• Title/Summary/Keyword: 운동학습

Search Result 450, Processing Time 0.026 seconds

A Machine Learning Algorithm Study for Predicting Time-Averaged Velocity Fluctuations in Turbulent Jets (난류 제트 내 시간 평균 속도 변동 예측을 위한 기계 학습 알고리즘)

  • Seongeun Choi;Jin Hwan Hwang
    • Proceedings of the Korea Water Resources Association Conference
    • /
    • 2023.05a
    • /
    • pp.130-130
    • /
    • 2023
  • 제트류는 다양한 크기와 운동량의 에디가 복잡하게 혼합되어 이루어져 있으며, 이를 정확하게 모델링하고 이해하기 위해서는 제트류의 다양한 특성들을 잘 반영하여 연구를 수행해야 한다. 다양한 연구 수행 방법 중 수치해석 방법은 상대적으로 공간 및 시간적 비용이 적게 들어서 널리 사용되고 있다. 이러한 수치해석 방법에는 DNS(Direct Numerical Simulation), LES(Large Eddy Simulation), RANS(Reynolds Averaged Navier Stokes) 등이 있으며, 그중 LES는 난류 모델링을 사용하는 RANS 방법에 비해 더욱 정확한 흐름 모델링을 제공하는 장점이 있다. 이러한 LES는 대규모 에디는 직접 해석하면서, 일정 크기 이하의 에디는 모델링을 사용해 해석하는 것이 특징이다. 하지만, LES를 사용하기 위해서는 적절한 그리드 크기를 결정하는 것이 중요하며, 이는 모델의 정확성과 연산 비용에 큰 영향을 미친다. 하지만, 여전히 적절한 그리드 크기를 결정하는 것은 어려운 문제이다. 이러한 LES 모델링을 사용할 때 적절한 그리드 크기를 결정하기 위해서는 정확한 시간 평균 속도 변동을 연구하는 것이 앞서 선행되어야 한다. 따라서, 본 연구에서는 기계학습 기반 접근 방식을 사용하여 난류 제트 내 시간 평균 속도 변동을 예측하는 연구를 진행하였다. 즉, 난류 제트 역학을 이해하는 데 중요한 파라미터인 시간 평균 유속을 이용하여 시간 평균 속도 변동을 예측하는 데 초점을 맞추었다. 모델의 성능은 평균 제곱 오차와 R-제곱 등 다양한 지표를 사용하여 평가되었다.

  • PDF

The Effect of Bilateral Eye Movements on Face Recognition in Patients with Schizophrenia (양측성 안구운동이 조현병 환자의 얼굴 재인에 미치는 영향)

  • Lee, Na-Hyun;Kim, Ji-Woong;Im, Woo-Young;Lee, Sang-Min;Lim, Sanghyun;Kwon, Hyukchan;Kim, Min-Young;Kim, Kiwoong;Kim, Seung-Jun
    • Korean Journal of Psychosomatic Medicine
    • /
    • v.24 no.1
    • /
    • pp.102-108
    • /
    • 2016
  • Objectives : The deficit of recognition memory has been found as one of the common neurocognitive impairments in patients with schizophrenia. In addition, they were reported to fail to enhance the memory about emotional stimuli. Previous studies have shown that bilateral eye movements enhance the memory retrieval. Therefore, this study was conducted in order to investigate the memory enhancement of bilaterally alternating eye movements in schizophrenic patients. Methods : Twenty one patients with schizophrenia participated in this study. The participants learned faces (angry or neutral faces), and then performed a recognition memory task in relation to the faces after bilateral eye movements and central fixation. Recognition accuracy, response bias, and mean response time to hits were compared and analysed. Two-way repeated measure analysis of variance was performed for statistical analysis. Results : There was a significant effect of bilateral eye movements condition in mean response time(F=5.812, p<0.05) and response bias(F=10.366, p<0.01). Statistically significant interaction effects were not observed between eye movement condition and face emotion type. Conclusions : Irrespective of the emotional difference of facial stimuli, recognition memory processing was more enhanced after bilateral eye movements in patients with schizophrenia. Further study will be needed to investigate the underlying neural mechanism of bilateral eye movements-induced memory enhancement in patients with schizophrenia.

