• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10a
• /
• pp.16-18
• /
• 2003

기계학습의 분류(classification) 기술을 실제 문제에 적용하기 위해서는 카테고리(category)를 부여한 학습예제를 상당수 준비하여야 한다. 예제에 카테고리를 부여(labeling)하는 작업에는 무시할 수 없는 시간과 인력을 필요로 한다. 능동적 학습(active learning)은 동일한 수의 학습예제로 최대한의 성능을 달성하기 위하여 카테고리를 부여할 학습예제를 선별하는 전략이다. 능동적 학습은 현재까지 파악된 정보에 기반하여 분류기(classifier)를 생성하고, 생성된 분류기를 활용하여 카테고리를 부여받았을 때 가장 이득이 큰 예제들을 선정하여 사용자에게 문의하는 과정을 반복하여 수행한다. 만일 능동적 학습의 첫 학습단계에서 학습에 보다 유용한 예제들을 최초학습예제집합으로 선정한다면 같은 수의 학습예제로 더 나은 성능을 달성할 수 있을 것이다. 본 논문에서는 유사한 예제들은 동일한 카테고리에 속할 가능성이 높다는 일반적인 가정에 기반하여 예제들을 군집화(clustering)한 후, 생성된 각 군집을 대표할 수 있는 예제로 최초학습예제집합으로 구성하는 방안을 제안한다. 제안한 방안을 문서분류 문제를 대상으로 실험해 본 결과 최초학습예제들을 임의로 선정하는 방식보다 정확도가 높은 분류기를 생성할 수 있음을 확인하였다.

• Proceedings of the KSPS conference
• /
• 2005.11a
• /
• pp.103-106
• /
• 2005

In this paper, we propose a language modeling approach to improve the performance of a large vocabulary continuous speech recognition system. The proposed approach is based on the active learning framework that helps to select a text corpus from a plenty amount of text data required for language modeling. The perplexity is used as a measure for the corpus selection in the active learning. From the recognition experiments on the task of continuous Korean speech, the speech recognition system employing the language model by the proposed language modeling approach reduces the word error rate by about 6.6 % with less computational complexity than that using a language model constructed with randomly selected texts.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.425-431
• /
• 2023

음성 어시스턴트 시스템에서 발화의 의도를 분류하고 새로운 의도를 탐지하는 것은 매우 중요한 작업이다. 끊임없이 인입되는 새로운 발화로 인해 기존에 학습된 모델의 의도 분류 성능은 시간이 지남에 따라 점차 낮아진다. 기존 연구들에서 새로운 의도 발견을 위해 제안되었던 클러스터링 방법은 최적의 클러스터 수 결정과 명명에 어려움이 있다. 이러한 제한 사항을 보완하기 위해, 본 연구에서는 대규모 언어 모델 기반의 효과적인 의도 발견 방법을 제안한다. 이 방법은 기존 의도 분류기로 판단하기 어려운 발화에 새로운 의도 레이블을 할당하는 방법이다. 새롭게 인입되는 OOD(Out-of-Domain) 발화 내에서 오분류를 찾아 기존에 정의된 의도를 탐지하고, 새로운 의도를 발견하는 효율적인 프롬프팅 방법도 분석한다. 이를 액티브 러닝 전략과 결합할 경우, 분류 가능한 의도의 개수를 지속 증가시면서도 모델의 성능 하락을 방지할 수 있고, 동시에 새로운 의도 발견을 자동화 할 수 있다.

• Journal of Mechanical Science and Technology
• /
• v.16 no.12
• /
• pp.1633-1642
• /
• 2002

Uninterrupted power supply has become indispensable during the maintenance task of active electric power lines as a result of today's highly information-oriented society and increasing demand of electric utilities. The maintenance task has the risk of electric shock and the danger of falling from high place. Therefore it is necessary to realize an autonomous robot system using electro-hydraulic manipulator because hydraulic manipulators have the advantage of electric insulation. Meanwhile it is relatively difficult to realize autonomous assembly tasks particularly in the case of manipulating flexible objects such as electric lines. In this report, a discrete event control system is introduced for automatic assembly task of electric lines into sleeves as one of the typical task of active electric power lines. In the implementation of a discrete event control system, LVQNN (linear vector quantization neural network) is applied to the insertion task of electric lines to sleeves. In order to apply these proposed control system to the unknown environment, virtual learning data for LVQNN is generated by fuzzy inference. By the experimental results of two types of electric lines and sleeves, these proposed discrete event control and neural network learning algorithm are confirmed very effective to the insertion tasks of electric lines to sleeves as a typical task of active electric power maintenance tasks.

