• Journal of the Korean Society of Radiology
• /
• v.16 no.5
• /
• pp.595-602
• /
• 2022

This study tried to suggest the most efficient learning rate for accurate and efficient automatic diagnosis of medical images for chest X-ray pneumonia images using deep learning. After setting the learning rates to 0.1, 0.01, 0.001, and 0.0001 in the Inception V3 deep learning model, respectively, deep learning modeling was performed three times. And the average accuracy and loss function value of verification modeling, and the metric of test modeling were set as performance evaluation indicators, and the performance was compared and evaluated with the average value of three times of the results obtained as a result of performing deep learning modeling. As a result of performance evaluation for deep learning verification modeling performance evaluation and test modeling metric, modeling with a learning rate of 0.001 showed the highest accuracy and excellent performance. For this reason, in this paper, it is recommended to apply a learning rate of 0.001 when classifying the presence or absence of pneumonia on chest X-ray images using a deep learning model. In addition, it was judged that when deep learning modeling through the application of the learning rate presented in this paper could play an auxiliary role in the classification of the presence or absence of pneumonia on chest X-ray images. In the future, if the study of classification for diagnosis and classification of pneumonia using deep learning continues, the contents of this thesis research can be used as basic data, and furthermore, it is expected that it will be helpful in selecting an efficient learning rate in classifying medical images using artificial intelligence.

• Korean Journal of Cognitive Science
• /
• v.20 no.3
• /
• pp.357-381
• /
• 2009

Although it is known that there are personality characteristic variances in the sensitivity to environmental feedback, the trait individual difference has scarcely been explored in the context of recognition memory decision. The present study investigated this issue by examining the relationship between the feedback-based adaptive flexibility of recognition criterion positioning and personality differences in general sensitivity to non-laboratory outcomes. Experiment 1 demonstrated that veridical feedback itself had little effect on the recognition decision criterion whereas Experiment 2 demonstrated that biased feedback manipulations selectively restricted to high confidence errors, induced shifts even in the overall Old/New category criterion. Critically, individual differences in stable personality characteristic linked to reward seeking(Behavioral Activation System-BAS) and anxiety avoidance (Behavioral Inhibition System-BIS) has been shown to predict the sensitivity of subjects to this form of feedback-induced criterion learning. This data further support the idea that incremental reinforcement-based learning mechanism not often considered important during explicit recognition decisions may play a key role in criterion setting.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.4
• /
• pp.7-16
• /
• 2023

For decades, creating a desired locomotive motion in a goal-oriented manner has been a challenge in character animation. Data-driven methods using generative models have demonstrated efficient ways of predicting long sequences of motions without the need for explicit conditioning. While these methods produce high-quality long-term motions, they can be limited when it comes to synthesizing motion for challenging novel scenarios, such as punching a random target. A state-of-the-art solution to overcome this limitation is by using a GAN Discriminator to imitate motion data clips and incorporating reinforcement learning to compose goal-oriented motions. In this paper, our research aims to create characters performing combat sports such as boxing, using a novel reward design in conjunction with existing GAN-based approaches. We experimentally demonstrate that both the Adversarial Motion Prior [3] and Adversarial Skill Embeddings [4] methods are capable of generating viable motions for a character punching a random target, even in the absence of mocap data that specifically captures the transition between punching and locomotion. Also, with a single learned policy, multiple task controllers can be constructed through the TimeChamber framework.

• Smart Media Journal
• /
• v.7 no.2
• /
• pp.47-53
• /
• 2018

Along with the development of information and communication technology, smart education that can learn without restrictions of time, place and equipment is activated even in the field of education. Although smart education is provided with content-based training solutions, construction of a system that grasps individual characteristics of learners and provides personalized learning is relatively weak. The activity of free choice is an important play activity of early childhood education, but it is not implemented efficiently by relying on the clinical observation of the teacher. If the IoT(Internet of Things) technology based on Hyper-Connected is applied to free-choice activities, it is possible to provide the child's personalized activity type and play-form analysis based on objective and stylized data. In this paper, we design and implement a system to monitor the child's activity of free choice by building an IoT environment that is based on open source hardware. The proposed system provides children's activity information as objective data and will be used as teacher's work mitigation and custom training material for each child.

