• 한국정보교육학회:학술대회논문집
• /
• 한국정보교육학회 2011년도 동계학술대회
• /
• pp.29-35
• /
• 2011

세계는 매우 빠른 속도로 변하고 있으며 이 변화 속도는 21세기를 살아 갈 어린이들에게 보다 더 효율적으로 빠르게 학습하고 창의적으로 사고 할 수 있는 능력을 요구한다. 이러한 상황 속에서 컴퓨터 교육 분야의 중요성은 점점 더 높아지고 있으나 학교 현장에서는 컴퓨터 교육과 관련한 시수가 줄고 있는 추세이다. 이와 같은 조건에서 학습자 중심으로 가속하는 변화를 다스릴 수 있도록 새로운 정보를 빠르게 흡수하고 이해할 수 있는 능력과 그 정보를 보유할 수 있는 능력을 중요시하는 교수 방법인 가속학습(Acceleated Learning)을 정보통신기술교육에 적용하여 시간적인 한계를 극복하고 제한된 시간 안에 최대한의 효과를 끌어내고자 한다. 이러한 목적을 달성하기 위해 가속학습이론에 기반을 둔 6단계 수업활동과 다중지능이론을 적용한 PLAY(Pre-processing, Learning how to recognize, Activating the problem solving, Yield product) 모형을 구안하였으며, 이를 경기도 남양주시에 소재하고 있는 초등학교 2학년 2개 학급 70명을 대상으로 10차시 분량의 정보통신기술교육 실험 수업을 실시하였다.

• 디지털콘텐츠학회 논문지
• /
• 제19권2호
• /
• pp.415-423
• /
• 2018

최근, 컴퓨터 분야의 기계 학습(Machine Learning)과 딥러닝(Deep Learning) 등 컴퓨터 관련 학습이 각광을 받고 있다. 이들은 인공 신경망(Artificial Neural Network)을 이용하여 가장 하위 레벨로부터 학습을 시작하여, 최상위 레벨까지 그 결과를 전달하여 최종 결과를 산출하는 방식이다. 하위레벨로부터의 체계적인 학습을 통한 효과적인 성장 및 교육 방안에 대한 연구는 다양한 분야에서 이루어지고 있으나, 체계적인 규칙과 방법에 기반한 모델은 찾아보기가 힘들다. 이에, 본 논문에서는 성장 및 융합 모델인, TNT 모델(Transitive Nested Triangle Model)을 처음으로 제안한다. 제안하는 모델은 기하학적인 형태를 통해 형성된 각 기능들이 유기적 계층 관계를 형성하여, 상위로 성장 및 융합하면서, 그 결과가 반복 사용되는 순환적 재귀 모델이다. 즉, '수평적 형제 병합에 이은 상위로의 융합(Horizontal Sibling Merges and Upward Convergence)'의 분석적 방법이다. 이러한 모델은 공학, 디지털공학, 인문학, 예술학 등에 모두 적용될 수 있는 기본기적 이론으로, 본 연구에서는 제안하는 TNT 모델을 설명하는 것에 그 초점을 둔다.

• 대한산업공학회지
• /
• 제40권4호
• /
• pp.404-414
• /
• 2014

Yield prediction is important to manage semiconductor quality. Many researches with machine learning algorithms such as SVM (support vector machine) are conducted to predict yield precisely. However, yield prediction using SVM is hard because extremely imbalanced and big data are generated by final test procedure in semiconductor manufacturing process. Using SVM algorithm with imbalanced data sometimes cause unnecessary support vectors from major class because of unselected support vectors from minor class. So, decision boundary at target class can be overwhelmed by effect of observations in major class. For this reason, we propose a under-sampling method with minor class based SVM (MCSVM) which overcomes the limitations of ordinary SVM algorithm. MCSVM constructs the model that fixes some of data from minor class as support vectors, and they can be good samples representing the nature of target class. Several experimental studies with using the data sets from UCI and real manufacturing process represent that our proposed method performs better than existing sampling methods.

• 한국인터넷방송통신학회논문지
• /
• 제20권1호
• /
• pp.219-223
• /
• 2020

딥 러닝 기반 객체 탐지 및 영상처리 분야에서 모델의 인식률과 정확도를 보장하기 위해 다량의 데이터 확보는 필수적이다. 본 논문에서는 학습데이터가 적은 경우에도 인공지능 모델의 높은 성능을 도출하기 위해 전이학습 기반 객체탐지 알고리즘을 제안한다. 본 논문에서는 객체탐지를 위해 사전 학습된 Resnet-50 네트워크와 YOLO(You Only Look Once) 네트워크를 결합한 전이학습 네트워크를 구성하였다. 구성된 전이학습 네트워크는 Leeds Sports Pose 데이터셋의 일부를 활용하여 이미지에서 가장 넓은 영역을 차지하고 있는 사람을 탐지하는 네트워크로 학습을 진행하였다. 실험결과는 탐지율 84%, 탐지 정확도 97%를 기록하였다.

