• Journal of the Korea Society of Computer and Information
• /
• v.23 no.11
• /
• pp.25-30
• /
• 2018

Pattern classification system is often an important component of intelligent systems. In this paper, we present a pattern classification system consisted of the feature selection module, knowledge base construction module and decision module. We introduce a feature impact evaluation selection method based on fuzzy cluster analysis considering computational approach and generalization capability of given data characteristics. A fuzzy neural network, OFUN-NET based on unsupervised learning data mining technique produces knowledge base for representative clusters. 240 blemish pattern images are prepared and applied to the proposed system. Experimental results show the feasibility of the proposed classification system as an automating defect inspection tool.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.920-924
• /
• 2005

This paper presents adaptive learning data of evolvable neural networks (ENNs) for time series prediction of nonlinear dynamic systems. ENNs are a special class of neural networks that adopt the concept of biological evolution as a mechanism of adaptation or learning. ENNs can adapt to an environment as well as changes in the environment. ENNs used in this paper are L-system and DNA coding based ENNs. The ENNs adopt the evolution of simultaneous network architecture and weights using indirect encoding. In general just previous data are used for training the predictor that predicts future data. However the characteristics of data and appropriate size of learning data are usually unknown. Therefore we propose adaptive change of learning data size to predict the future data effectively. In order to verify the effectiveness of our scheme, we apply it to chaotic time series predictions of Mackey-Glass data.

• Journal of Information Processing Systems
• /
• v.14 no.5
• /
• pp.1114-1135
• /
• 2018

A multiple classification system based on a new boosting technique has been approached utilizing different biometric traits, that is, color face, iris and eye along with fingerprints of right and left hands, handwriting, palm-print, gait (silhouettes) and wrist-vein for person authentication. The images of different biometric traits were taken from different standard databases such as FEI, UTIRIS, CASIA, IAM and CIE. This system is comprised of three different super-classifiers to individually perform person identification. The individual classifiers corresponding to each super-classifier in their turn identify different biometric features and their conclusions are integrated together in their respective super-classifiers. The decisions from individual super-classifiers are integrated together through a mega-super-classifier to perform the final conclusion using programming based boosting. The mega-super-classifier system using different super-classifiers in a compact form is more reliable than single classifier or even single super-classifier system. The system has been evaluated with accuracy, precision, recall and F-score metrics through holdout method and confusion matrix for each of the single classifiers, super-classifiers and finally the mega-super-classifier. The different performance evaluations are appreciable. Also the learning and the recognition time is fairly reasonable. Thereby making the system is efficient and effective.

• The Journal of Korean Association of Computer Education
• /
• v.10 no.5
• /
• pp.53-60
• /
• 2007

To enhance learning efficiency, we implement a hybrid experimental system for electrical and electronic circuits where web-based virtual laboratory system and distant education system are properly integrated. In the first stage, we developed web-based virtual laboratory systems for electrical/electronic circuit experiments, which are composed of three important sessions and their management system: concept learning, virtual experiment, assessment. In the second stage, we have implemented cost-effective distant laboratory systems for practicing electric/electronic circuits, which can be used to eliminate the lack of reality occurred during virtual laboratory session. The proposed virtual/ distant laboratory systems can be used in stand-alone fashion, but to enhance learning efficiency we integrated them and developed a creative hybrid experimental system for electric and electronic circuits.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.6
• /
• pp.177-186
• /
• 2020

Recent advances in deep learning technology have improved image recognition performance in the field of computer vision, and serverless computing is emerging as the next generation cloud computing technology for event-based cloud application development and services. Attempts to use deep learning and serverless computing technology to increase the number of real-world image recognition services are increasing. Therefore, this paper describes how to develop an efficient deep learning based image recognition service system using serverless computing technology. The proposed system suggests a method that can serve large neural network model to users at low cost by using AWS Lambda Server based on serverless computing. We also show that we can effectively build a serverless computing system that uses a large neural network model by addressing the shortcomings of AWS Lambda Server, cold start time and capacity limitation. Through experiments, we confirmed that the proposed system, using AWS Lambda Serverless Computing technology, is efficient for servicing large neural network models by solving processing time and capacity limitations as well as cost reduction.

