• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.3
• /
• pp.344-355
• /
• 2002

Having a variety of good characteristics against other pattern recognition techniques, Multilayer Perceptron (MLP) has been used in wide applications. But, it is known that Error Backpropagation (EBP) algorithm which MLP uses in learning has a defect that requires relatively long leaning time. Because learning data in pattern recognition contain abundant redundancies, in order to increase learning speed it is very effective to use online-based teaming methods, which update parameters of MLP pattern by pattern. Typical online EBP algorithm applies fixed learning rate for each update of parameters. Though a large amount of speedup with online EBP can be obtained by choosing an appropriate fixed rate, fixing the rate leads to the problem that the algorithm cannot respond effectively to different leaning phases as the phases change and the learning pattern areas vary. To solve this problem, this paper defines learning as three phases and proposes a Instant Learning by Varying Rate and Skipping (ILVRS) method to reflect only necessary patterns when learning phases change. The basic concept of ILVRS is as follows. To discriminate and use necessary patterns which change as learning proceeds, (1) ILVRS uses a variable learning rate which is an error calculated from each pattern and is suppressed within a proper range, and (2) ILVRS bypasses unnecessary patterns in loaming phases. In this paper, an experimentation is conducted for speaker verification as an application of pattern recognition, and the results are presented to verify the performance of ILVRS.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06a
• /
• pp.76-78
• /
• 2006

현재 심혈관 질환은 암 다음으로 높은 사망 원인으로 기록되고 있어 심혈관 질환에 대한 초기 진단은 질환의 치료에 매우 중요한 문제로 대두되고 있다. 본 논문에서는 SVM을 이용하여 심혈관질환 환자의 질환 단계를 예측하였다. 일반적으로 이진분류에 사용되는 SVM을 이용하여 정상 및 질환 $1{\sim}3$기의 총 4가지 분류가 필요한 다분류 분류문제를 처리하기 위해서 논문에서는 독립적 학습된 단일 SVM 분류기들을 결합하여 분류를 수행하는 SVM 앙상블 방법을 사용하였다. 단일 분류기의 결합은 Majority voting, 최소자승에러기반 가중치 부여, 2단계층 결합 등의 방법으로 수행하여 심혈관 질환 분류에 적합한 앙상블의 구성을 시도하였다. 실험 데이터는 (주)제노프라의 압타머 칩 데이터를 사용하였다. 서로 다른 데이터를 이용하여 학습된 이종의 SVM들을 결합한 결과 질환단계 예측에 있어서 단일 SVM을 이용하여 질환 단계를 예측하는 경우 보다 향상된 질환단계 예측 성능을 관찰할 수 있었으며, 심혈관 질환의 예측에 대해서는 단일 SVM 분류기의 2단 계층 결합법이 가장 좋은 성능을 보임을 확인하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2013.11a
• /
• pp.1328-1331
• /
• 2013

본 논문에서는 스마트폰 내장 가속도 센서를 이용한 2단계 행위 인식 시스템을 제안한다. 제안하는 행위 인식 시스템에서는 행위 별 시간에 따른 가속도 센서 데이터의 변화 패턴을 충분히 반영하기 위해, 1단계 분류에서는 결정트리 모델 학습과 분류를 수행하고, 2단계 분류에서는 1단계 분류 결과들의 시퀀스를 이용하여 HMM모델 학습과 분류를 수행하였다. 또한, 본 논문에서는 특정 사용자나 스마트폰의 특정 위치, 방향 변화에도 견고한 행위 인식을 위하여, 동일한 행위에 대해 사용자와 스마트폰의 위치, 방향을 변경하면서 다양한 훈련 데이터를 수집하였다. 6720개의 가속도 센서 데이터를 이용하여 총 6가지 실내 행위들을 인식하기 위한 실험들을 수행하였고, 그 결과 높은 인식 성능을 확인 할 수 있었다.