• Journal of Korean Home Economics Education Association
• /
• v.26 no.3
• /
• pp.113-131
• /
• 2014

This study aimed to explore the effects of private tutoring expenses, parents' monitoring affection, their children's learning value and self-regulated learning abilities on middle-school boys' and girls' English Math academic achievement. The subjects were the 3rd middle-school 1,123 students taking the private tutoring of English and Math who participated in the Korea Child Youth Panel Surveys(KCYPS). The data were analyzed with descriptive statistics, correlations and hierarchical regressions. The main results of this study were as follows. Firstly, regardless of middle-school students' sex, as monthly average private tutoring expenses were more, the levels of parents' monitoring, and their children's learning value self-regulated learning abilities were higher, so middle-school students' academic achievement was higher. Secondly, regardless of middle-school students' sex, their self-regulated learning abilities were the highest predictors of English Math achievement. Also, their learning value and parents' monitoring influenced middle-school boys' English Math achievement in order. On the other hand, monthly average private tutoring expenses influenced middle-school girls' English Math achievement. Furthermore there were no moderating effects of parents' monitoring affection, their children's learning value and self-regulated learning abilities between monthly average private tutoring expenses and middle-school boys' and girls' English Math achievement. Finally, based on the results, the importance of parents and Home Economics was suggested in attaining middle-school students' higher academic achievement. Especially, Home Economics can play an important role of enhancing middle-school students' self-regulated learning abilities and learning value necessary for middle-school students' higher academic achievement.

• The Journal of the Acoustical Society of Korea
• /
• v.43 no.2
• /
• pp.214-224
• /
• 2024

Sonobuoys are disposable devices that utilize sound waves for information gathering, detecting engine noises, and capturing various acoustic characteristics. They play a crucial role in accurately detecting underwater targets, making them effective detection systems in anti-submarine warfare. Existing sonobuoy deployment methods in multistatic systems often rely on fixed patterns or heuristic-based rules, lacking efficiency in terms of the number of sonobuoys deployed and operational time due to the unpredictable mobility of the underwater targets. Thus, this paper proposes an optimal sonobuoy placement strategy for Unmanned Aerial Vehicles (UAVs) to overcome the limitations of conventional sonobuoy deployment methods. The proposed approach utilizes reinforcement learning in a simulation-based experimental environment that considers the movements of the underwater targets. The Unity ML-Agents framework is employed, and the Proximal Policy Optimization (PPO) algorithm is utilized for UAV learning in a virtual operational environment with real-time interactions. The reward function is designed to consider the number of sonobuoys deployed and the cost associated with sound sources and receivers, enabling effective learning. The proposed reinforcement learning-based deployment strategy compared to the conventional sonobuoy deployment methods in the same experimental environment demonstrates superior performance in terms of detection success rate, deployed sonobuoy count, and operational time.

• The Korean Journal of Coaching Psychology
• /
• v.6 no.2
• /
• pp.75-92
• /
• 2022

There are many studies that intrinsic learning motivation and social support play an important role in the study of children and adolescents. However, studies examining the effects of intrinsic learning motivation and social support by measuring the actual academic performance of elementary school students are rare. This study attempted to verify the effect of intrinsic learning motivation on Korean language performance and moderating effect of social support in 5th and 6th graders in elementary school. 122 elementary school students in local county-level areas participated in this study. The Korean language test was conducted about 5 months after intrinsic learning motivation and social support of families and teachers were measured. The results revealed that Korean language performance showed a significant positive correlation with intrinsic learning motivation and social support, and also showed a significant correlation between learning motivation and social support. In the regression analysis with control variables, it was found that intrinsic learning motivation had a significant effect on Korean language performance. The moderating effect of social support was analyzed by dividing it into family support and teacher support. The interaction effect of learning motivation and social support was significant only in teacher support, not in family support. In specific, when teacher support was high, Korean language performance was high regardless of the student's learning motivation level, but when teacher support was low, the student's learning motivation mattered in the performance. Based on the results of this study, implications and limitations were discussed.

