• Journal of Platform Technology
• /
• v.6 no.4
• /
• pp.34-40
• /
• 2018

Currently, artificial neural networks perform well for a single task, but NN have the problem of forgetting previous learning by learning other kinds of tasks. This is called catastrophic forgetting. To use of artificial neural networks in general purpose this should be solved. There are many efforts to overcome catastrophic forgetting. However, even though there was a lot of effort, it did not completely overcome the catastrophic forgetting. In this paper, we propose sequential iterative learning using core concepts used in elastic weight consolidation (EWC). The experiment was performed to reproduce catastrophic forgetting phenomenon using EMNIST data set which extended MNIST, which is widely used for artificial neural network learning, and overcome it through sequential iterative learning.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.31 no.2
• /
• pp.25-37
• /
• 2023

The purpose of this study is to develop and technically implement the design and scenario of cabin safety contents for virtual reality (VR)-based cabin safety learning for aviation service majors. The process for developing VR cabin safety learning contents consisted of a total of four stages: learning stage, research stage, verification stage, and application stage. The cabin safety scenario items for the production of VR learning contents reflected the occurrence of an emergency, the procedure for survival from impact, and the evacuation procedure from the aircraft as the core. For the technical implementation of learning contents, modeling work is conducted by checking scenario reviews, types and numbers of objects related to equipment and facilities, and items of interaction. In addition, the connection work with the actual metaverse platform is carried out to enable the utilization of the manufactured facilities and equipment objects. Finally, application tests were carried out to reconfirm supplementary items.

• Journal of information and communication convergence engineering
• /
• v.18 no.3
• /
• pp.176-182
• /
• 2020

In this paper, we propose a web-based open artificial intelligence (AI) platform which provides high convenience in input data pre-processing, artificial neural network training, and the configuration of subsequent operations according to inference results. The proposed platform has the advantages of the GUI-based environment which can be easily utilized by a user without complex installation. It consists of a web server implemented with the JavaScript Node.js library and a client running the tensorflow.js library. Using the platform, many users can simultaneously create, modify and run their projects to apply AI functionality into various smart services through an open web interface. With our implementation, we show the operability of the proposed platform. By loading a web page from the server, the client can perform GUI-based operations and display the results performed by three modules: the Input Module, the Learning Module and the Output Module. We also implement two application systems using our platform, called smart cashier and smart door, which demonstrate the platform's practicality.

• Journal of Power Electronics
• /
• v.12 no.5
• /
• pp.842-850
• /
• 2012

Learning Power Electronics is essential in both electrical and electronic engineering fields and the introductory courses are similar in many universities. Taking this premise into account, an educational computer-aided platform for power electronics will be presented in this paper. This educational platform includes an e-book, a set of power electronics animations, Java simulations, as well as several hands-on training sessions. The main advantages of this platform are twofold. First, all necessary teaching tools are combined on a single platform. And secondly, access to this platform is available free of charge and with no complicated registration requirements. In addition to traditional teaching techniques, the use of this platform has demonstrated an increase in student participation and has consistently improved their academic performance. Data consist of surveys, which guarantee both reliability and validity through psychometric techniques.

• International Journal of Advanced Culture Technology
• /
• v.10 no.3
• /
• pp.360-370
• /
• 2022

Demand for machine learning systems continues to grow, and cloud machine learning platforms are widely used to meet this demand. Recently, the performance improvement of the application processor of smartphones has become an opportunity for the machine learning platform to move from the cloud to On-Device AI, and mobile applications equipped with machine learning functions are required. In this paper, machine learning tools for mobile applications are investigated and compared the characteristics of these tools.