• Journal of the Korean Chemical Society
• /
• v.63 no.5
• /
• pp.376-386
• /
• 2019

The purpose of this study was to explore the learning progression of logical thinking in acid and base chemical reactions and to evaluate its validity. For this purpose, we collected 387 participants in 9 schools of elementary, middle and high schools nationwide. The questionnaire developed in this study was composed of nine items. The questionnaire presented the acid and base reactants and products, and the students pictured their thoughts on how these substances change, and answered the reasons of their thoughts. Situation contexts of the questionnaire were divided into two groups: one kind of solute dissolved in a solvent, and two kinds of solute dissolved in a solvent. In this study, six levels of learning progression were assumed by combining material conservation logic, combination logic, proportion logic, and particle number conservation logic. By analyzing the data, Infit and Outfit values of Person reliability, Item reliability, MNSQ and ZSTD were obtained from the Rasch model. As a result of the analysis of data, it was found that lower levels of learning progression prevailed up to the younger grade students till $8^{th}$ grade. The higher levels of learning progression(Level 2~Level 5) prevailed up to the older grade students. However, higher levels of learning progression dropped sharply in Grade 12. The 5 level of learning progression was very low in all grades, and $9^{th}$ grade had highest percentage of students belonging to the 5 level. Interpretation of these unusual results suggests a future research related to explanation differences of textbooks.

• Journal of Engineering Education Research
• /
• v.8 no.1
• /
• pp.99-112
• /
• 2005

Problem-Based Learning is focused, experiential learning organized around the investigation and resolution of messy, real-world problem. It is both a curriculum organizer and instructional strategy, two complementary processes. The PBL model developed in this study was composed the two components of Problem Design(curriculum organizer) and Problem Implementation(instructional strategy). The basic process of Problem Implementation Model were composed the 8 steps ; 1) the identification of problem, 2) the specification of problem, 3) the exploration and generation for solution, 4) the selecting of best idea, 5) the specific planning of best idea, 6) the implementation and realization, 7) the evaluation, 8) the applying and reflection.

• Journal of the Korean earth science society
• /
• v.25 no.4
• /
• pp.205-213
• /
• 2004

The purpose of this study was to design the coaching strategies for the problem based learning and examine 'the problem situation' to analyze the process of learning as it applies to the students' perception on problem based learning. The steps of this model were as follows: 1) presentation of the problem situation 2) confrontation of the problem 3) know/ need to know 4) definition of the problem statement 5) collection and sharing of information 6)generation of possible solutions 7) assessment of the best fit of solutions 8) presentation of the solution. Problem-based learning steps and coaching strategies were designed and implemented to 2nd grade high school students for the environment teaching unit. The results demonstrated that group discussion in the know/need to know step was most helpful for students to review what they know and generate solutions. At first students tend to state problems widely but through repeated group discussions they gradually clarified the problems. In the students' personal reflection notes and perception questionnaire of problem-based learning, many students especially showed difficulties in defining the problem statement. In contrast they participated actively in the learning process and express their opinions enthusiastically. Therefore, this study suggests that developing problem situation based on real context is of great importance for implementing a problem based teaming model continuously.

• Journal of The Korean Association of Information Education
• /
• v.21 no.4
• /
• pp.451-462
• /
• 2017

This study selected 3D printing that is highly likely to be adopted in schools. This research was conducted in two stages: 1) proposing the learning activity framework for utilizing 3D printing in education, and 2) exploring the potential of integrating 3D printing in the school field. The '3D printing learning activity framework' proposed in this study includes four phases that are categorized according to the complexity of problem-solving processes and collaborative interaction: Step 1 as production through replication, Phase 2 as means of imaginary expression, Phase 3 as near problem-solving, and Phase 4 as expanded problem-solving. Next, we conducted the field study with 23 students in the 6th grade math class where they learned the various solid shapes and volumes through 3D printing-integrated activities. The lesson was considered as Phase 1, which is the production through replication. Overall, the results showed that the participants had positive perceptions about the efficacy of 3D printing activities, the quality of learning experience, and satisfaction. On the other hand, it was found that the usability of 3D printers and CAD program needs further improvement The contribution of this study can be found in the learning activity framework that can guide 3D printing activity design in school, and in the exploration of enhancing the connection between 3D printing activities and curricular relevance beyond simple interest toward a novel technology.

