• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.5
• /
• pp.251-257
• /
• 2024

Securing various types of training data in image Classification is important for improving performance. However, increasing the amount of original data is cost-limited, so data diversity can be secured by transforming images through data augmentation. Recently, a new deep learning approach using Generative AI models like GAN or Diffusion Based models has emerged in the Data Augmentation task, and reinforcement learning-based methods such as AutoAugment and Deep AutoAugment using existing basic Augmentation techniques are also showing good performance. However, this method has the disadvantage of having a complicated optimization procedure and high computational cost. This paper conducted various experiments with existing methods Cutmix, Mixup, RandAugment. By combining these techniques appropriately, we obtained higher performance than existing method without much effort. Additionally, our ablation experiment shows that additional hyper-parameter adjustments are needed for the basic augmentation types when we use RandAugment. Our code is available at https://github.com/lliee1/Randomness_Analysis.

• Journal of Digital Contents Society
• /
• v.18 no.2
• /
• pp.303-309
• /
• 2017

The purpose of this study is to determine the cause and result of the smartphone use in the classroom, focusing on students from Child Care departments in colleges. The targets of this study were 10 college students from child care departments and data was collected through in-depth interviews. The results of the study are as follows. [Cause of use] was divided into and . [Negative result] has been identified by and . [positive result] has been identified by < Learning Support >, < Memory Support >. The study is believed to contribute to identifying the problems of smartphone use of college students in the future and contributing to the provision of basic materials for the use of favorable purposes.

• Journal of the Korean earth science society
• /
• v.34 no.7
• /
• pp.727-737
• /
• 2013

The purpose of this study was to investigate the middle school students' perception and their learning satisfaction in SMART learning based science instruction. Three types of modules on the solar system and lunar phases unit at the middle school level were developed and lessons on each module were taught to 207 student participants. All participants were provided with tabletPC(iPad2) with iOS5 installed, and using astronomy app Solar Walk, mirroring function, QR code, and Google Presentation, the lessons were carried out both in classroom and at home. The instrument for assessing students' perception on the SMART learning-based instruction was developed based on 4 factors including Self-directed, Motivation, Adaptiveness, and Technology Embedded, with a Likert scale from 1-5 on 20 items. The learning satisfaction survey instrument was originally from Keller's work (1987), and its test items were adapted and modified. To reveal the perception and learning satisfaction about SMART learning-based science lessons, the participants were comparatively analyzed by gender and science achievement levels. Results indicated that male students showed positive perception for the SMART learning-based instruction. Group with higher science achievement scores showed more positive perception of the SMART learning-based instruction in terms of Self-directed and Motivation factor. Also, the learning satisfaction of male students was higher than female students and group with higher academic ability more satisfied with the SMART learning-based instruction than the low group. The results provide implications for future development of programs and help set a direction of increasing the use of a SMART learning-based science in school.

• Journal of the Korean Society of Earth Science Education
• /
• v.11 no.3
• /
• pp.203-211
• /
• 2018

The purpose of this study is to investigate the effects of Project-Based Learning on Science Concepts and Science Learning Motivation. This particular study was proceeded to 4th grader at S elementary school, there was a mutual agreement with a homeroom teacher about assigning a research group and comparison group and it was agreed to students by explaining the reason and purpose of the study. There searcher visited in person to pick 22 students for research group and another 20 students for comparison group. For a research group, an experimental group, homeroom teacher, proceeded a science class with the application of Project-Based Learning. The experimental period was set up as a 40 minutes class unit for 12 weeks. After an experimental group, Science Concepts and Science Learning Motivation were examined, data collection and data analysis were proceeded by order. The following experimental results are as below. First, the application of Project-Based Learning method in a class was effective in improvement of Science Concepts acquisition. Second, the application of Project-Based Learning method in a class was effective in cultivation of Science Learning Motivation. Third, the application of Project-Based Learning method in a class had a positive cognition from the learners in the experimental group. Based on the discussions and implications of the results of this study, some suggestions in the follow - up study are as follows. First, applying Project-Base Learning to various science lessons and learning effects can be suggested as one of the new teaching methods. Second, the use of the Project-Based Learning to test the effects of elementary school students' different grades may be regarded as another teaching method for science class.

• Advances in Energy Research
• /
• v.5 no.1
• /
• pp.1-12
• /
• 2017

The world today has reached a certain level where it is impossible to get the quality education at the tertiary level without the use of Information and Communication Technology (ICT). ICT has made life better, communication easier and faster, teaching and learning more practical through computers and other technology based learning tools. However, despite these benefits ICT has equally contributed immensely to environmental problems. Therefore there is the need to use ICT resources efficiently in universities for environmental sustainability so as to save both the university environment and the world at large from the effects of global warming. This paper evaluates the carbon footprints from the use of ICT devices and comes up with a proposed green ICT framework to reduce the carbon footprints in universities. The framework contains techniques and approaches to achieve greenness in the data center, personal computers (PCs) and monitors, and printing in order to make ICT more environmentally friendly, cheaper, safer and ultimately more efficient. Concerned experts in their respective departments at Asia Pacific University of Technology and Innovation (APU) Malaysia evaluated the proposed framework. It was found to be effective for achieving efficiency, reducing energy consumption and carbon emissions.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.3
• /
• pp.273-283
• /
• 2000

