• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.1
• /
• pp.136-141
• /
• 2022

A very complex task in deep learning such as image classification must be solved with the help of neural networks and activation functions. The backpropagation algorithm advances backward from the output layer towards the input layer, the gradients often get smaller and smaller and approach zero which eventually leaves the weights of the initial or lower layers nearly unchanged, as a result, the gradient descent never converges to the optimum. We propose a two-factor non-saturating activation functions known as Bea-Mish for machine learning applications in deep neural networks. Our method uses two factors, beta (𝛽) and alpha (𝛼), to normalize the area below the boundary in the Mish activation function and we regard these elements as Bea. Bea-Mish provide a clear understanding of the behaviors and conditions governing this regularization term can lead to a more principled approach for constructing better performing activation functions. We evaluate Bea-Mish results against Mish and Swish activation functions in various models and data sets. Empirical results show that our approach (Bea-Mish) outperforms native Mish using SqueezeNet backbone with an average precision (AP50val) of 2.51% in CIFAR-10 and top-1accuracy in ResNet-50 on ImageNet-1k. shows an improvement of 1.20%.

• Journal of Industrial Convergence
• /
• v.21 no.1
• /
• pp.147-157
• /
• 2023

In this study, we approached from the perspective of AI convergence education in elementary, middle and high schools to understand AI convergence education. We examined what capabilities AI convergence education ultimately seeks to pursue, and analyzed various examples of AI convergence education in three dimensions: core curriculum, convergence model, AI learning elements and learning activities. In addition, factors to be considered in order for AI convergence education to be actively carried out include the cultivation of AI convergence education capabilities of teachers, the development and dissemination of AI teaching and learning methods and teaching and learning models, and evaluation methods for AI convergence education.

• Communications for Statistical Applications and Methods
• /
• v.30 no.5
• /
• pp.485-499
• /
• 2023

This study aimed to predict the number of meals served in a group cafeteria using machine learning methodology. Features of the menu were created through the Word2Vec methodology and clustering, and a stacking ensemble model was constructed using Random Forest, Gradient Boosting, and CatBoost as sub-models. Results showed that CatBoost had the best performance with the ensemble model showing an 8% improvement in performance. The study also found that the date variable had the greatest influence on the number of diners in a cafeteria, followed by menu characteristics and other variables. The implications of the study include the potential for machine learning methodology to improve predictive performance and reduce food waste, as well as the removal of subjective elements in menu classification. Limitations of the research include limited data cases and a weak model structure when new menus or foreign words are not included in the learning data. Future studies should aim to address these limitations.

• Journal of Korean society of Dental Hygiene
• /
• v.23 no.6
• /
• pp.501-509
• /
• 2023

Objectives: This study aimed to analyze the relation between factors related to the clinical performance competency of dental hygiene students and their clinical learning environment. Methods: The study conducted a survey of dental hygiene students from October 18 to 30, 2023. The data were analyzed using one way analysis of variance, t-test, and stepwise multiple regression. Results: The total practicum lasted ≤10 weeks, 11-15 weeks, and ≥16 weeks for 41.7%, 33.5%, and 24.8% of the students, respectively. Half of them had experience at only one clinical institution. Clinical learning environment had an average score of 3.46 points, whereas the average clinical performance competency of the participants was 3.60 points. The major influencing factors on clinical performance competency were identified as preceptor' s guidance (β=0.277), work participation opportunities (β=0.213), and perceived importance of clinical practice (β=0.136). Conclusions: Efforts are required to provide students with prior education on the importance of clinical practice, improve the clinical learning environment with a focus on preceptor's guidance and work participation opportunities. And standardize various elements to resolve differences in the practice of clinical institutions across regions.

• Hwankyungkyoyuk
• /
• v.20 no.4
• /
• pp.12-26
• /
• 2007

The purpose of this research is to develop a water environmental education (EE) program using streams, based on the core ideas of ENVISION and materializing elements that were extracted in this research. This research realized the elements and presented a model of the water EE program using a local stream. First, this research developed a basic model of a water EE program using streams by extracting 10 materializing elements and realizing the elements in 4 stage-procedural model. The 10 materializing elements were 1. experiencing the process of inquiry, 2. inquiring local environments, 3. self-directing learning and mutual interaction with colleagues, 4. collecting real data and interpreting, 5. utilizing the ICT(information and communication technology), 6. inquiring with the view point of the 'Environmental Studies for EE', 7. inquiring with the watershed concept, 8. inquiring with the integrating and the holistic view point, 9. pursuing the macroscopic understanding about environment, and 10. connecting the real world phenomena with the environmental concepts and theories. This research materialized these 10 elements in 4 stage model, following the previous ENVISION research, which are 1. preparing stage and visual assessment, 2. writing the report of the inquiry plan, 3. collecting the real data in the environment and performing the investigation, and 4. presenting the inquiry results. Second, with using this basic model, this research developed and presented a model of the specific water EE program using a case stream called 'Baig Cheon' stream, which is a local stream. This research is considered to have a considerable meaning in developing a EE program with ENVISION ideas for the watershed concept and inquiry with environmental science using local streams. The developed model can help the professional development of teachers and teacher education of water EE.

