으며 냉각팬의 경우 일반적으로 장치 내부에 대해서만 모델링을 하는 데 실제로 전체 시스템의 주변에서 공기의 흐름을 넓게 해석해야 실제 냉각 효과를 파악할 수 있다. 심한 경우 냉각용 공기 흡입구와 토출구의 간격이 좁아서 열원에 의해서 가열된 공기의 상당량이 다시 냉각용 공기 흡입구로 재순환 되는 경우도 발생하기 쉽다.

• The Journal of the Korea Contents Association
• /
• v.13 no.6
• /
• pp.67-76
• /
• 2013

By integrating of intrinsic motivation theory and flow theory, this study sought to investigate the causal linkage among optimal balance of video game player's perceived skill and challenge, flow experience, intrinsic motivation and video game genre usage. A path analysis showed that the different types of video game usage were directly influenced by players' intrinsic motivations, which were also reinforced by players' flow experiences of enjoyment. However, the path model with a multiple group approach provided no strong evidence for gender differences in the relationships between the variables of interest.

• Journal of Information Management
• /
• v.37 no.3
• /
• pp.85-97
• /
• 2006

In a Digital Age contents are comprized of knowledge and information. Also distinction between the sinner and loser is very clear thus only those who can rise up to challenge of changes can become the strong and a principal player. In this study this author presented a model for forecasting information service program(FISP) which is devised for the purpose of innovation of service pattern related to providing information which is ultimate goal of a library. This model is an innovative strategy escaping boldly from negative and inactive information service pattern known in analog age. Through this model this author attempted to outline methodology for heightening ideal of professional librarians and for assuring success in organizational system.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.11
• /
• pp.2032-2038
• /
• 2006

The recognition of human gestures in image sequences is an important and challenging problem that enables a host of human-computer interaction applications. This paper describes a gesture recognition algorithm based on the particle filters, namely CONDENSATION. The particle filter is more efficient than any other tracking algorithm because the tracking mechanism follows Bayesian estimation rule of conditional probability propagation. We used two models for the evaluation of particle Inter and apply the MATLAB for the preprocessing of the image sequence. But we implement the particle filter using the C++ to get the high speed processing. In the experimental results, it is demonstrated that the proposed algorithm prove to be robust in the cluttered environment.