Analysis of the Science Words Used by Science Teachers in Teaching the Unit of 'Force and Motion' (중학교 과학 교사가 '힘과 운동' 단원 수업 중에 사용하는 과학용어 분석)

  • Yun, Eunjeong;Park, Yunebae
    • Journal of The Korean Association For Science Education
    • /
    • v.35 no.2
    • /
    • pp.209-216
    • /
    • 2015
  • In science classrooms, using science terminology is a very important aspect of communications between science teachers and students, as well as in the science learning of students. This study was conducted to investigate the usage of the science terminology in the lectures of science teachers, and identify the problem in the aspect of both communication and teaching. To do this, we have recorded 13 hours of class teaching 'Motion' part in unit of 'Force and Motion' from three science teachers, and extracted science terminologies from the science teachers' lectures by using an analysis program. We performed qualitative analysis, such as kind of science terminology used, and linkage between curriculum and textbook, and quantitative analysis, such as number of science terminology, and frequency of use. With respect to communication, there appears some problems in its proportion in the teacher's lecture in class. It is deemed that science terminology in teachers' lectures were too many, that the frequency of usage of important conceptual terminology was low, and that teachers use higher level terminologies to explain key concepts. And in respect to science learning, there were problems where terminologies including important concepts were used separately by the teachers and textbooks, terminologies of higher level concept were used, and there might be differences between teachers in majority of teachers.

Correlations of Elementary Students’ Spatial Abilities with their Conceptions of Celestial Motion and Science Process Skills (초등학교 학생들의 공간능력과 천체운동개념 및 과학탐구능력과의 관계)

  • Kim, Sang-Dal;Lee, Young-Seob;Lee, Sang-Gyun
    • Journal of the Korean earth science society
    • /
    • v.26 no.6
    • /
    • pp.461-468
    • /
    • 2005
  • The purpose of this study is to investigate the abilities of 6th grade students to assess the relationship between spatial capability, concept of celestial motion, and science process skills, which can help find a better teaching strategy for students in understanding the concept of celestial motion. The results are as follows. First, in terms of level of accomplishment of these three skills, male students show higher level of accomplishment than female students, but significant differences are found. Second, according to the analysis of the effect of spatial capability and concept of the movement of heavenly bodies, the former has a stronger influence on the students' cognition of celestial motion. Minor elements of spatial capability that influence the conception of celestial motion are device analogy, calculation of wood cut, and revolving light. Third, spatial capability is very influential on the level of accomplishment in science process skills. Among the minor elements of spatial capability that is influential to science process skills, calculation of wood cut is the highest, especially when various elements are interactively related to each other.

Transfer Learning Backbone Network Model Analysis for Human Activity Classification Using Imagery (영상기반 인체행위분류를 위한 전이학습 중추네트워크모델 분석)

  • Kim, Jong-Hwan;Ryu, Junyeul
    • Journal of the Korea Society for Simulation
    • /
    • v.31 no.1
    • /
    • pp.11-18
    • /
    • 2022
  • Recently, research to classify human activity using imagery has been actively conducted for the purpose of crime prevention and facility safety in public places and facilities. In order to improve the performance of human activity classification, most studies have applied deep learning based-transfer learning. However, despite the increase in the number of backbone network models that are the basis of deep learning as well as the diversification of architectures, research on finding a backbone network model suitable for the purpose of operation is insufficient due to the atmosphere of using a certain model. Thus, this study applies the transfer learning into recently developed deep learning backborn network models to build an intelligent system that classifies human activity using imagery. For this, 12 types of active and high-contact human activities based on sports, not basic human behaviors, were determined and 7,200 images were collected. After 20 epochs of transfer learning were equally applied to five backbone network models, we quantitatively analyzed them to find the best backbone network model for human activity classification in terms of learning process and resultant performance. As a result, XceptionNet model demonstrated 0.99 and 0.91 in training and validation accuracy, 0.96 and 0.91 in Top 2 accuracy and average precision, 1,566 sec in train process time and 260.4MB in model memory size. It was confirmed that the performance of XceptionNet was higher than that of other models.

