• Journal of Platform Technology
• /
• v.11 no.4
• /
• pp.71-79
• /
• 2023

In this paper, we analyze the characteristics of machine learning workloads and, based on them, propose a distributed in-memory caching technique to improve the performance of machine learning workloads. The core of machine learning workload is model training, and model training is a computationally intensive task. Performing machine learning workloads in a Kubernetes-based cloud environment in which the computing framework and storage are separated can effectively allocate resources, but delays can occur because IO must be performed through network communication. In this paper, we propose a distributed in-memory caching technique to improve the performance of machine learning workloads performed in such an environment. In particular, we propose a new method of precaching data required for machine learning workloads into the distributed in-memory cache by considering Kubflow pipelines, a Kubernetes-based machine learning pipeline management tool.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.23 no.5
• /
• pp.438-444
• /
• 2013

This paper presents the replication of a desired vibration response by iterative learning control (ILC) system for a vibration motion replication actuator. The vibration motion replication actuator has parameter uncertainties including system nonlinearity and joint nonlinearity. Vehicle manufacturers worldwide are increasingly relying on road simulation facilities that put simulated loads and stresses on vehicles and subassemblies in order to reduce development time. Road simulation algorithm is the key point of developing road simulation system. With the rapid progress of digital signal processing technology, more complex control algorithms including iterative learning control can be utilized. In this paper, ILC algorithm was utilized to produce simultaneously the six channels of desired responses using the Stewart platform composed of six linear electro-magnetic actuators with Halbach magnet array. The convergence rate and accuracy showed reasonable results to meet the requirement. It shows that the algorithm is acceptable to replicate multi-channel vibration responses.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.974-992
• /
• 2021

Recently, the healthcare field has undergone rapid changes owing to the accumulation of health big data and the development of machine learning. Data mining research in the field of healthcare has different characteristics from those of other data analyses, such as the structural complexity of the medical data, requirement for medical expertise, and security of personal medical information. Various methods have been implemented to address these issues, including the machine learning model and cloud platform. However, the machine learning model presents the problem of opaque result interpretation, and the cloud platform requires more in-depth research on security and efficiency. To address these issues, this paper presents a recent technology for Internet-of-Things-based (IoT-based) health big data processing. We present a cloud-based IoT health platform and health big data processing technology that reduces the medical data management costs and enhances safety. We also present a data mining technology for health-risk prediction, which is the core of healthcare. Finally, we propose a study using explainable artificial intelligence that enhances the reliability and transparency of the decision-making system, which is called the black box model owing to its lack of transparency.

• Journal of Digital Contents Society
• /
• v.19 no.7
• /
• pp.1349-1355
• /
• 2018

Recently, a variety of contents services over the internet are becoming popular, among which MCN(Multi Channel Network) platform services have become popular with the generalization of smart phones. The MCN platform is based on streaming, and various factors are added to improve the service. Among them, augmented reality sticker service using face recognition is widely used. In this paper, we implemented the MCN platform that masks the augmented reality sticker on the face through facial emotion recognition in order to further increase the interest factor. We analyzed seven facial emotions using deep learning technology for facial emotion recognition, and applied the emotional sticker to the face based on it. To implement the proposed MCN platform, emotional stickers were applied to the clients and various servers that can stream the servers were designed.

• Communications of the Korean Institute of Information Scientists and Engineers
• /
• v.34 no.3
• /
• pp.31-34
• /
• 2016

This article reviews Apache SINGA, a general distributed deep learning (DL) platform. The system components and its architecture are presented, as well as how to configure and run SINGA for different types of distributed training using model/data partitioning. Besides, several features and performance are compared with other popular DL tools.

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.9
• /
• pp.35-43
• /
• 2021

This study aims to investigate factors affecting eBook acceptance and learning outcomes among students experiencing online distance learning. As conventional textbooks are now switched into eBooks, the effects of contextual factors including lecturer, student computer competency, content and design of the course, access ability, infrastructure, and university support on eBook acceptance and learning outcome needs to be evaluated. The sample of this study is represented by students at the Universiti Teknologi MARA, City Campus Melaka, undertaking 'strategic management course'. Non-probability random sampling was selected as the sampling technique and a purposive sampling method was chosen to select the samples. The samples comprised 171 students randomly selected through Google Form. The questionnaire data was analyzed by using PLS-SEM. The results indicated that these factors contributed 62.3% variations in the eBook acceptance and 67.1% variations in the learning outcomes. The strongest factor affecting both dependent variables was content and design of course. Managerial implication suggested that the content for all courses taught through the eBook platform needs to be revisited and improved in accordance with the mode of online deliverance. Tutorial on how to navigate the eBook platform is important to all users as this would enhance acceptance and produce better learning outcomes among students.

• Journal of Practical Engineering Education
• /
• v.11 no.1
• /
• pp.75-86
• /
• 2019

Globally, the change in higher education is gradually moving toward a trend that seeks a change in innovative higher education through the revitalization of digital-based education. Accordingly, this study designed a customized education model based on e-learning that can be used in undergraduate education and development of lifelong vocational skills. The use of online learning platforms and the expansion of education are major factors that change the overall higher education system as the form and content of curriculum changes around the world. In order to establish a customized education model using online learning platform, this study analyzed major overseas advanced education cases and selected the basic direction of customized learning as personalized learning, competency based learning, and training for talents leading the 4th Industrial Revolution. Then, FGI was conducted for undergraduate and lifelong vocational ability development experts. As a result, a customized education model using an online learning platform was derived from a degree-type model available in undergraduate education and a non-degree-type model available in the field of lifelong vocational ability development, and each operation strategy was suggested.

• Communications of Mathematical Education
• /
• v.36 no.3
• /
• pp.417-438
• /
• 2022

The purpose of this study is to suggest implications for mathematics teaching and learning when using AI-based educational platforms that support personalized mathematics learning. To this end, we selected five platforms(Knock-knock! Math Expedition, knowre, Khan Academy, MATHia, CENTURY) and analyzed how the AI-based educational platforms for mathematics reflect the three elements(PLP, PLN, PLE) to support personalized learning. The results of this study showed that although the characteristics of PLP, PLN, and PLE implemented on each platform varied, they were designed to form PLEs that allow learners to make their autonomous decisions about learning based on PLP and PLN. The significance of this study can be found in that it has improved the understanding and practicability of personalized mathematics learning with the AI-based educational platforms.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2004.05b
• /
• pp.199-202
• /
• 2004

In order to efficiently provide the learning ability for engineering education, electromagnetic field simulator have been developed on lava 2 platform. Each simulation module based on lava applet can be easily utilized with independent platform and provide GUI environment to set up physical conditions. The numerical results using computer animation technique are provided in real time. To aggregate and manage the learning objects, application of SCORM which is e-Learning standards is also studied.

• The Mathematical Education
• /
• v.59 no.4
• /
• pp.373-387
• /
• 2020

The purpose of this study is to verify the effects of personalized learning platforms applied with machine learning recommendation modules that upgrade recommended algorithms by themselves through learning big data analysis on students' learning time, self-directed learning ability, mathematics achievement, and attitudes toward mathematics, and the correlation between them. According to the study, customized learning affected learning time, self-directed learning ability and mathematics attitude, while learning time affected self-directed learning ability. Self-directed learning ability has had a significant impact on the attitude of mathematics and mathematical achievements. As a result of the mediated effectiveness test, the indirect impact of customized learning on mathematics attitude and mathematics performance was significant through the medium of learning time and self-directed learning ability.