• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.12
• /
• pp.499-508
• /
• 2022

A lot of research has been carried out on the Hangeul generation model using deep learning, and recently, research is being carried out how to minimize the number of characters input to generate one set of Hangul (Few-Shot Learning). In this paper, we propose a CKFont2 model using only 14 letters by analyzing and improving the CKFont (hereafter CKFont1) model using 28 letters. The CKFont2 model improves the performance of the CKFont1 model as a model that generates all Hangul using only 14 characters including 24 components (14 consonants and 10 vowels), where the CKFont1 model generates all Hangul by extracting 51 Hangul components from 28 characters. It uses the minimum number of characters for currently known models. From the basic consonants/vowels of Hangul, 27 components such as 5 double consonants, 11/11 compound consonants/vowels respectively are learned by deep learning and generated, and the generated 27 components are combined with 24 basic consonants/vowels. All Hangul characters are automatically generated from the combined 51 components. The superiority of the performance was verified by comparative analysis with results of the zi2zi, CKFont1, and MX-Font model. It is an efficient and effective model that has a simple structure and saves time and resources, and can be extended to Chinese, Thai, and Japanese.

• Journal of Science Education
• /
• v.46 no.2
• /
• pp.178-194
• /
• 2022

This study analyzes the trend of domestic research related to science writing education. The subjects of analysis were 152 research papers related to science writing education in Korea from 2004 to 2021. The analysis criteria were set as the research problem, research subject, research method and research application etc. Result of the analysis shows a steady increase until 2014, but decreased afterwards. In the result of the research problems, it was found that most studies were about finding out the effects of scientific writing activities. The research subjects were mostly elementary, middle, and high school students. Qualitative research occupied a large proportion in the results of the research method analysis, and there were many mixed studies that combined quantitative and qualitative research. As for the research application method, the most applied research in regular classes. As a result of analyzing the effect of application, most of the studies were on science concepts, attitudes towards science, thinking skills, and creative problem-solving skills. Writing education such as experimental and observational writing in science classes has been steadily conducted since before the introduction of the 2007 revised curriculum. In particular, the importance of scientific writing as a text-based education is being emphasized from the 2007 revised curriculum to the 2022 revised curriculum overview. Writing is an important learning strategy in science education for students to generate, share, explain, and expand their ideas. Therefore, examining domestic research trends related to science writing education can provide important basic data for setting the future direction of science writing education.

• Journal of the Korean Society for information Management
• /
• v.40 no.2
• /
• pp.183-209
• /
• 2023

The purpose of this study is to evaluate the Dublin Core metadata generated by ChatGPT using book covers, title pages, and colophons from a collection of books. To achieve this, we collected book covers, title pages, and colophons from 90 books and inputted them into ChatGPT to generate Dublin Core metadata. The performance was evaluated in terms of completeness and accuracy. The overall results showed a satisfactory level of completeness at 0.87 and accuracy at 0.71. Among the individual elements, Title, Creator, Publisher, Date, Identifier, Rights, and Language exhibited higher performance. Subject and Description elements showed relatively lower performance in terms of completeness and accuracy, but it confirmed the generation capability known as the inherent strength of ChatGPT. On the other hand, books in the sections of social sciences and technology of DDC showed slightly lower accuracy in the Contributor element. This was attributed to ChatGPT's attribution extraction errors, omissions in the original bibliographic description contents for metadata, and the language composition of the training data used by ChatGPT.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.6D
• /
• pp.711-717
• /
• 2009

This study develops model for analysis of relationship between major node (Interchange in expressway) and land price formation of apartments along with Seoul Belt Expressway by using CHAID analysis. The results show that first, regions(outer side: Gyeongido, inner side: Seoul) on the line of Seoul Belt Expressway are different and a graph generally show llinear relationships between land price and traffic node but it does not; second, CHAID analysis shows two different spatial distribution at the point of 2.6km in the outer side, but three different spatial distribution at the point of 1.4km and 3.8km in the inner side. In other words, traffic access does not necessarily guarantee high housing price since the graphs shows land price related to composite spatial distribution. This implies that residential environments (highway noise and regional discontinuity) and traffic accessibility cause mutual interaction to generate this phenomenon. Therefore, the highway IC landprice model will be beneficial for calculation of land price in New Town which constantly is being built along the highway.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.4
• /
• pp.329-344
• /
• 2023

In this study, we comprehensively analyze the generalization performance of various deep learning-based active sonar target classifiers when applied to small and imbalanced active sonar datasets. To generate the active sonar datasets, we use data from two different oceanic experiments conducted at different times and ocean. Each sample in the active sonar datasets is a time-frequency domain image, which is extracted from audio signal of contact after the detection process. For the comprehensive analysis, we utilize 22 Convolutional Neural Networks (CNN) models. Two datasets are used as train/validation datasets and test datasets, alternatively. To calculate the variance in the output of the target classifiers, the train/validation/test datasets are repeated 10 times. Hyperparameters for training are optimized using Bayesian optimization. The results demonstrate that shallow CNN models show superior robustness and generalization performance compared to most of deep CNN models. The results from this paper can serve as a valuable reference for future research directions in deep learning-based active sonar target classification.

