• 지식경영연구
• /
• 제21권2호
• /
• pp.119-135
• /
• 2020

Owing to pandemic (COVID-19), the traditional face-to-face education method has been changed to the non-face-to-face real-time online education methods. Using a real time-based video conference system, synchronous education can be adopted by face-to-face class easily. Specially, it is very important to minimize the difference in learning effects between face-to-face and non-face-to-face in Entrepreneurship education. In this study, in order to derive the factors that affect the satisfaction of learners in synchronous online education, authors collected data from learners taking a synchronous entrepreneurship course. Through previous research, learned the reality of education and the composition of lessons. Spatiotemporal effectiveness, mentor ability, and educational environment influence learning satisfaction. PLS-SEM results revealed that it was confirmed that only spatiotemporal effects affect learner satisfaction. However, the education environment (fluent operation and convenience of function use of real-time based online conference system) effect teaching presence, class structure, and spatiotemporal effects. Through this research, we hope to provide theoretical and practical support for developing effective teacher activities, proper lesson structure, convenient function of the conference system, and learner-centered online learning environment when developing synchronous online classes.

• 한국지능시스템학회논문지
• /
• 제13권3호
• /
• pp.316-321
• /
• 2003

최근들어, 서포트 벡터 학습은 패턴 분류, 함수 근사 및 비정상 상태 탐지 등의 분야에서 상당한 관심을 끌고 있다. 여러가지 서포트 벡터 학습 방법들 중 누-버전(nu-versions)으로 불리는 방법들은 서포트 벡터의 개수를 제어해야할 필요가 있는 경우에는 특히 유용한 것으로 알려져 있다. 본 논문에서는, $\nu-SVR$로 불리는 누-버전 서포트 벡터 학습 방법과 미리 정해진 기저함수를 모두 활용하는 함수 근사 문제를 고려한다. $\varepsilon-SVR$, $\nu-SVR$ 및 세미-파라메트릭 함수 근사 방법론등을 복습한 후에, 본 논문은 정해진 기저함수를 이용할 수 있는 방향으로 기존의 $\nu-SVR$ 방법을 확장하는 방안을 제시한다. 그리고, 제안된 방법의 적용가능성이 예제를 통하여 보여진다.

• 한국초등과학교육학회지:초등과학교육
• /
• 제38권3호
• /
• pp.387-394
• /
• 2019

The purpose of this study was to investigate the effects of writing science diary on science learning motivation, science academic achievement and ecological sensitivity of elementary students. Teaching unit was focused on 'The structure and function of plants' in 6th grade science text book. The subjects of study were 51 students of two classes. One class of 25 students, experimental group, wrote science diaries as homework. While the other class of 26 students, comparative group, performed homeworks with fill-in-the-blank worksheets. The results of this study can be summarized as follows: First, writing science diary had a meaningful effect on improvement of science learning motivation. Second, writing science diary had a meaningful effect on improvement of science academic achievement. Third, writing science diary had a meaningful effect on improvement of ecological sensitivity. And we could find that students had a favorable impression and high satisfaction level about writing science diary from the questionnaire.

• 전기전자학회논문지
• /
• 제25권1호
• /
• pp.139-147
• /
• 2021

본 논문에서는 저조도 영상을 개선하기 위한 영상 분해 기반 심층 학습 방법 및 분해 채널 특성에 따른 손실함수를 제안한다. 기존 기법들의 문제점인 색신호 왜곡 및 할로 현상을 제거하기 위해, 입력 영상의 휘도 채널을 반사 성분과 조도 성분으로 분해하고, 반사 성분, 조도 성분 및 색차 신호를 신호 특성에 적합한 심층학습 과정을 적용하는 분해 기반 다중 구조 심층 학습 방법을 제안한다. 더불어, 분해 채널들의 특성에 따른 혼합 놈 기반의 손실함수를 정의하여 복원 영상의 안정성을 증대하고 열화 현상을 제거하기 위한 기법에 대해 기술한다. 실험 결과를 통해 제안한 방법이 다양한 저조도 영상을 효과적으로 개선하였음을 확인할 수 있었다.