• The Journal of the Korea Contents Association
• /
• v.15 no.5
• /
• pp.36-46
• /
• 2015

The purpose of this study is to investigate the characteristics of diverse contents service of EBS providing on the multi-platform in the era of smart media. This study investigated contents service of EBS in order to seek and suggest EBS's competitiveness in the era of smart media environment. As a result, EBS is providing real-time service, VOD and e-learning, but its current contents service is concentrated on providing e-learning service. This study recommended that EBS need to reinforce contents suitable for various multi-platform uses. It also suggested EBS need to cooperate with diverse webs service and SNS service in order to widen openness of its platform, and to provide not only diverse contents, but also applications by offering mobile and on-demand contents and services. Finally, this study proposed that EBS is needed to move towards a more public and comprehensive media portal.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.11
• /
• pp.19-27
• /
• 2022

In this study, we propose a language and platform to describe and manage the MLOps(Machine Learning Operations) workflow for time series data anomaly detection. Time series data is collected in many fields, such as IoT sensors, system performance indicators, and user access. In addition, it is used in many applications such as system monitoring and anomaly detection. In order to perform prediction and anomaly detection of time series data, the MLOps platform that can quickly and flexibly apply the analyzed model to the production environment is required. Thus, we developed Python-based AI/ML Modeling Language (AMML) to easily configure and execute MLOps workflows. Python is widely used in data analysis. The proposed MLOps platform can extract and preprocess time series data from various data sources (R-DB, NoSql DB, Log File, etc.) using AMML and predict it through a deep learning model. To verify the applicability of AMML, the workflow for generating a transformer oil temperature prediction deep learning model was configured with AMML and it was confirmed that the training was performed normally.

• The Mathematical Education
• /
• v.63 no.2
• /
• pp.273-294
• /
• 2024

Recently, interest in Edu-Tech, which applies new technologies to the educational field, is growing. Edu-Tech is now being naturally used in schools, allowing both teachers and students to adapt to these changes. Particularly, there's significant attention on using Edu-Tech to bridge the educational gap through various teaching and learning strategies. This study focuses on the importance of self-directed task management by students for supplementary learning. It developed and utilized a math learning platform that enables teachers to easily provide and manage necessary tasks for students. Initially, the study developed "Sussam-MathTeacher" a problem-based learning application for middle school students, aimed at enhancing problem-solving abilities. This platform operates as a task management system, allowing teachers to assign or recommend problems to either the entire class or individual students. It aims to improve students' problem-solving abilities through a process that includes presenting necessary tasks, monitoring their own progress in solving problems, and self-assessing growth. Through this study, students demonstrated improved problem-solving skills by tackling tasks suited to their levels using "Sussam" highlighting the critical role of teachers in the digital educational environment.

• The Journal of the Korea Contents Association
• /
• v.21 no.12
• /
• pp.517-535
• /
• 2021

The development of digital technology has not only brought many changes to our lives, but also many changes to the online education environment. The emergence of micro-learning is to meet the needs of individual learners who hopes to receive personalized learning content immediately when they need it. Therefore, Micro-learning can be said to be 'adaptive' education. This research attempts to explore the development principles of adaptive micro-learning through literature research and case analysis. The results of the research draw four aspects of the development principles, including adaptive learning environment, adaptive learning content, adaptive learning sequence and adaptive learning evaluation, as well as detailed elements of each aspect. Micro-learning is a new form of e-learning that reflects the needs of the current society. As exploratory research, this research attempts to point out the direction for future follow-up research.

• Communications of Mathematical Education
• /
• v.36 no.1
• /
• pp.1-21
• /
• 2022

The purpose of this study is to analyze the current situation of EduTech, which is proposed as a way to build a flexible learning environment regardless of time and place according to the use of digital technology in mathematics subjects. The process of designing classes to use the EduTech platform, which is still in the development introduction stage, in public education is still difficult, and research to observe its effects and characteristics is also in its early stages. However, in the stage of preparing for future education, it is a meaningful process to grasp the current situation and point out the direction in preparation for the future in which EduTech will be actively applied to education. Accordingly, the current situation and utilization trends of EduTech at home and abroad were confirmed, and the functions and roles of EduTech platforms used in mathematics were analyzed. As a result of the analysis, the EduTech platform was pursuing learners' self-directed learning by constructing its functions so that they could be useful for individual learning of learners in hierarchical mathematics education. In addition, we have confirmed that the platform is evolving to be useful for teachers' work reduction, suitable activities, and evaluations learning management. Therefore, it is necessary to implement instructional design and individual customized learning support measures for students that can efficiently utilize these platforms in the future.