In this paper we suggest an optimal design method of Fuzzy-Neural Networks(FNN) model for complex and nonlinear systems. The FNNs use the simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rule. And we use a HCM(Hard C-Means) Clustering Algorithm to find initial parameters of the membership function. The parameters such as parameters of membership functions learning rates and momentum weighted value is proposed to achieve a sound balance between approximation and generalization abilities of the model. According to selection and adjustment of a weighting factor of an aggregate objective function which depends on the number of data and a certain degree of nonlinearity (distribution of I/O data we show that it is available and effective to design and optimal FNN model structure with a mutual balance and dependency between approximation and generalization abilities. This methodology sheds light on the role and impact of different parameters of the model on its performance (especially the mapping and predicting capabilities of the rule based computing). To evaluate the performance of the proposed model we use the time series data for gas furnace the data of sewage treatment process and traffic route choice process.

• The Journal of Korean Association of Computer Education
• /
• v.13 no.2
• /
• pp.59-70
• /
• 2010

This study aims to analyze research trend of Web 2.0 use in education and to suggest directions for further studies. Sixty representative studies were selected and analyzed in terms of a developed framework of Web 2.0 based learning environments and an analysis scheme. This scheme is divided into five dimensions: research targets, research themes, types of Web 2.0 tools, learning theories and research methodology. The findings indicate that a majority of the previous studies aimed to share information and reflect thoughts in collaborative contexts through blogs and wikis at universities. Some implications were discussed for further studies.

• Asia pacific journal of information systems
• /
• v.32 no.4
• /
• pp.770-802
• /
• 2022

With the development of deep learning techniques, text mining is producing breakthrough performance improvements, promising future applications, and practical use cases across many fields. Likewise, even though several attempts have been made in the field of financial information, few cases apply the current technological trends. Recently, companies and government agencies have attempted to conduct research and apply text mining in the field of financial information. First, in this study, we investigate various works using text mining to show what studies have been conducted in the financial sector. Second, to broaden the view of financial application, we provide a description of several text mining techniques that can be used in the field of financial information and summarize various paradigms in which these technologies can be applied. Third, we also provide practical cases for applying the latest text mining techniques in the field of financial information to provide more tangible guidance for those who will use text mining techniques in finance. Lastly, we propose potential future research topics in the field of financial information and present the research methods and utilization plans. This study can motivate researchers studying financial issues to use text mining techniques to gain new insights and improve their work from the rich information hidden in text data.

• Computers and Concrete
• /
• v.30 no.3
• /
• pp.225-236
• /
• 2022

In this article, Multi-Gene Genetic Programming (MGGP) is proposed for the estimation of the compressive strength of concrete. MGGP is known to be a powerful algorithm able to find a relationship between certain input space features and a desired output vector. With respect to most conventional machine learning algorithms, which are often used as "black boxes" that do not provide a mathematical formulation of the output-input relationship, MGGP is able to identify a closed-form formula for the input-output relationship. In the study presented in this article, MGPP was used to predict the compressive strength of plain concrete, concrete with fly ash, and concrete with furnace slag. A formula was extracted for each mixture and the performance and the accuracy of the predictions were compared to the results of Artificial Neural Network (ANN) and Extreme Learning Machine (ELM) algorithms, which are conventional and well-established machine learning techniques. The results of the study showed that MGGP can achieve a desirable performance, as the coefficients of determination for plain concrete, concrete with ash, and concrete with slag from the testing phase were equal to 0.928, 0.906, 0.890, respectively. In addition, it was found that MGGP outperforms ELM in all cases and its' accuracy is slightly less than ANN's accuracy. However, MGGP models are practical and easy-to-use since they extract closed-form formulas that may be implemented and used for the prediction of compressive strength.

• International Journal of Computer Science & Network Security
• /
• v.22 no.9
• /
• pp.316-326
• /
• 2022

The prompt development in the field of video is the outbreak of online services which replaces the television media within a shorter period in gaining popularity. The online videos are encouraged more in use due to the captions displayed along with the scenes for better understandability. Not only entertainment media but other marketing companies and organizations are utilizing videos along with captions for their product promotions. The need for captions is enabled for its usage in many ways for hearing impaired and non-native people. Research is continued in an automatic display of the appropriate messages for the videos uploaded in shows, movies, educational videos, online classes, websites, etc. This paper focuses on two concerns namely the first part dealing with the machine learning method for preprocessing the videos into frames and resizing, the resized frames are classified into multiple actions after feature extraction. For the feature extraction statistical method, GLCM and Hu moments are used. The second part deals with the deep learning method where the CNN architecture is used to acquire the results. Finally both the results are compared to find the best accuracy where CNN proves to give top accuracy of 96.10% in classification.