• Journal of Convergence for Information Technology
• /
• v.11 no.11
• /
• pp.335-344
• /
• 2021

The purpose of this paper is the development of course data models and learning achievement computation algorithms for enabling the course-embedded assessment(CEA), which is essential of competency-based education in higher education. The previous works related CEA have weakness in the development of the systematic solution for CEA computation. In this paper, we propose data models and algorithms to implement competency-based assessment system. Our data models are composed of a layered architecture of learning outcomes, learning modules and activities, and an associative matrix of learning outcomes and activities. The proposed methods can be applied to the development of the course-embedded assessment system as core modules. We evaluated the effectiveness of our proposed models through applying the models to a practical course, Java Programing. From the result of the experiments we found that our models can be used in the assessment system as a core module.

• Journal of The Korean Association of Information Education
• /
• v.8 no.2
• /
• pp.167-176
• /
• 2004

Due to the development of information communication technology, educational environment has undergone much change and various types of teaching and learning methods based on information communication technology has been suggested. Recently, remote education using the internet are also spreading. However, in current classrooms, students are asked to do an teacher-centered assignment, which they are required to collect and report some information using the internet. This method does not help students use the advantages of learning using the internet, which stimulate students-students interaction and teacher-students interaction.Thus, this study focused on the problem-based learning system based on video communication technology. The researcher designed the problem-based learning system based on video communication technology and applied the system to classes at elementary school. The results were analyzed in terms of students-students interaction and teacher-students interaction in the internet. This research found that the problem-based learning system stimulates teacher and students communication and has positive effects on students' attitude and interest in learning. This research proposes that the traditional teacher-centered teaching method can be supplemented with cyber space learning, which has the merit of problem-based learning model.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.112-115
• /
• 2022

The process of deep learning usually needs to deal with massive data which has greatly limited the development of deep learning technologies today. Convolutional Neural Network (CNN) structure is often used to solve image classification problems. However, a large number of images may be required in order to train an image in CNN, which is a heavy burden for existing computer systems to handle. If the image data can be compressed under the premise that the computer hardware system remains unchanged, it is possible to train more datasets in deep learning. However, image compression usually adopts the form of lossy compression, which will lose part of the image information. If the lost information is key information, it may affect learning performance. In this paper, we will analyze the effect of image compression on deep learning performance on CNN-based cat and dog classification. Through the experiment results, we conclude that the compression of images does not have a significant impact on the accuracy of deep learning.

• Educational Technology International
• /
• v.8 no.2
• /
• pp.53-71
• /
• 2007

This study identified the effectiveness of e-learning by comparing learning outcome in conventional face-to-face lecture with the selected e-learning methods. Two e-learning contents (animation based and video based) were developed based on the rapid prototyping model and loaded onto the learning management system (LMS), which is http://www.enaged.co.kr. Fifty-four Korean agricultural high school students were randomly assigned into three groups (face-to-face lecture, animation based e-learning, and video based e-learning group). The students of the e-learning group logged on the LMS in school computer lab and completed each e-learning. All students were required to take a pretest and posttest before and after learning under the direction of the subject teacher. A one-way analysis of covariance was administered to verify whether there was any difference between face-to-face lecture and e-learning in terms of students' learning outcomes after controlling the covariate variable, pretest score. According to the results, no differences between animation based and video based e-learning as well as between face-to-face learning and e-learning were identified. Findings suggest that the use of well designed e-learning could be worthy even in agricultural education, which stresses hands-on experience and lab activities if e-learning was used appropriately in combination with conventional learning. Further research is also suggested, focusing on a preference of e-learning content type and its relationship with learning outcome.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.1
• /
• pp.9-19
• /
• 2024

Deep learning-based computer vision anomaly detection algorithms are widely utilized in various fields. Especially in the manufacturing industry, the difficulty in collecting abnormal data compared to normal data, and the challenge of defining all potential abnormalities in advance, have led to an increasing demand for unsupervised learning methods that rely on normal data. In this study, we conducted a comparative analysis of deep learning-based unsupervised learning algorithms that define and detect abnormalities that can occur when transparent contact lenses are immersed in liquid solution. We validated and applied the unsupervised learning algorithms used in this study to the existing anomaly detection benchmark dataset, MvTecAD. The existing anomaly detection benchmark dataset primarily consists of solid objects, whereas in our study, we compared unsupervised learning-based algorithms in experiments judging the shape and presence of lenses submerged in liquid. Among the algorithms analyzed, EfficientAD showed an AUROC and F1-score of 0.97 in image-level tests. However, the F1-score decreased to 0.18 in pixel-level tests, making it challenging to determine the locations where abnormalities occurred. Despite this, EfficientAD demonstrated excellent performance in image-level tests classifying normal and abnormal instances, suggesting that with the collection and training of large-scale data in real industrial settings, it is expected to exhibit even better performance.