• 한국문헌정보학회지
• /
• 제50권2호
• /
• pp.309-336
• /
• 2016

본 논문에서는 지지벡터기계(Support Vector Machines, SVM) 기반의 기계 학습 모듈을 활용하여 특정 문장 내에서의 두 개체 간의 관계를 자동으로 식별하고 분류하는 바이오 분야 관계 추출 시스템을 제안한다. 제안된 시스템의 특징은 개체를 포함하고 있는 문장 내에서 풍부한 언어 자질을 추출하여 학습에 활용함으로써 그 성능을 극대화할 수 있는 다양한 기능들을 포함하고 있다는 점이다. 제안된 시스템의 성능 측정을 위해서 전 세계적으로 많이 활용되고 있는 바이오 분야 관계 추출 표준 컬렉션 3가지를 활용하여 심층적인 실험을 수행한 결과 모든 컬렉션에서 높은 성능을 획득하여 그 우수성을 입증하였다. 결론적으로, 본 논문에서 수행한 바이오 분야 관계 추출에 대한 광범위하고 심층적인 실험 연구가 향후 기계학습 기반의 바이오 분야 텍스트 분석 연구에 많은 시사점을 제공할 것으로 보인다.

• 수산해양교육연구
• /
• 제10권1호
• /
• pp.100-113
• /
• 1998

The information society is characterized by rapidly increasing and changing information. Traditional models of learning and instruct on have emphasized forms of mastering the information in content domains. Storing information and being able to recall it has been central to formal education. But it is no longer possible to master content domains. This paper discusses constructivism as a shifting paradigm for educational research and practice in information society. Constructivism provides an alternative epistemological base to the objectivist tradition. Constructivism holds that there is a real world that we experience. However, the argument is that meaning is imposed on the world by us, rather than existing in the world independently of us. Meaning is seen as rooted in experience. The experience in which an idea is embedded is critical to the individual's understanding of that idea. From the constructivist perspective, learning is not a stimulus-response phenomenon. It requires self-regulation and the building of conceptual structures through reflection and abstraction. Problems are not solved by the retrieval of rote-learned right answers. The effective motivation to continue learning can be fostered by leading students to experience the pleasure that is inherent in solving problems chosen as one's own. Constructivism requires the change of the teacher's role from a knowledge transmitter to a coach or facilitator of student's understanding. Constructivist teachers inquire about students' understanding of concepts before sharing their own understandings of those concepts, and encourage students to engage in dialogue, both with them arid with one another. In Korea, the educational reform called open education has been spreading through out the country. There should be a paradigm shift in learning and instruction from objectivism to constructivism for better educational reform in Korea.

• 한국멀티미디어학회논문지
• /
• 제4권3호
• /
• pp.280-285
• /
• 2001

귀납적학습 알고리즘과 분석적학습 알고리즘을 결합한 지식기반인공신경망이 제안된 후, 이를 개선한 TopGen, TR-KBANN, THRE-KBANN과 같은 영역이론정련알고리즘이 제시되었다. 그러나 이들은 모두 KBANN과 같이 영역이론이 있을 경우에만 사용할 수 있다. 본 연구에서는 영역이론이 없이 예제만 있는 경우 KBANN으로 표기하는 알고리즘을 제시하였다. KBANN으로 표현된 영역 이론은 THRE-KBANN으로 정련화될 수 있다. 이 알고리즘을 귀납적 학습에서 사용하는 몇 개의 문제영역에 적용하여 실험한 결과 C4.5보다 좋은 성능을 나타냈다.

• 한국멀티미디어학회논문지
• /
• 제9권6호
• /
• pp.762-772
• /
• 2006

Instance based learning algorithm is the best known lazy learner and has been successfully used in many areas such as pattern analysis, medical analysis, bioinformatics and internet applications. However, its feature weighting scheme is too naive that many other extensions are proposed. Our version of IB3 named as eXtended IBL (XIBL) improves feature weighting scheme by backward stepwise regression and its distance function by VDM family that avoids overestimating discrete valued attributes. Also, XIBL adopts leave-one-out as its noise filtering scheme. Experiments with common artificial domains show that XIBL is better than the original IBL in terms of accuracy and noise tolerance. XIBL is applied to two important applications - intrusion detection and spam mail filtering and the results are promising.

• 한국융합학회논문지
• /
• 제5권3호
• /
• pp.45-51
• /
• 2014

Computer science and computing technologies are applied into mathematical, science, medical, engineering and educational applications. The models are used to solve the issues in all the domains. Educational systems are used top down, bottom up, Gap Analysis model in the educational learning system. Educational learning process integrated with Lerner, content and the methodology. The Learners and content are same in the educational system or similar courses but the teaching methodologies are differing one with another. The determinations of teaching methodologies are based on the factors related to that particular model or subject. The learning model influencing determinations are made by the surveys, analysis and observation of data to maximize the learning outcome. This paper attempted to evaluate the SNE learners cognitive using NASA Scaling.