• Steel and Composite Structures
• /
• 제29권3호
• /
• pp.309-318
• /
• 2018

Steel girders are the structural members often used for passing long spans. Mostly being subjected to patch loading, or concentrated loading, steel girders are likely to face sudden deformation or damage e.g., web breathing. Horizontal or vertical stiffeners are employed to overcome this phenomenon. This study aims at assessing the feasibility of a machine learning method, namely the support vector machines (SVM) in predicting the patch loading resistance of longitudinally stiffened webs. A database consisting of 162 test data is utilized to develop SVM models and the model with best performance is selected for further inspection. Existing formulations proposed by other researchers are also investigated for comparison. BS5400 and other existing models (model I, model II and model III) appear to yield underestimated predictions with a large scatter; i.e., mean experimental-to-predicted ratios of 1.517, 1.092, 1.155 and 1.256, respectively; whereas the selected SVM model has high prediction accuracy with significantly less scatter. Robust nature and accurate predictions of SVM confirms its feasibility of potential use in solving complex engineering problems.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제16권2호
• /
• pp.81-86
• /
• 2016

Term weighting is a popular technique that effectively weighs the term features to improve accuracy in document classification. While several successful term weighting algorithms have been suggested, none of them appears to perform well consistently across different data domains. In this paper we propose several reasonable methods to combine different term weight vectors to yield a robust document classifier that performs consistently well on diverse datasets. Specifically we suggest two approaches: i) learning a single weight vector that lies in a convex hull of the base vectors while minimizing the class prediction loss, and ii) a mini-max classifier that aims for robustness of the individual weight vectors by minimizing the loss of the worst-performing strategy among the base vectors. We provide efficient solution methods for these optimization problems. The effectiveness and robustness of the proposed approaches are demonstrated on several benchmark document datasets, significantly outperforming the existing term weighting methods.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 1996년도 추계학술대회발표논문집; 고려대학교, 서울; 26 Oct. 1996
• /
• pp.207-210
• /
• 1996

Learning methodologies such as neural networks or genetic algorithms usually require long training times. Case based reasoning, however, attains peak performance swiftly and is often appropriate for learning even with small data sets. Previous work has shown that an extended case reasoning methodology can yield superior performance in the task of predicting financial data series. This paper examines the impact of reasoning procedures on stock price prediction. The following characteristics are evaluated: size of input vector, multiplicity of neighboring states, and a scaling factor for growth. The concepts are illustrated in the context of predicting the price of an individual price.

• 한국경영과학회지
• /
• 제18권2호
• /
• pp.57-81
• /
• 1993

This paper is concerned with analyzing the bankruptcy prediction power of three methods : Multivariate Discriminant Analysis (MDA), Inductive Learning, Neural Network, MDA has been famous for its effectiveness for predicting bankrupcy in accounting fields. However, it requires rigorous statistical assumptions, so that violating one of the assumptions may result in biased outputs. In this respect, we alternatively propose the use of two AI models for bankrupcy prediction-inductive learning and neural network. To compare the performance of those two AI models with that of MDA, we have performed massive experiments with a number of Korean bankrupt-cases. Experimental results show that AI models proposed in this study can yield more robust and generalizing bankrupcy prediction than the conventional MDA can do.

• 반도체디스플레이기술학회지
• /
• 제19권3호
• /
• pp.12-18
• /
• 2020

In semiconductor manufacturing, defect detection is critical to maintain high yield. Currently, computer vision systems used in semiconductor photo lithography still have adopt to digital image processing algorithm, which often occur inspection faults due to sensitivity to external environment. Thus, we intend to handle this problem by means of using Mask R-CNN instead of digital image processing algorithm. Additionally, Mask R-CNN can be trained with image dataset pre-processed by means of the specific designed digital image filter to extract the enhanced feature map of Convolutional Neural Network (CNN). Our approach converged advantage of digital image processing and instance segmentation with deep learning yields more efficient semiconductor photo lithography inspection system than conventional system.

• IEIE Transactions on Smart Processing and Computing
• /
• 제4권4호
• /
• pp.265-271
• /
• 2015

This study proposes an algorithm for recognizing apple trees in images and detecting apples to measure the number of apples on the trees. The proposed algorithm explores whether there are apple trees or not based on the number of image block-unit edges, and then it detects apple areas. In order to extract colors appropriate for apple areas, the CIE $L^*a^*b^*$ color space is used. In order to extract apple characteristics strong against illumination changes, modified census transform (MCT) is used. Then, using the AdaBoost learning algorithm, characteristics data on the apples are learned and generated. With the generated data, the detection of apple areas is made. The proposed algorithm has a higher detection rate than existing pixel-based image processing algorithms and minimizes false detection.