• Journal of Digital Contents Society
• /
• v.10 no.1
• /
• pp.73-85
• /
• 2009

In this paper, we propose integrated guidelines of effective e-Learning management system based on LMS(Learning Management System), LCMS(Learning Contents Management System), Authoring Tools by linking e-Learning system to management. The purpose of the study is to describe the present condition and major characteristics of domestic e-Learning System and to recommend quality assurance related to main functions and practical use of the corporate e-learning in multiple ways and implementation strategies to fit facilitator, tutor, learner.

• Korean Chemical Engineering Research
• /
• v.61 no.3
• /
• pp.388-393
• /
• 2023

The analysis of engineering data using neural network based on supervised learning has been utilized in various engineering fields such as optimization of chemical engineering process, concentration prediction of particulate matter pollution, prediction of thermodynamic phase equilibria, and prediction of physical properties for transport phenomena system. The supervised learning requires training data, and the performance of the supervised learning is affected by the composition and the configurations of the given training data. Among the frequently observed engineering data, the data is given in log-scale such as length of DNA, concentration of analytes, etc. In this study, for widely distributed log-scaled training data of virtual 100×100 images, available loss functions were quantitatively evaluated in terms of (i) confusion matrix, (ii) maximum relative error and (iii) mean relative error. As a result, the loss functions of mean-absolute-percentage-error and mean-squared-logarithmic-error were the optimal functions for the log-scaled training data. Furthermore, we figured out that uniformly selected training data lead to the best prediction performance. The optimal loss functions and method for how to compose training data studied in this work would be applied to engineering problems such as evaluating DNA length, analyzing biomolecules, predicting concentration of colloidal suspension.

• The Journal of the Korea Contents Association
• /
• v.21 no.3
• /
• pp.616-625
• /
• 2021

Deep neural networks are an approximation method that approximates an arbitrary function to a linear model and then repeats additional approximation using a nonlinear active function. In this process, the method of evaluating the performance of approximation uses the loss function. Existing in-depth learning methods implement approximation that takes into account loss functions in the linear approximation process, but non-linear approximation phases that use active functions use non-linear transformation that is not related to reduction of loss functions of loss. This study proposes parametric activation functions that introduce scale parameters that can change the scale of activation functions and location parameters that can change the location of activation functions. By introducing parametric activation functions based on scale and location parameters, the performance of nonlinear approximation using activation functions can be improved. The scale and location parameters in each hidden layer can improve the performance of the deep neural network by determining parameters that minimize the loss function value through the learning process using the primary differential coefficient of the loss function for the parameters in the backpropagation. Through MNIST classification problems and XOR problems, parametric activation functions have been found to have superior performance over existing activation functions.

• The Mathematical Education
• /
• v.62 no.3
• /
• pp.303-326
• /
• 2023

Mathematics is a discipline with a strong systemic structure, and learning deficits in previous stages have a great influence on the next stages of learning. Therefore, it is necessary to frequently check whether students have learned well and to provide immediate feedback, and for this purpose, intelligent tutoring system(ITS) can be used in math education. For this reason, it is necessary to reveal how the intelligent tutoring system is effective in personalized adaptive learning. The purpose of this study is to investigate the functions and applications of intelligent tutoring system for personalized adaptive learning in mathematics. To achieve this goal, literature reviews and surveys with students were applied to derive implications. Based on the literature reviews, the functions of intelligent tutoring system for personalized adaptive learning were derived. They can be broadly divided into diagnosis and evaluation, analysis and prediction, and feedback and content delivery. The learning and lesson plans were designed by them and it was applied to fifth graders in elementary school for about three months. As a result of this study, intelligent tutoring system was mostly supporting personalized adaptive learning in mathematics in several ways. Also, the researcher suggested that more sophisticated materials and technologies should be developed for effective personalized adaptive learning in mathematics by using intelligent tutoring system.

• International Journal of Advanced Culture Technology
• /
• v.10 no.4
• /
• pp.412-419
• /
• 2022

The amount of digital text data is growing exponentially, and many machine learning solutions are being used to monitor and manage this data. Artificial intelligence and machine learning are used in many areas of our daily lives, but the underlying processes and concepts are not easy for most people to understand. At a time when many experts are needed to run a machine learning solution, no-code machine learning tools are a good solution. No-code machine learning tools is a platform that enables machine learning functions to be performed without engineers or developers. The latest No-Code machine learning tools run in your browser, so you don't need to install any additional software, and the simple GUI interface makes them easy to use. Using these platforms can save you a lot of money and time because there is less skill and less code to write. No-Code machine learning tools make it easy to understand artificial intelligence and machine learning. In this paper, we examine No-Code machine learning tools and compare their features.