• International Journal of Internet, Broadcasting and Communication
• /
• 제14권1호
• /
• pp.136-141
• /
• 2022

A very complex task in deep learning such as image classification must be solved with the help of neural networks and activation functions. The backpropagation algorithm advances backward from the output layer towards the input layer, the gradients often get smaller and smaller and approach zero which eventually leaves the weights of the initial or lower layers nearly unchanged, as a result, the gradient descent never converges to the optimum. We propose a two-factor non-saturating activation functions known as Bea-Mish for machine learning applications in deep neural networks. Our method uses two factors, beta (𝛽) and alpha (𝛼), to normalize the area below the boundary in the Mish activation function and we regard these elements as Bea. Bea-Mish provide a clear understanding of the behaviors and conditions governing this regularization term can lead to a more principled approach for constructing better performing activation functions. We evaluate Bea-Mish results against Mish and Swish activation functions in various models and data sets. Empirical results show that our approach (Bea-Mish) outperforms native Mish using SqueezeNet backbone with an average precision (AP50val) of 2.51% in CIFAR-10 and top-1accuracy in ResNet-50 on ImageNet-1k. shows an improvement of 1.20%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권6호
• /
• pp.1545-1559
• /
• 2023

In the cloud environment, microservices are implemented through Kubernetes, and these services can be expanded or reduced through the autoscaling function under Kubernetes, depending on the service request or resource usage. However, the increase in the number of nodes or distributed microservices in Kubernetes and the unpredictable autoscaling function make it very difficult for system administrators to conduct operations. Artificial Intelligence for IT Operations (AIOps) supports resource management for cloud services through AI and has attracted attention as a solution to these problems. For example, after the AI model learns the metric or log data collected in the microservice units, failures can be inferred by predicting the resources in future data. However, it is difficult to construct data sets for generating learning models because many microservices used for autoscaling generate different metrics or logs in the same timestamp. In this study, we propose a cloud data refining module and structure that collects metric or log data in a microservice environment implemented by Kubernetes; and arranges it into computing resources corresponding to each service so that AI models can learn and analogize service-specific failures. We obtained Kubernetes-based AIOps learning data through this module, and after learning the built dataset through the AI model, we verified the prediction result through the differences between the obtained and actual data.

• International Journal of Computer Science & Network Security
• /
• 제24권1호
• /
• pp.1-8
• /
• 2024

In contemporary education, the rapid advancement of digital technologies elevates demands for integrating the latest tools into the learning process. Mathematical analysis, as a discipline, benefits from computer mathematics in distance education, enhancing practical aspects and enabling individualized learning. This article addresses the integration of the Maple computer mathematics system into higher education, specifically in teaching "Mathematical Analysis." Emphasizing its role in distance learning, computer mathematics optimizes the educational environment, reducing the time required for knowledge acquisition. The article showcases the application of Maple in finding extremum points and introduces an educational software simulator, enabling students to practice the method. The simulator, developed within Maple, facilitates self-checking and enhances the study of functions. Conclusions drawn from the study highlight the positive impact of these tools on distance education, affirming Maple's role in enhancing professional training and information culture among higher education students.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제14권2호
• /
• pp.73-83
• /
• 2014

Many recent theoretical developments in the field of machine learning and control have rapidly expanded its relevance to a wide variety of applications. In particular, a variety of portfolio optimization problems have recently been considered as a promising application domain for machine learning and control methods. In highly uncertain and stochastic environments, portfolio optimization can be formulated as optimal decision-making problems, and for these types of problems, approaches based on probabilistic machine learning and control methods are particularly pertinent. In this paper, we consider probabilistic machine learning and control based solutions to a couple of portfolio optimization problems. Simulation results show that these solutions work well when applied to real financial market data.

• 컴퓨터교육학회논문지
• /
• 제6권3호
• /
• pp.75-85
• /
• 2003

본 연구는 SCORM 기반 학습 객체의 메타데이타 생성 즉 Asset, SCO, Contents Aggregation과 Contents Package에 대한 메타데이터를 생성하는 시스템(MetaGene)을 개발한다. SCORM 을 지원하는 LMS내 API 어댑터와 인터페이스를 위한 학습 객체 내에 API 활성화 함수를 내장시키고, 데이터 모델을 기반으로 학습 과정을 트래킹 하는 코드도 포함 시킨다. 또한 학습 객체들이 LMS에 전송되게 PIF(Package Interchange File)로 패키징 시킨다. MetaGene에 생성된 학습객체의 메타데이터와 컨텐츠 패키지의 manifest file을 $SCORM^{(TM)}$ Conformance Testsuite을 이용하여 유효성을 검증한다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅲ
• /
• pp.1239-1242
• /
• 2003

The neural networks are a famous model to learn the nonlinear function or nonlinear system. The main point of neural network is that the difference actual output from desired output is used to update weights. Usually, the gradient descent method is used for the learning process. On training process, if learning rate is too large, neural networks hardly guarantee convergence of neural networks. On the other hand, if learning rate is too small, the training spends much time. Therefore, one major problem in use of neural networks are to decrease the teaming time while neural networks are guaranteed convergence. In this paper, we suggest the model of fuzzy logic to neural networks to calibrate learning rate. This method is to tune learning rate dynamically according to error and demonstrates the optimization of training.