• Journal of Korea Game Society
• /
• v.14 no.5
• /
• pp.7-14
• /
• 2014

Since educational and fun elements coexist in an educational game, it is important to come up with standards that can elicit fun with the game as well as educational standards. Many of the existing developed educational games have received low marks in the characteristics of entertainment and game design. Thus, how to make up for their entertainment characteristics is important. This paper analyzes gaming elements of puzzle games that were released for smart phones and gained massive popularity from their users and will apply them to educational games. Such games that have gained popularity and won a commercial success include Anipang, Anipang 2, Candy Crush Saga, Farm Heroes Saga and Pokopang. This study will analyze gaming elements of these games. By analyzing the educational games, it is applied to the game elements of casual games. In addition, we have proposed a element that can enhance the immersive. Game design elements that can improve entertainment characteristics include game rules, stages and missions. Elements that can increase immersion include items, background maps and new learning environments. This paper has significance in that it analyzes entertainment characteristics, which mean the fun of games through those that have been a commercial success and applies them to educational games.

• The Journal of Sustainable Design and Educational Environment Research
• /
• v.7 no.1
• /
• pp.39-55
• /
• 2008

The purpose of this study is to suggest improvement of certification standards of environment friendly architecture in ecological environment field by surveying and analyzing actual status of actual certification schools of environment friendly architecture per ecological environment elements through extracting ecological environment elements which is possible to analyze in quantity certification standards of environment friendly architecture. As contents of the study, first, certification as an examination on certification system of environment friendly architecture, summary of certification system of environment friendly architecture, ecological environment elements which is possible for quantitative analysis of ecological environment certification standards were extracted. Second, actual status of ecological environment elements per school grade, region, and population increase by collecting actual data of certification schools of environment friendly architecture. Third, using analysis data of ecological environment elements, ecological area ratio of certification schools of environment friendly architecture per weighted values of space types was calculated and the results were analyzed as the follows. (1) In case of preliminary certification out of total objects of investigation schools was average 34.06% but in this case of certification, it was reduced to 33.07%. However, it is difficult to make judgment as actual evidence data since ecological area ratio of actual preliminary certification is data based on drawing planes. Therefore, in case of this certification, elementary schools was investigated at average 33.11%, middle schools 34.81%, and high schools 31.28% and middle school showed the highest relatively. (2) Besides, average composition ratio of ecological area ratio was 63.04% in case of this certification regardless of school grade and region and it was analyzed that about 70% of ecological area ratio is composed of the natural based green area including pure natural based green land and 6.89% of ecological learning institute.

• Journal of Digital Convergence
• /
• v.12 no.9
• /
• pp.313-323
• /
• 2014

Recently, there has been an effort to convert standardized educational system to customized one considering students' individual differences and aptitude. Flipped learning has been regarded as a teaching method fitting to this movement and has gained much interest by educators. Despite its increasing interest, its studies fall short in the quantity above all. This study investigated the concept of flipped learning and explored its educational value in terms of student' and teacher' perspectives. The study also examined the requirements to apply flipped learning to its practice. Flipped learning was found to accomplish mastery learning and to promote interaction between teacher and students and also peer-to-peer interaction. It was expected to reduce the burden of teacher's duties and secure his or her interest in students and the tutorial time for individual students. Essential elements in its application in practice included students' voluntary participation, the consent of their parents and support from their school and government. In addition, teachers' professional development was found to be important to improve their understanding and application of flipped learning in practice.

• Journal of Science Education
• /
• v.46 no.3
• /
• pp.278-292
• /
• 2022

Researchers in science education have recently focused on service learning as an important educational approach to address environmental challenges including climate change. However, there are a limited number of studies focused on service learning in science education in Korea. In this study, based on existing related literature, the theoretical background of service learning is introduced and a conceptual framework that can be applied in the science education environment in Korea as a new approach to address environmental challenges is proposed. Given this, 112 relevant academic documents collected by searching a global database were analyzed. As a result, the theoretical background of service learning rooted in Dewey's experiential learning is introduced. Specifically, service learning is defined as a form of experiential learning that supports students in helping communities with structured services related to learning content, while at the same time fostering their in-depth understanding of content and engaging in civic and social responsibility as members of their community. In addition, a conceptual framework for service learning to address environmental challenges that can be applied to the science education environment in Korea is proposed. The proposed framework classifies the members of service learning into schools, students, and communities, and presents knowledge, experience, and learning through critical reflection as its core elements. It is important to actively implement service learning in science education to address various environmental challenges including climate change, biodiversity loss, air pollution, and deforestation, as just a few pressing challenges. Therefore, based on the proposed conceptual framework for service learning, studies that apply and verify diverse forms of educational approaches to addressing environmental challenges in local communities at school sites are needed.

• Journal of The Korean Association of Information Education
• /
• v.11 no.1
• /
• pp.1-9
• /
• 2007

The purpose of this study is to develope a criterion for evaluation on game-type learning programs by the application of HCI(Human Computer Interaction) theory. And to analyze game-type learning programs for elementary students on the criterion developed in this study. The HCI theory, which deals with principles or methods for developing systems people can use conveniently and pleasantly, has been applied to overall area of program development. And it also has been widely used to evaluate learning programs. However, there have been few studies on a game-type learning program evaluation on the basis of the HCI theory. This paper shows that evaluation criteria are developed on three viewpoint bases : usefulness, usability, and affect which are as elements in HCI. And analyzes the game-type learning programs from these three points of view. The evaluation criteria developed in this study can be applied to a basis for evaluation on game-type learning programs, and the analysis will be able to be a useful guide to game programers as well as its users.