• Journal of The Korean Association of Information Education
• /
• v.15 no.4
• /
• pp.579-587
• /
• 2011

In this study, we developed the filed-experience learning support application. Because teachers want to use LBS in education area. In order to select fit functions, we carried out survey about functions that teachers want. Then, we analysed the result of survey and implemented the functions. Also, for survey about application's effectiveness, we selected elementary school teachers. Then we demonstrated and explained the application to them. The result of survey about application's effectiveness shows that application have higher utilization in education area.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.3
• /
• pp.107-115
• /
• 2006

This paper presents a new class of activation functions for Cascade Correlation learning algorithm, which herein will be called CosGauss function. This function is a cosine modulated gaussian function. In contrast to the sigmoidal, hyperbolic tangent and gaussian functions, more ridges can be obtained by the CosGauss function. Because of the ridges, it is quickly convergent and improves a pattern recognition speed. Consequently it will be able to improve a learning capability. This function was tested with a Cascade Correlation Network on the two spirals problem and results are compared with those obtained with other activation functions.

• Journal of Multimedia Information System
• /
• v.9 no.1
• /
• pp.21-32
• /
• 2022

Early-stage diagnosis of Alzheimer's Disease (AD) from Cognitively Normal (CN) patients is crucial because treatment at an early stage of AD can prevent further progress in the AD's severity in the future. Recently, computer-aided diagnosis using magnetic resonance image (MRI) has shown better performance in the classification of AD. However, these methods use a traditional machine learning algorithm that requires supervision and uses a combination of many complicated processes. In recent research, the performance of deep neural networks has outperformed the traditional machine learning algorithms. The ability to learn from the data and extract features on its own makes the neural networks less prone to errors. In this paper, a dense neural network is designed for binary classification of Alzheimer's disease. To create a classifier with better results, we studied result of different activation functions in the prediction. We obtained results from 5-folds validations with combinations of different activation functions and compared with each other, and the one with the best validation score is used to classify the test data. In this experiment, features used to train the model are obtained from the ADNI database after processing them using FreeSurfer software. For 5-folds validation, two groups: AD and CN are classified. The proposed DNN obtained better accuracy than the traditional machine learning algorithms and the compared previous studies for AD vs. CN, AD vs. Mild Cognitive Impairment (MCI), and MCI vs. CN classifications, respectively. This neural network is robust and better.

• Journal of Internet Computing and Services
• /
• v.23 no.5
• /
• pp.17-23
• /
• 2022

Network slicing is a promising paradigm and significant evolution for adjusting the heterogeneous services based on different requirements by placing dynamic virtual network functions (VNF) forwarding graph (VNFFG) and orchestrating service function chaining (SFC) based on criticalities of Quality of Service (QoS) classes. In system architecture, software-defined networks (SDN), network functions virtualization (NFV), and edge computing are used to provide resourceful data view, configurable virtual resources, and control interfaces for developing the modified deep reinforcement learning agent (MDRL-A). In this paper, task requests, tolerable delays, and required resources are differentiated for input state observations to identify the non-critical/critical classes, since each user equipment can execute different QoS application services. We design intelligent slicing for handing the cross-domain resource with MDRL-A in solving network problems and eliminating resource usage. The agent interacts with controllers and orchestrators to manage the flow rule installation and physical resource allocation in NFV infrastructure (NFVI) with the proposed formulation of completion time and criticality criteria. Simulation is conducted in SDN/NFV environment and capturing the QoS performances between conventional and MDRL-A approaches.

• The Korean Journal of Applied Statistics
• /
• v.27 no.3
• /
• pp.445-459
• /
• 2014

As a certain job is repeatedly done by a worker, the outcome comparative to the effort to complete the job gets more remarkable. The outcome may be the time required and fraction defective. This phenomenon is referred to a learning-curve effect. We focus on the parametric modeling of the learning-curve effects on count data using a logistic cumulative distribution function and some probability mass functions such as a Poisson and negative binomial. We conduct various simulation scenarios to clarify the characteristics of the proposed model. We also consider a real application to compare the two discrete-type distribution functions.