• Knowledge Management Research
• /
• v.12 no.2
• /
• pp.17-34
• /
• 2011

As organization's success depends on how much employees are active in recognizing new knowledge and transform into organization's value, this study is to examine the influence of employees' self-directed learning capacity on organizational absorptive capacity, organizational commitment and job involvement. The study has shown that managerial level employees have high level of self-directed learning capacity, and employees who have such high level capacity also shown to have high level of absorptive capacity and organizational affective commitment and job involvement. There were static correlations between all the variables, and absorptive capacity has a complete meditated effect on each variables. Hence, in order to enhance absorptive capacity, organizations should put their effort to develop employees' self-directed learning capacity, which will not only improve individual competency but also lead to organization's success in this competitive global market.

• 한국정보통신설비학회:학술대회논문집
• /
• 2009.08a
• /
• pp.151-154
• /
• 2009

Enhanced active learning through information technology is becoming important for business and education sector. This paper describes one case of the continuing works for educational services through information technologies on the two-way learning and communication experience for live classes. As we expected, we faced some critical challenges but we believe that this work could make a significant impact on teaching, learning and creativity for better education in the near future.

• The Journal of Korean Academic Society of Nursing Education
• /
• v.6 no.2
• /
• pp.359-375
• /
• 2000

PBL can be defined as an active, self-directed and student-centered learning, and an opposite way of classroom teacher-centered learning which has been traditional role learning. PBL enables students think more efficiently and effectively when puzzling through the patient problems. The purpose of this study is to find out the perception of student nurse about PBL, the characteristics and the structure of the type for PBL. The research process is as follow : First, the researcher selected 35 statements for PBL with the content analysis of in depth interview and the literature review. Second, the researcher asks 38 student nurse to classify the statement cards. The result of the research is that the type of student nurse's PBL perception is divided into 4 types(Affirmative type, Negative type, Suspicious type, and Preferable type), and the explanative total variance is 44 percent. In relation to this, if PBL well combined and adapted in our traditional curriculum will change our nursing education in better direction.

• Education of Primary School Mathematics
• /
• v.10 no.2
• /
• pp.151-161
• /
• 2007

There were a lot of attempt to reform mathematics education. These may include reforms of teaching and learning methods, development of mathematics curriculum and textbooks, materials and resources for teaching. But the most important thing is reform of teaching and learning method. The teaching and learning in mathematics is related with culture of mathematics classroom. Thus the culture of mathematics classroom plays a critical role in the learning of mathematics. Although there is consensus what desirable culture of mathematics classroom is, the role of teachers is important to make desirable culture of mathematics classroom. So, this paper suggested the role of teachers in reforming the culture of mathematics classroom. Avoiding from routine and technical teaching, the teachers should play active role in constructing mathematical understanding of students.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.24.2-24
• /
• 2001

In this paper, the authors present a row dynamic route planning by Q-learning. The proposed algorithm is executed in a cellular automation based traffic simulator, which is also newly created. In Vehicle Information and Communication System(VICS), which is an active field of Intelligent Transport System(ITS), information of traffic congestion is sent to each vehicle at real time. However, a centralized navigation system is not realistic to guide millions of vehicles in a megalopolis. Autonomous distributed systems should be more flexible and scalable, and also have a chance to focus on each vehicles demand. In such systems, each vehicle can search an own optimal route. We employ Q-learning of the reinforcement learning method to search an optimal or sub-optimal route, in which route drivers can avoid traffic congestions. We find some applications of the reinforcement learning in the "static" environment, but there are ...

• Journal of Engineering Education Research
• /
• v.22 no.1
• /
• pp.22-28
• /
• 2019

The purpose of this study is to develop instructional design model of flipped learning suitable for engineering education field and to draw out effects and improvements by applying it to actual lessons for engineering college students. Literature review and case studies were conducted to achieve the purpose of the study. For a case study, flipped learning was applied to 'creative problem solving methodology' which is a liberal arts course of engineering college at D university in Gyeonggi-do. As a result of the literature review, the PARTNER model was applied and weekly instructional guide was presented by each stage. In addition, the results of analysis on the reflection journal showed that the students were more able to achieve the deepening learning stage through active participation in class than the existing class, and found that they had a more challenging plan after the class.