• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.6
• /
• pp.179-186
• /
• 2008

In this paper we analyse the effects of Bayesian algorithm in teaming class imbalance problems and compare the performance evaluation methods. The teaming performance of the Bayesian algorithm is evaluated over the class imbalance problems generated by priori data distribution, imbalance data rate and discrimination complexity. The experimental results are calculated by the AUC(Area Under the Curve) values of both ROC(Receiver Operator Characteristic) and PR(Precision-Recall) evaluation measures and compared according to imbalance data rate and discrimination complexity. In comparison and analysis, the Bayesian algorithm suffers from the imbalance rate, as the same result in the reported researches, and the data overlapping caused by discrimination complexity is the another factor that hampers the learning performance. As the discrimination complexity and class imbalance rate of the problems increase, the learning performance of the AUC of a PR measure is much more variant than that of the AUC of a ROC measure. But the performances of both measures are similar with the low discrimination complexity and class imbalance rate of the problems. The experimental results show 4hat the AUC of a PR measure is more proper in evaluating the learning of class imbalance problem and furthermore gets the benefit in designing the optimal learning model considering a misclassification cost.

• Proceedings of the KAIS Fall Conference
• /
• 2006.11a
• /
• pp.239-241
• /
• 2006

비디오 데이터를 효율적으로 처리하기 위해서는 비디오 데이터가 가지고 있는 내용에 대한 정보를 데이터베이스에 저장하고 사용자들의 다양한 질의를 처리할 수 있는 의미기반 검색 기법이 요구된다. 본 논문에서는 사용자의 키워드 학습과 비교 영역 학습을 이용하여 대용량의 비디오 데이터에 대한 사용자의 다양한 의미검색을 지원하는 에이전트 기반에서의 자동화된 비디오 검색 시스템을 제안한다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.11
• /
• pp.305-313
• /
• 2019

Recently, spatial information is being constructed actively based on the images obtained by drones. Because occlusion areas occur due to buildings as well as many obstacles, such as trees, pedestrians, and banners in the urban areas, an efficient way to resolve the problem is necessary. Instead of the traditional way, which replaces the occlusion area with other images obtained at different positions, various models based on deep learning were examined and compared. A comparison of a type of feature descriptor, HOG, to the machine learning-based SVM, deep learning-based DNN, CNN, and RNN showed that the CNN is used broadly to detect and classify objects. Until now, many studies have focused on the development and application of models so that it is impossible to select an optimal model. On the other hand, the upgrade of a deep learning-based detection and classification technique is expected because many researchers have attempted to upgrade the accuracy of the model as well as reduce the computation time. In that case, the procedures for generating spatial information will be changed to detect the occlusion area and replace it with simulated images automatically, and the efficiency of time, cost, and workforce will also be improved.

• Journal of Multimedia Information System
• /
• v.7 no.1
• /
• pp.1-10
• /
• 2020

The image analysis is an important and predominant task for classifying the different parts of the image. The analysis of complex image analysis like histopathological define a crucial factor in oncology due to its ability to help pathologists for interpretation of images and therefore various feature extraction techniques have been evolved from time to time for such analysis. Although deep reinforcement learning is a new and emerging technique but very less effort has been made to compare the deep learning and deep reinforcement learning for image analysis. The paper highlights how both techniques differ in feature extraction from complex images and discusses the potential pros and cons. The use of Convolution Neural Network (CNN) in image segmentation, detection and diagnosis of tumour, feature extraction is important but there are several challenges that need to be overcome before Deep Learning can be applied to digital pathology. The one being is the availability of sufficient training examples for medical image datasets, feature extraction from whole area of the image, ground truth localized annotations, adversarial effects of input representations and extremely large size of the digital pathological slides (in gigabytes).Even though formulating Histopathological Image Analysis (HIA) as Multi Instance Learning (MIL) problem is a remarkable step where histopathological image is divided into high resolution patches to make predictions for the patch and then combining them for overall slide predictions but it suffers from loss of contextual and spatial information. In such cases the deep reinforcement learning techniques can be used to learn feature from the limited data without losing contextual and spatial information.

