• 한국컴퓨터산업학회논문지
• /
• 제2권2호
• /
• pp.151-156
• /
• 2001

우리는 개념 학습에 있어서 전통적으로 사용되어 온 연역 트리 구성법이나 규칙 학습법과 다른 새로운 개념 표현 기법을 소개하고자 한다. 우리의 PROLEARN 알고리즘은 각 클래스로부터 주어진 예제를 가장 잘 설명할 수 있는 가상 예제, 즉, 프로토타입을 하나 이상 학습하고 이것을 마치 주어진 예제처럼 취급하여 일반적인 개체 중심 학습법처럼 분류하도록 한다. 우리의 프로토타입 개념은 인지 심리학에서 사용한 같은 용어와는 하나의 개념이 하나 이상의 프로토타입을 가질 수 있도록 한 점에서 다르며 학습된 프로토타입은 근본적으로 ‘가상 예제’라는 점에서 다른 개체 중심 학습법과 다르다. 실험 결과 이 알고리즘은 정확도에서 다른 알고리즘에 뒤지지 않으며 실제 학습 문제에서 자주 발생하는 불안정성 문제, 즉 훈련 예제 집합이 바뀌면 알고리즘의 정확도도 영향 받는 부분도 해소하였다.

• 기본간호학회지
• /
• 제23권3호
• /
• pp.283-291
• /
• 2016

Purpose: This study was done to evaluate the persistency of effects of an integrated nursing simulation program on interest in learning, recognition of importance of communication, communication skills, and problem-solving abilities. Method: Forty-seven nursing students were recruited for this quasi-experimental design research. The experimental group (n=23) performed the simulation program for two weeks, and the control group (n=24) performed traditional clinical nursing practice for two weeks. Data were collected at baseline, immediately after the intervention, at 4 weeks, and finally at 8 weeks. Results: With respect to all variables, no significant differences were found between the experimental group and the control group. Interest in learning showed a significant increase in the control group (F=3.59, p=.018) at 4 weeks, and there was a significant increase in problem-solving abilities in the experimental group (F=4.98, p=.004) immediately after the intervention. Conclusion: Findings from this study suggest that the integrated nursing simulation program is as effective as the traditional clinical nursing practice, and the integrated nursing simulation program could be used as an alternative.

• 한국실천공학교육학회논문지
• /
• 제4권2호
• /
• pp.24-32
• /
• 2012

본 연구에서는 온라인 학습환경에서 수강자들간 발생하는 스터디 그룹의 유형과 그 특성을 밝히고자 한다. 이를 위하여 Neiderman & Beise(1999)의 네 가지 이론적 유형(typology)을 바탕으로 연구분석틀을 구성하고, 실제 데이터를 활용하여 실증적 유형(taxanomy)을 도출하고자 하였다. 데이터는 259명의 두 개 대학교 대학생을 대상으로 구성된 46개 팀을 대상으로 하였다. 군집분석 결과 실증적 유형은 Neiderman & Beise(1999)의 이론적 유형에서 수동적 유형(inactive type)을 제외하고 세 가지의 유형이 도출되었다. 또한 학습성과 측면에서는 온오프라인 전방위활용(fully-supported) 유형의 성과가 상대적으로 가상중심형(highly-virtual) 유형 보다 높은 것으로 나타났다.

• International Journal of Contents
• /
• 제15권4호
• /
• pp.59-64
• /
• 2019

In recent years, emotion recognition has been an interesting and challenging topic. Compared to facial expressions and speech modality, gesture-based emotion recognition has not received much attention with only a few efforts using traditional hand-crafted methods. These approaches require major computational costs and do not offer many opportunities for improvement as most of the science community is conducting their research based on the deep learning technique. In this paper, we propose an end-to-end deep learning approach for classifying emotions based on bodily gestures. In particular, the informative keyframes are first extracted from raw videos as input for the 3D-CNN deep network. The 3D-CNN exploits the short-term spatiotemporal information of gesture features from selected keyframes, and the convolutional LSTM networks learn the long-term feature from the features results of 3D-CNN. The experimental results on the FABO dataset exceed most of the traditional methods results and achieve state-of-the-art results for the deep learning-based technique for gesture-based emotion recognition.

• 한국초등과학교육학회지:초등과학교육
• /
• 제20권2호
• /
• pp.217-228
• /
• 2001

The purpose of this study was to investigate the effect of learning cycle model on the changes of electricity conceptions of elementary students. Four classes in forth grade of an elementary school in Busan were selected and two of them were served as experimental group and the others as control group. The experimental group were taught the unit of "Light an electric bulb" in elementary science textbook with teaching model based on teaming cycle and the control group with traditional teaching style. The instruction effects were analyzed through pre and post-test results using questionnaire on the electricity. The results of pre-test showed that there was not a significant difference between experimental group and control group at .05 level, so two groups could be regarded as homogeneous. The mean score of experimental group was significantly higher than that of control group on the post-test at .05 level. And within-group comparison revealed that both groups made improvement on the mean score and that the improvement of each group had significant difference at .05 level. Above results said that the teaching model based on learning cycle, which focuses on hands-on activity and considers each student as an active subject, was more effective than traditional teaching style in improving the formation of scientific conceptions on electricity.ectricity.