• Human Ecology Research
• /
• 제52권2호
• /
• pp.189-198
• /
• 2014

The purpose of this study was to examine the effects of the We Start center visiting intervention program on the positive changes in the developmental areas and literacy interests of young children from low-income families. The subjects were 195 young children (109 in the experimental group, M=71.7 months; 86 in the control group, M=73.3 months) living in We Start areas (Ansan, Gwangmyeong, and Suwon cities) of Gyeonggi-do. The We Start center visiting intervention programs were conducted for 30-34 weeks in each city, and pre-tests and post-tests were conducted before and after the intervention programs. The instruments used were the developmental checklist and the literacy interests test The developmental checklist consisted of 7 domains (personality & sociality; language, reading, & writing; scientific thinking; mathematical thinking; arts; social learning; and physical development). The literacy interests test consisted of 2 factors (interest in literacy and interaction during activity). The scores on the developmental checklist showed positive changes in several domains (personality & social development; language, reading, & writing ability; scientific thinking; and social learning), but not in mathematical thinking, arts, and physical development. Second, the results of the literacy interests test showed positive effects on interaction during activity and the total score. In conclusion, the We Start center visiting program for young children from low-income families is an effective early intervention program to end the intergenerational transference of poverty in Korea.

• 한국산학기술학회논문지
• /
• 제12권11호
• /
• pp.4788-4795
• /
• 2011

본 연구는 전통적인 방식으로부터 U-Learning으로 학습 패러다임을 전환하는 효과 및 가능성을 검토하고 제안하기 위한 것이며, E-Learning으로부터 M-Learning 및 U-Learning으로의 교수법 플랫폼 전환을 고려하기 위한 것이다. 개인별 학습 프로세스에 학습 환경이 어떤 영향을 미치는가에 대한 적절한 연구없이는 양질의 교육을 제공하기 어려울 것이다. 현대는 언제 어디서나 교육을 받을 수 있는 새로운 학습 환경 시대를 맞이하고 있으며, 누구나 평생교육을 받을 수 있게 되었다. 이러한 경향의 장점을 최대화하고, 양질의 교육을 진행하기 위한 제한 사항들을 확인하여야 하는데, 이들 요소들은 U-Learning 및 이를 가능하게 하는 기술과 함께 논의되어야 한다. 보급교육 혹은 평생교육은 많은 연구기관의 관심을 받고 있는바, 본 논문에서는 학습 모드 및 학습 양상의 유형에 대한 논의도 포함하였다.

• World Technopolis Review
• /
• 제7권1호
• /
• pp.2-14
• /
• 2018

Nowadays, we are standing in front of the $4^{th}$ Industrial Revolution that is featured by a great range of new and advanced technologies that influences all the domains of economies and industries. The great question that this revolution raises is how it can lead to a future that reflects the peoples' common objectives and values on how these advanced technologies can affect the life and transform the economic, social, cultural, and human environment. It is commonly agreed that to be adapted to these changes and needs and shape a society with competitive economies with highly-skilled individuals, we need to encourage innovation, entrepreneurship, new knowledge generation and exchange and true and effective collaboration and communication. In this complex scene, education seems to have a central and critical role on finding new ways of developing expertise and innovation within the existing knowledge procedures, with more and better cooperation between the key players. This paper argues the concepts, opportunities and challenges that are related to the learning ecosystem towards the needs raised by the $4^{th}$ Industrial Revolution. The education is discussed as catalyst but also as carrier of innovation and innovation practices and the basis of a relevant framework is presented that takes into account all the aspects, domains and key players of educational world and interacting domains. Having introduced the ideas of innovation, collaboration and technology advancement in this environment, this paper also presents a real case of practice, focusing on how more than 5.000 schools around Europe succeeded the last four (4) years to implement innovation activities in a collaborative way and under a unique but also flexible pedagogical innovation framework.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• 제18권1호
• /
• pp.58-65
• /
• 2007

Objectives : Motivational factor is a unique contributor to the typically poor academic performance of children with ADHD. However, few study has directly intervened learning motivation in children with ADHD. We conducted this study to explore the direct effects of the learning motivation improvement program applied to children with ADHD. Method The program was designed in order to increase an interest-inducing educational intervention, an academic skills integration, a basic learning activity (reading, writing, and math), and children's self-esteem. We conducted the program twice a week (total 10 sessions) and assessed learning motivation, teaming attitude, self-esteem, academic performance, and problem behaviors of participating children. Results : After the program, teachers reported improvement in teaming motivation. In addition, parents notified sisnificant reduction of problem behaviors. Children reported improvement in a few domains of teaming motivation and learning attitude. Conclusion : While loaming motivation is regarded as an important factor in education, there have been few studies considering this issue in both educational and psychiatric fields. The teaming motivation improvement would be needed in both field in order to reduce the deficits in academic performance in children with ADHD.

• Asia pacific journal of information systems
• /
• 제29권4호
• /
• pp.789-816
• /
• 2019

Text classification is a challenging task, especially when dealing with a huge amount of text data. The performance of a classification model can be varied depending on what type of words contained in the document corpus and what type of features generated for classification. Aside from proposing a new modified version of the existing algorithm or creating a new algorithm, we attempt to modify the use of data. The classifier performance is usually affected by the quality of learning data as the classifier is built based on these training data. We assume that the data from different domains might have different characteristics of noise, which can be utilized in the process of learning the classifier. Therefore, we attempt to enhance the robustness of the classifier by injecting the heterogeneous data artificially into the learning process in order to improve the classification accuracy. Semi-supervised approach was applied for utilizing the heterogeneous data in the process of learning the document classifier. However, the performance of document classifier might be degraded by the unlabeled data. Therefore, we further proposed an algorithm to extract only the documents that contribute to the accuracy improvement of the classifier.