• Journal of The Korean Digital Architecture Interior Association
• /
• v.12 no.3
• /
• pp.101-109
• /
• 2012

Recently, the roof garden of the large commercial facilities has been changed as a complex cultural space follow the users' diverse needs. It is focused on large commercial facilities on Seoul and Gyeonggi area but large commercial facilities of the province still has not been able to respond to the needs of users. Thus, the present condition of roof gardens was compared space components with physical environmental factors in Seoul, Gyeonggi and Daejeon, Chungcheong. Throughout the comparison, roof gardens in Daejeon and Chungcheong area were focused on finding ways to improve As a result, roof gardens in Seoul and Gyeonggi area were applied to a variety of factors in space components experience, learning, performance, exhibition and etc. It is composed to take place of various acts in the roof garden. On the other hand, roof gardens in Daejeon and Chungcheong were not applied a variety space components by comparing Seoul and Gyeonggi area. Physical environmental factors also were insufficient by comparing Seoul and Gyeonggi area the landscape, awning facilities, amenities for handicapped. As described above, space components and physical environmental factors in Daejeon and Chungcheong area roof gardens were insufficient. To perform the role of the roof garden changing as a complex cultural space and to confront to the users' needs space component such as learning, performance, exhibition should be considered during planning. And physical environmental factors supportable space components should be considered together.

• Journal of the Korean Society of Earth Science Education
• /
• v.8 no.3
• /
• pp.399-408
• /
• 2015

This study is to find out that the effects of a creative experience activity program to scientific concepts and self-directed learning skills. This study has been aimed at 2 class 40 students of 4th grade in D metropolitan city A elementary school in 2015, one class 20 students are the research group to apply Scientific research program using creative experience activity, another class 20 students were comparison groups to apply general science classes. The related class section of this study is 4th grade 2 semester of science 4 chapters, 'The Earth and the moon' This section is in fourth grade elementary science curriculum revision in 2009 is a Sections to learn for the first time about astronomical area. Target research group in club activities as part of the creative activities implemented using scientific inquiry and analyzed the results. In addition, in order to better research based on the results of this study as follows. First, the science curriculum in elementary schools, as well as applied research about the creative experience activity classes in other subjects is required. The ongoing research is needed to classes utilizing the characteristics of creative experience activities in several subjects of the elementary school curriculum. Second, Creative experiential learning is only effective when it is done consistently, it is worth studying for long periods of time.

• Journal of Gifted/Talented Education
• /
• v.13 no.3
• /
• pp.1-17
• /
• 2003

Development and evaluation of distance learning for the gifted students in science and mathematics In this study, we developed and administrated the distance learning for the gifted students in science and mathematics, and analysed their responses. To do this, four types of teaching programs - lectures using program for distance learning, practice activities using simulation program, tasks solving programs based on discussions, and problem solving activities - were developed and students responses were analysed in eight area - stimulus, difficulties, structure, learning circumstances, involvement, interaction, learning outcomes, comparison with other learning -. As results, it was found that many students responded positively and thought programs helped their creativity, logical thinking, intelligent ability, and information searching ability. Students preferred practice activities based on appropriate guidances to lectures providing detailed explanations. And interaction could be stimulated by inducing discussion.

• Journal of Elementary Mathematics Education in Korea
• /
• v.16 no.2
• /
• pp.227-252
• /
• 2012

Given the lack of research on the measurement domain, this paper analyzed the statements related to length and area in the curricular materials developed under the 7th and the current mathematics curriculum in terms of the degree of guidance and the key learning elements of measurement. The results showed that despite the similarity of the most prevalent guidance type and learning elements, the current materials used open-ended or combined types in place of guided types and employed measurement reasoning and components while decreasing mere calculation in measurement, in comparison with the previous textbooks and workbooks. This paper close with implications on the revision of curricular materials related to the measurement domain as well as methodological suggestions of textbook analysis.

• Education of Primary School Mathematics
• /
• v.8 no.1
• /
• pp.33-50
• /
• 2004

This study is intended to investigate the effect on the development of multiplicative thinking and multiplicative ability by teaching repeated addition, rate, comparison, area-array, and combination problems. Two research questions are established: first, is there any difference of multiplicative thinking between the experimental group(the modeling of problem situation learning group) and the control group(the traditional learning group)\ulcorner Second, is there any difference of multiplicative ability between the experimental group and the control group\ulcorner The treatment process for the experimental group is based on modeling problem situations for nine lesson periods. In order to answer the research questions the chi-square analysis was used for the first research question and the t-test was used for the second one. The findings are summarized as follows: there is no significant difference of multiplicative thinking be1ween the experimental and the control group but there is significant difference of multiplicative ability.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.4
• /
• pp.519-524
• /
• 2009

Recently, reinforcement learning methods have drawn much interests in the area of machine learning. Dominant approaches in researches for the reinforcement learning include the value-function approach, the policy search approach, and the actor-critic approach, among which pertinent to this paper are algorithms studied for problems with continuous states and continuous actions along the line of the actor-critic strategy. In particular, this paper focuses on presenting a method combining the so-called ACFRL(actor-critic fuzzy reinforcement learning), which is an actor-critic type reinforcement learning based on fuzzy theory, together with the RLS-NAC which is based on the RLS filters and natural actor-critic methods. The presented method is applied to a control problem for crawling robots, and some results are reported from comparison of learning performance.