• Smart Media Journal
• /
• v.10 no.2
• /
• pp.76-83
• /
• 2021

With the development and use of smart devices such as smartphones and tablets increases, the mobile application market based on mobile devices is growing rapidly. Mobile application users write reviews to share their experience in using the application, which can identify consumers' various needs and application developers can receive useful feedback on improving the application through reviews written by consumers. However, there is a need to come up with measures to minimize the amount of time and expense that consumers have to pay to manually analyze the large amount of reviews they leave. In this work, we propose to collect delivery application user reviews from Google PlayStore and then use machine learning and deep learning techniques to classify them into four categories like application feature advantages, disadvantages, feature improvement requests and bug report. In the case of the performance of the Hugging Face's pretrained BERT-based Transformer model, the f1 score values for the above four categories were 0.93, 0.51, 0.76, and 0.83, respectively, showing superior performance than LSTM and GRU.

• Journal of Biomedical Engineering Research
• /
• v.43 no.5
• /
• pp.361-368
• /
• 2022

The purpose of this study was to develop a hand rehabilitation training system for hemiplegic patients. We also tried to find out five hand postures (WF: Wrist Flexion, WE: Wrist Extension, BG: Ball Grip, HG: Hook Grip, RE: Rest) in real-time using multi-channel EMG-based deep learning. We performed a pre-processing method that converts to Spider Chart image data for the classification of hand movement from five test subjects (total 1,500 data sets) using Convolution Neural Networks (CNN) deep learning with an 8-channel armband. As a result of this study, the recognition accuracy was 92% for WF, 94% for WE, 76% for BG, 82% for HG, and 88% for RE. Also, ten physical therapists participated for the usability evaluation. The questionnaire consisted of 7 items of acceptance, interest, and satisfaction, and the mean and standard deviation were calculated by dividing each into a 5-point scale. As a result, high scores were obtained in immersion and interest in game (4.6±0.43), convenience of the device (4.9±0.30), and satisfaction after treatment (4.1±0.48). On the other hand, Conformity of intention for treatment (3.90±0.49) was relatively low. This is thought to be because the game play may be difficult depending on the degree of spasticity of the hemiplegic patient, and compensation may occur in patient with weakened target muscles. Therefore, it is necessary to develop a rehabilitation program suitable for the degree of disability of the patient.

• Education of Primary School Mathematics
• /
• v.12 no.1
• /
• pp.39-55
• /
• 2009

For this study, the 'Game Activities' lessons presented in the math textbooks from the 1st grade to the 6th were examined in terms of learning materials, the learning members' make-up, the playing structures, and the relation with the contents. In addition, the survey by means of questionnaires was conducted to analyze the actual condition of teachers' guidance in the field. The findings from this research were as follows: First, as for the activities presented in the textbooks, it turned out that too much emphasis is placed upon plays mainly using learning materials such as cards and dice played by teams of two. In addition, there have been shown negative aspects in various ways of plays putting too much emphasis on certain types of plays such as and structures. As for the relation with the contents, although lots of efforts were taken to connect the playing activity to the lesson contents, there were units presenting plays based on the preceding lesson's repeated activity, ones that have weak link with the contents. Second, it turned out that the teachers had negative attitude on the guidance using the 'Game Activities' lesson, although they were aware of the effects of playing in math learning. This seemed to result from the delicate variety and insufficient preparation for the play. Besides, the findings indicate that the appreciation and activity of the 'Game Activities' lesson presented as a way of performance evaluation. for play need to be provided in school or classrooms for teachers and students to make good use of them.