The Effects of Small Group Drawing in Learning the Particulate Nature of Matter (물질의 입자성에 대한 학습에서 소집단 그림 그리기의 효과)

  • Han, Jae-Young;Kim, Hun-Sik;Kim, Bo-Kyung;Noh, Tae-Hee
    • Journal of The Korean Association For Science Education
    • /
    • v.25 no.7
    • /
    • pp.721-727
    • /
    • 2005
  • This study investigated the effects of small group drawing in learning the concepts of particulate nature of matter. Three classes of seventh graders (N = 126) at a coed middle school were randomly assigned to a pair drawing group, an individualistic drawing group, and a control group. The students were taught the 'three states of matter' and 'motion of molecules' for eight class periods. Prior to these classes, student self-efficacy, learning motivation, and attitude toward science instruction were examined. After instruction, tests assessing achievement, conception, learning motivation, and attitude toward science instruction were administered. Two-way ANCOVA results revealed that scores of achievement and conception for the pair drawing group were significantly higher than those for the control group. However, scores of the three groups did not significantly differ in learning motivation and attitude toward science instruction. Furthermore, no significant interactions surfaced between instruction and the level of self-efficacy in all dependent variables.

Research on DNN Modeling using Feature Selection on Frequency Domain for Vital Reaction of Breeding Pig (모돈 생체 반응 신호의 주파수 영역 Feature selection을 통한 DNN 모델링 연구)

  • Cho, Jinho;Oh, Jong-woo;Lee, DongHoon
    • Proceedings of the Korean Society for Agricultural Machinery Conference
    • /
    • 2017.04a
    • /
    • pp.166-166
    • /
    • 2017
  • 모돈의 건강 상태를 정량 지수화 하기 위한 연구를 수행 중이다. 지제이상, 섭식 불량, 수면 패턴 등의 운동 특성 분석을 위하여 복수의 초음파 센서를 이용하였다. 시계열 계측 신호를 분석하여 정량 지수화를 수행하는 과정에서 주파수 도메인 분석을 시도하였다. 이 과정에서 주파수 도메인의 분해능에 따른 편차 극복을 위한 비선형 모델링을 수행하였다. 또한 인접한 시계열 데이터 구간 간의 상관성 분석이 가능하면 대용량 데이터의 실시간 처리로 인한 지연 시간 극복 및 기대되는 예후에 대한 조기 진단이 가능할 것이다. 본 연구에서는 구글에서 제공하는 Tensorflow와 NVIDIA에서 제공하는 CUDA 엔진을 동시 적용한 심층 학습 시스템을 이용하였다. 전 처리를 위하여 주파수 분해능 (2분, 3분, 5분, 7분, 11분, 13분, 17분, 19분)에 따른 데이터 집합을 1단계로 두고, 상위 10 순위 안에 드는 파워 스펙트럼 밀도의 크기를 2단계로 하여, 총 2~10개의 입력 노드를 순차적으로 선정하였고, 동일한 방식으로 인접한 시계열의 파워 스펙터럼 밀도를 순위를 변화시켜 지정하였다. 대표적인 심층학습 모델인 Softmax regression with a multilayer convolutional network를 이용하여 Recursive feature selection 경우의 수를 $8{\times}9{\times}9$로 총 648 가지 선정하고, Epoch는 10,000회로 지정하였다. Calibration 모델링의 경우 Cost function이 10% 이하인 경우 해당 경우의 학습을 중단하였으며, 모델 간 상호 교차 검증을 수행하기 위하여 $_8C_2{\times}_8C_2{\times}_8C_2$ 경우의 수에 대한 Verification test를 수행하였다. Calibration 과정 상 모든 경우에 대하여 10% 이하의 Cost function 값을 보였으나, 검증 테스트 과정에서 모든 경우에 대하여 $r^2$ < 0.5 인 결정 계수 값이 나타났다. 단적으로 심층학습 모델의 과도한 적합(Over fitting) 방식의 한계를 보인 것이라 판단할 수 있다. 적합한 Feature selection 및 심층 학습 모델에 대한 지속적이고 추가적인 고려를 통해 과도적합을 해소함과 동시에 실효적이고 활용 가능한 Classification을 위한 입, 출력 노드 단의 전후 Indexing, Quantization에 대한 고려가 필요할 것이다. 이를 통해 모돈 생체 정보 정량화를 위한 지능형 현장 진단 기술 연구를 지속할 것이다.