• Journal of Korean Home Economics Education Association
• /
• v.34 no.2
• /
• pp.41-58
• /
• 2022

This study aims to provide lesson plans that can simultaneously achieve both the learning goals of the 'Human Development & Family' domain of the 2015 revised Technology & Home Economics curriculum and the Sustainable Development Goals. Four steps including analysis, design, development, and evaluation and revision were followed. In the analysis step, the 'Changing Families and Healthy Families' unit was selected as it is relevant to the nine subgoals of the SDGs. In the design step, three sessions were planned with a problem-solving project approach. In the development step, three lesson plans for each session, individual and group activity worksheets for students, and guidelines for teachers were constructed. In the evaluation and revision step, criteria for evaluating the lesson plans were developed reflecting both the goals of the Home Economic curriculum and the SDGs. The validity of the lesson plans was reviewed by a panel of experts. Then, the revised lesson plans were finalized. This study provides an illustrative example of the lesson plans in the secondary education context that can be used to achieve the learning goals of the Home Economic curriculum and the Sustainable Development Goals at the same time.

• Journal of Korean Home Economics Education Association
• /
• v.22 no.3
• /
• pp.1-18
• /
• 2010

The purpose of this study was to develop a teaching-learning process plan for sustainable residing creating with neighbors and to apply it to the housing section of Technology-Home Economics according to the 2007 Revised Curriculum. Teachinglearning method solving practical problems was used for the teaching-learning process plans of 6-session lessons according to the ADDIE model. In the development stage, 17 activity materials and 15 teaching learning materials (6 reading texts, 6 moving pictures, 2 internet and 1 image materials) were developed. for the 6-session lessons, based on the stages of solving practical problems. The plans applied to the 3 classes of 8, 9, and 10th grade of the H. junior and senior high school in Myun district in Kyungbook during Sept. 1st to 14th, 2009. The results showed that students actively participated when the contents and materials were related to their own experience. The 6-session lessons about sustainable residing creating with neighbors was significantly increased the sense of community between before and after. Each of the 4 stages of the teachinglearning method solving practical problems were highly participated by the students. The satisfaction with the contents and methods of the 6-session lessons were evaluated over medium to somewhat higher levels. The practical activities to solve the community space and programs were got positive comments. Problem solving process and presentation and discussion were needed to learn more. Those results might support that the teachinglearning process plan this research developed. would be appropriate to the lessons for sustainable residing creating with neighbors.

• Journal of Science Education
• /
• v.36 no.2
• /
• pp.167-185
• /
• 2012

This study aimed at finding points of improvement in teaching expertise by analyzing the question patterns that appeared during teaching demonstrations which applied science teaching models prepared by a pre-service biology teacher. The question analysis frame for analyzing question types were categorized largely into the question types of Category 1 (questions in cognitive domain, questions with research function, questions in affective domain), Category 2 (repeated questions, questions for narrowing the range, practice questions), and Category 3 (questions on student activity progress, memory questions, and thinking questions). The results of analyzing question patterns from five different science teaching models revealed a high frequency of questions in the fields of cognition and memory. For the circular learning model, questions from the cognitive field appeared the most often, while, student activity progressive questions in particular were used mostly in the 'preliminary concept introduction stage' of the circular learning model and the 'secondary exploratory stage', in which experiments were conducted, and displayed the characteristics of these stages. The discovery learning model combined the courses of observation, measurement, classification and generalization, but, during teaching demonstrations, memory questions turned up the most, while the portion of inquisitive function questions was low. There were many questions from the inquisitive learning model, and, compared to other learning models, many exploratory function questions turned up during the 'experiment planning stage' and 'experiment stage'. Definitional questions and thought questions for the STS learning model turned up more than other learning models. During the change of concept learning model, the five concepts of students were stimulated and the modification of scientific concepts was very much aided by using many memory questions.