• 대한안전경영과학회지
• /
• 제7권5호
• /
• pp.155-174
• /
• 2005

Parametric life-cycle cost(LCC) models have been integrated with traditional design tools, and used in prior work to demonstrate the rapid solution of holistic, analytical tradeoffs between detailed design variations. During early designs stages there may be competing concepts with dramatic differences. Additionally, detailed information is scarce, and decisions must be models. for a diverse range of concepts, and the lack of detailed information make the integration make the integration of traditional LCC models impractical. This paper explores an approximate method for providing preliminary life-cycle cost. Learning algorithms trained using the known characteristics of existing products be approximated quickly during conceptual design without the overhead of defining new models. Artificial neural networks are trained to generalize on product attributes and life cycle cost date from pre-existing LCC studies. The Product attribute data to quickly obtain and LCC for a new and then an application is provided. In additions, the statistical method, called regression analysis, is suggested to predict the LCC. Tests have shown it is possible to predict the life cycle cost, and the comparison results between a learning LCC model and a regression analysis is also shown

• 대한간호학회지
• /
• 제48권6호
• /
• pp.743-753
• /
• 2018

Purpose: This study aimed to explore the effects of a community-based first and third Intergenerational Exchange Program (IGEP) on older adults' health-related quality of life (HRQoL), loneliness, depression, and walking speed, and on 4~5-year-old preschool children's learning-related social skills. Methods: This study employed a non-equivalent control group pre-post-test design. The experimental group included 42 older adults and 42 children who participated in the IGEP for 8 weeks, and the control group included 39 older adults. The experimental group participated in the IGEP once a week for 8 weeks. It comprised a traditional play program based on the intergroup contact theory. Results: Compared to the control group, there was a significant increase in scores on the HRQoL-Visual analogue scale (VAS) and a decrease in loneliness and depression in older adults in the experimental group (p<.05). Children who participated in the IGEP showed an improvement in their learning-related social skills (p<.001). Conclusion: These results confirm that the IGEP is an effective intervention to improve HRQoL-VAS, loneliness, and depression among older adults and learning-related social skills among preschool children in the community.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2021년도 제64차 하계학술대회논문집 29권2호
• /
• pp.313-315
• /
• 2021

Timely and accurate diagnosis of lung diseases using Chest X-ray images has been gained much attention from the computer vision and medical imaging communities. Although previous studies have presented the capability of deep convolutional neural networks by achieving competitive binary classification results, their models were seemingly unreliable to effectively distinguish multiple disease groups using a large number of x-ray images. In this paper, we aim to build an advanced approach, so-called Ensemble Knowledge Distillation (EKD), to significantly boost the classification accuracies, compared to traditional KD methods by distilling knowledge from a cumbersome teacher model into an ensemble of lightweight student models with parallel branches trained with ground truth labels. Therefore, learning features at different branches of the student models could enable the network to learn diverse patterns and improve the qualify of final predictions through an ensemble learning solution. Although we observed that experiments on the well-established ChestX-ray14 dataset showed the classification improvements of traditional KD compared to the base transfer learning approach, the EKD performance would be expected to potentially enhance classification accuracy and model generalization, especially in situations of the imbalanced dataset and the interdependency of 14 weakly annotated thorax diseases.

• Korean Journal of Radiology
• /
• 제25권6호
• /
• pp.550-558
• /
• 2024

Hepatocellular carcinoma (HCC) is a biologically heterogeneous tumor characterized by varying degrees of aggressiveness. The current treatment strategy for HCC is predominantly determined by the overall tumor burden, and does not address the diverse prognoses of patients with HCC owing to its heterogeneity. Therefore, the prognostication of HCC using imaging data is crucial for optimizing patient management. Although some radiologic features have been demonstrated to be indicative of the biologic behavior of HCC, traditional radiologic methods for HCC prognostication are based on visually-assessed prognostic findings, and are limited by subjectivity and inter-observer variability. Consequently, artificial intelligence has emerged as a promising method for image-based prognostication of HCC. Unlike traditional radiologic image analysis, artificial intelligence based on radiomics or deep learning utilizes numerous image-derived quantitative features, potentially offering an objective, detailed, and comprehensive analysis of the tumor phenotypes. Artificial intelligence, particularly radiomics has displayed potential in a variety of applications, including the prediction of microvascular invasion, recurrence risk after locoregional treatment, and response to systemic therapy. This review highlights the potential value of artificial intelligence in the prognostication of HCC as well as its limitations and future prospects.

• 한국문헌정보학회지
• /
• 제40권2호
• /
• pp.103-123
• /
• 2006

금융기관의 지식경영 초기 단계 이후부터는 지속적인 지식 창출과 효율적인 지식 검색이 지식경영의 핵심 요인으로 보고, 지식 창출의 한 방안으로 e-러닝을 제시하고, 효율적인 지식 검색 체제를 구축하기 위해서 리포지토리에 저장된 학습객체, 지식객체, 자료실 정보객체를 유사성에 따라 분류하고 상호 연관관계를 맺음으로써 키워드 검색은 물론 분류 검색과 연관 검색을 가능하게 하는 토픽맵 개념에 기반을 둔 지식맵을 활용한 통합 리포지토리 모형을 제안해 보았다. 모형 구현을 위해서 사용된 연구 방법에는 지식 관리 현황을 파악하기 위해서 세 보험회사들을 대상으로 사례 연구를 실시하였고, 기존의 토픽맵 기반의 실험적인 정보시스템들도 분석, 참조하였다. 디렉토리 형식의 전통적인 지식맵은 관련된 지식을 연계시키기가 어려워 지식관리시스템의 효율적인 브라우징이나 검색에 걸림돌로 작용하고 있는데 본 연구에서 제안된 모형은 이러한 문제점들을 개선할 하나의 안으로 이용될 수 있을 것이다.