• Journal of Service Research and Studies
• /
• v.11 no.1
• /
• pp.21-30
• /
• 2021

This study examined methodologies for establishing a customer experience-focused innovation strategy, which is essential to transforming existing collection and preservation-oriented museum service into a visitor-oriented service. To this end, a review of studies examining good customer experience and bad customer experience was conducted, focusing on change in museum environments and customer experience from the service science and customer experience management perspectives. Research was conducted to present and prove. Implementing a customer experience-oriented innovation strategy in a museum requires exploring the various approaches that can be used to reach the target state from the present state, focusing on the customer, and selecting the most appropriate transformation plan. This study found that effective approaches are distinguished by the fact that they generate positive emotions in customers and use customer experience data to make important decisions regarding the establishment of practical resources in museums. This innovation model was developed and validated by analyzing how it differed from existing evaluation methods. Finally, a regression analysis using the customer experience measurement procedure, customer experience diagnosis/evaluation, customer experience innovation strategy, and museum performance as variables revealed a significant causal relationship.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.2
• /
• pp.389-397
• /
• 2023

This study examines how macro-economic conditions, fashion momentum, and inter-genre competition affect movie genre's popularity, thus shaping fashion trends in the feature film market in Korea. Using panel data analysis of genre-specific audience sizes with 6 genre cateories and 132 monthly time points, we found that favorable economic conditions generate the fashion trend in the action/crime genre, while the deterioration of the economic conditions leads to the decline of action/crime genre. The finding implies that economic situations influence cultural consumers' psychological states, which in turn shape the fashion trend in certain direction. Furthermore, we found that the action/crime genre has a greater fashion momentum and its competitive power is stronger than other genres, suggesting that this genre has longer fashion cycle even if other genres rise to the top in their popularity. We argue that such enlengthened fahion cycle and competitive stength of the action/crime genre are associated with its breadth of niche width and audience loyalty. Scholarly and practical implications are discussed.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.2B
• /
• pp.173-179
• /
• 2009

Industrial water supplied by water resource project is essential input materials along with labor, capital and land for companies. It is very important to stably secure these input materials in order for the industry to generate additional values. If the supply of industrial water is stopped, it is known damage for the industry is greater than domestic water or agriculture water based on same amount of supply. Like this, the actual value of industrial water has been highly acknowledged from the intuitive perspective, but study on the value and benefits of industrial water has been rarely conducted. Therefore, this study verified the value of industrial water supplied from water resource project, and used marginal production value as a measure to estimate the benefits of industrial water in the analysis of economic efficiency. As a result of empirical analysis using Cobb-Douglas production function and Translog production function, industries' average marginal production value was $5,427KRW/m^3$ and $5,583KRW/m^3$ respectively. The marginal production value for eleven industries were estimated by using same method. The marginal production value by industries presented by this study will be used as important data to calculate benefits of industrial water in the future. Moreover, the result of this study will provide reasonable criteria for decision making on the allocation of water in emergency situation, and problem of resource supply from water resource project.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.3B
• /
• pp.289-301
• /
• 2009

In this work, wave run-up heights and resultant wave forces on a vertical revetment due to tsunami (solitary wave) are investigated numerically using a numerical wave tank model called CADMAS-SURF (CDIT, 2001. Research and Development of Numerical Wave Channel (CADMAS-SURF). CDIT library, No. 12, Japan.), which is based on a 2-D Navier-Stokes solver, coupled to a volume of fluid (VOF) method. The third order approximate solution (Fenton, 1972. A ninth-order solution for the solitary wave. J. of Fluid Mech., Vol. 53, No.2, pp.257-271) is used to generate solitary waves and implemented in original CADMAS-SURF code. Numerical results of the wave profiles and forces are in good agreements with available experimental data. Using the numerical results, the regression curves determined from the least-square analysis are proposed, which can be used to determine the maximum wave run-up height and force on a vertical revetment due to tsunami. In addition, the capability of CADMAS-SURF is demonstrated for tsunami wave forces acting on an onshore structure using various configuration computations including the variations of the crown heights of the vertical wall and the position of the onshore structure. Based on the numerical results such as water level, velocity field and wave force, the direct effects of tsunami on an onshore structure are discussed.

• Korean Journal of Legislative Studies
• /
• v.27 no.1
• /
• pp.5-40
• /
• 2021

This study attempts to analyze gender gaps in voter turnout for three different types of elections held since 2017 at the aggregate level using the Central Election Management Commission's turnout data, paying attention to the importance of women's voting. The findings are as follows. First, modern gender gaps in voter turnout at the aggregate level are confirmed in most regions regardless of election types. Second, the gender gap in turnout varies with age. The gender difference is verified in the "widowhood effect," where turnout decreases in the oldest-old. In the new voter group, modern gender gaps appear in most regions. The reversed gender difference in turnout in the late 20s, which reflects the Korean society's characteristics, is confirmed in all elections. Third, it is unclear whether the reverse gender gap in turnout becomes more pronounced in urbanized districts. As urbanization progressed, modern gender differences in voter turnout across age groups are observed at the population-based size level. Paradoxically, the modern gender gap tends to be weak and turns into the traditional gender gap in younger age groups (in the late thirties) in Gangnam-gu and Seocho-gu, the most modernized districts in Seoul. These results show that the modern gender gap in turnout is now a common phenomenon and continues to be strengthened by newly recruited voters in Korea. Thus we should pay more attention to female voters' political behavior and a new approach beyond the developmental theory to understand the causal mechanism to generate the modern gender gap in voter turnout.