  • PDF

A Study on the Effect of Cooperative Computer-Assisted Instruction by Previous Achievement Level (사전 성취 수준에 따른 협동적 컴퓨터 보조 수업의 효과)

  • No, Tae-Hui;Cha, Jeong-Ho;Yun, Seon-Ae;Gang, Seok-Jin
    • Journal of the Korean Chemical Society
    • /
    • v.46 no.4
    • /
    • pp.377-384
    • /
    • 2002
  • In this study, the effect of cooperative computer-assisted instruction upon students' conceptual under-standing,application ability, and learning motivation were investigated by a previous achievement level. The treatment and the control groups (2 classes) were selected from a middle school in Seoul, and taught about the motion of molecule for 5 class periods. Prior to the instructions, a learning motivation test was administered and used as covariate. The scores of a previous achievement test were also used as covariate. The scores of the mid-term science examination were used as blocking variable. After the instructions, the conceptions test, the application test, and the learning motivation test were administered. Two-way ANCOVA results revealed that there were no significant differences in the scores of the con-ceptions test and the application test. However, the scores of the treatment group were found to be significantly higher than those of the control group in the learning motivation test.

Study on the Unsuitable Case for the Hierarchy of the Curriculum through the Analysis of the Science Teaching-Learning Lesson Plan Focused on 'Three States of Matter' Unit of the 7th Grade (과학과 교수.학습 과정안의 분석을 통한 교육 과정 위계 부적합 사례 연구 - 7학년 물질의 세 가지 상태)

  • Seo, Min-Hee;Choi, Won-Ho
    • Journal of the Korean Chemical Society
    • /
    • v.53 no.6
    • /
    • pp.784-792
    • /
    • 2009
  • In this study, we investigated the unsuitable cases for the hierarchy of the curriculum in science 'teaching-learning lesson plan' which is on the web site of the each city's support center for teaching and learning and Seoul Science Park with a purpose of giving helpful data for science teachers at lessons. The investigation is limited to 'Three states of matter' unit for the 7th grade and the content elements used at analysis was the "state changes of matter", "molecule" and "molecular model". The case found for unsuitable case for the hierarchy of the curriculum was phase equilibrium, ice crystal theory, peculiar property of water, classification of solid, thermal energy and physical change at the "state changes of matter". While the "molecular model" showed molecular motion and density. On the other hand, no case was found at "molecule".

The effects of CAI adapting to the level of students' conceptual understanding in concept learning (개념 학습에서 학생들의 개념 이해 수준에 적응적인 CAI의 효과)

  • Kim, Kyung-sun;Kang, Yi-young;Kwon, Hyeok-soon;Wang, Hye-nam;Noh, Tae-hee
    • The Journal of Korean Association of Computer Education
    • /
    • v.9 no.2
    • /
    • pp.79-88
    • /
    • 2006
  • This study investigated the effects of computer-assisted instruction adapting to the level of students' conceptual understanding upon students' conceptual understanding, retention of conceptions, learning motivation, and perception about computer-assisted instruction in concept learning. 94 seventh grade students from a coed middle school in Seoul were randomly assigned to control, CAI, adaptive CAI groups, and were taught about 'motion of molecules' for 7 class periods. Two-way ANCOVA results revealed that the scores of a conception test and a learning motivation test for the adaptive CAI group were significantly higher than those for the control group. The scores of a retention test of conceptions for the adaptive CAI group were significantly higher than those for other two groups. There were no significant interactions between the instruction and the level of previous achievement in the scores of the conception test, the learning motivation test, and the retention test of conceptions. The perception about computer-assisted instruction for the students of the adaptive CAI group were more positive than those for the students of the CAI group.

  • PDF