• Journal of Service Research and Studies
• /
• v.13 no.1
• /
• pp.127-143
• /
• 2023

Currently, the domestic traditional market has not escaped the swamp of stagnation that began in the early 2000s despite various projects promoted by many related players such as the central government and local governments. In order to overcome the crisis faced by the traditional market, various R&Ds have recently been conducted on how to build a smart traditional market that combines information and communication technologies such as big data analysis, artificial intelligence, and the Internet of Things. This study analyzes various previous studies, users of traditional markets, and application cases of ICT technology in foreign traditional markets since 2012 and proposes a model to build a smart traditional market using ICT technology based on the analysis. The model proposed in this study includes building a traditional market metaverse that can interact with visitors, certifying visits to traditional markets through digital signage with NFC technology, improving accuracy of fire detection functions using IoT and AI technology, developing smartphone apps for market launch information and event notification, and an e-commerce system. If a smart traditional market platform is implemented and operated based on the smart traditional market platform model presented in this study, it will not only draw interest in the traditional market to MZ generation and foreigners, but also contribute to revitalizing the traditional market in the future.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.2
• /
• pp.104-108
• /
• 2023

Because of recent expansion of the electric car market. it is highly growing that should be supplemented its performance and safely issue. The EMI problem due to the interlocking of electrical components that causes various safety problems such as fire in electric vehicles is emerging every time. We strive to achieve optimal charging efficiency by combining various technologies and reduce radioactive noise among the EMI noise of a weirless charging control module, one of the important parts of an electric vehicle was designed and tested. In order to analyze the EMI problems occurring in the wireless charging control module, the optimized wireless charging control module by applying the optimization design technology by learning the accumulated test data for critical factors by utilizing the Python-based script function in the Ansys simulation tool. It showed an EMI noise improvement effect of 25 dBu V/m compared to the charge control module. These results not only contribute to the development of a more stable and reliable weirless charging function in electric vehicles, but also increase the usability and efficiency of electric vehicles. This allows electric vehicles to be more usable and efficient, making them an environmentally friendly alternative.

• Journal of Industrial Convergence
• /
• v.21 no.3
• /
• pp.195-202
• /
• 2023

This study analyzes the results of a survey based on classes conducted in the field to understand the educational needs of learners, and reflects the elements necessary for SW education. In this study, various experimental elements according to learning motivation and learning achievement were constructed and designed through previous studies. As a survey applied to this study, experimental elements in three categories: Faculty Competences(FC), Learner Competences(LC), and Educational Conditions(EC) were analyzed by primary area and secondary major, respectively. As a result of analyzing CT-based SW education by area, the development of educational materials, understanding of lectures, and teaching methods showed high satisfaction, while communication with students, difficulty of lectures, and the number of students were relatively low. The results of the analysis by major were found to be more difficult and less interesting in the humanities than in the engineering field. In this study, Based on these statistical results proposes the need for non-major SW education to improve into an interesting curriculum for effective liberal arts education in the future in terms of enhancing learners' problem-solving skills.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.125-132
• /
• 2022

Sentence compression is a natural language processing task that generates concise sentences that preserves the important meaning of the original sentence. For grammatically appropriate sentence compression, early studies utilized human-defined linguistic rules. Furthermore, while the sequence-to-sequence models perform well on various natural language processing tasks, such as machine translation, there have been studies that utilize it for sentence compression. However, for the linguistic rule-based studies, all rules have to be defined by human, and for the sequence-to-sequence model based studies require a large amount of parallel data for model training. In order to address these challenges, Deleter, a sentence compression model that leverages a pre-trained language model BERT, is proposed. Because the Deleter utilizes perplexity based score computed over BERT to compress sentences, any linguistic rules and parallel dataset is not required for sentence compression. However, because Deleter compresses sentences only considering perplexity, it does not compress sentences by reflecting the linguistic information of the words in the sentences. Furthermore, since the dataset used for pre-learning BERT are far from compressed sentences, there is a problem that this can lad to incorrect sentence compression. In order to address these problems, this paper proposes a method to quantify the importance of linguistic information and reflect it in perplexity-based sentence scoring. Furthermore, by fine-tuning BERT with a corpus of news articles that often contain proper nouns and often omit the unnecessary modifiers, we allow BERT to measure the perplexity appropriate for sentence compression. The evaluations on the English and Korean dataset confirm that the sentence compression performance of sentence-scoring based models can be improved by utilizing the proposed method.

• 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.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.145-151
• /
• 2023

3D-CNN is one of the deep learning techniques for learning time series data. Such three-dimensional learning can generate many parameters, so that high-performance machine learning is required or can have a large impact on the learning rate. When learning dynamic hand-gestures in spatiotemporal domain, it is necessary for the improvement of the efficiency of dynamic hand-gesture learning with 3D-CNN to find the optimal conditions of input video data by analyzing the learning accuracy according to the spatiotemporal change of input video data without structural change of the 3D-CNN model. First, the time ratio between dynamic hand-gesture actions is adjusted by setting the learning interval of image frames in the dynamic hand-gesture video data. Second, through 2D cross-correlation analysis between classes, similarity between image frames of input video data is measured and normalized to obtain an average value between frames and analyze learning accuracy. Based on this analysis, this work proposed two methods to effectively select input video data for 3D-CNN deep learning of dynamic hand-gestures. Experimental results showed that the learning interval of image data frames and the similarity of image frames between classes can affect the accuracy of the learning model.

• Journal of Internet of Things and Convergence
• /
• v.9 no.3
• /
• pp.93-98
• /
• 2023

In this paper, an artificial neural network based on simulated annealing was constructed. The mapping relationship between the parameters of micro-scaled flow channels electrochemical machining and the channel shape was established by training the samples. The depth and width of micro-scaled flow channels electrochemical machining on stainless steel surface were predicted, and the flow channels experiment was carried out with pulse power supply in NaNO3 solution to verify the established network model. The results show that the depth and width of the channel predicted by the simulated annealing artificial neural network with "4-7-2" structure are very close to the experimental values, and the error is less than 5.3%. The predicted and experimental data show that the etching degree in the process of channels electrochemical machining is closely related to voltage and current density. When the voltage is less than 5V, a "small island" is formed in the channel; When the voltage is greater than 40V, the lateral etching of the channel is relatively large, and the "dam" between the channels disappears. When the voltage is 25V, the machining morphology of the channel is the best.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.12
• /
• pp.489-498
• /
• 2022

Recently, the textile industry, which is labor-intensive, is attempting to reduce process costs and optimize quality through artificial intelligence. However, the fiber spinning process has a high cost for data collection and lacks a systematic data collection and processing system, so the amount of accumulated data is small. In addition, data imbalance occurs by preferentially collecting only data with changes in specific variables according to the purpose of fiber spinning, and there is an error even between samples collected under the same fiber spinning conditions due to difference in the measurement environment of physical properties. If these data characteristics are not taken into account and used for AI models, problems such as overfitting and performance degradation may occur. Therefore, in this paper, we propose an outlier handling technique and data augmentation technique considering the characteristics of the spinning process data. And, by comparing it with the existing outlier handling technique and data augmentation technique, it is shown that the proposed technique is more suitable for spinning process data. In addition, by comparing the original data and the data processed with the proposed method to various models, it is shown that the performance of the tensile tenacity and elongation prediction model is improved in the models using the proposed methods compared to the models not using the proposed methods.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.104-123
• /
• 2022

Style transfer based on neural network provides very high quality results by reflecting the high level structural characteristics of images, and thereby has recently attracted great attention. This paper deals with the problem of resolution limitation due to GPU memory in performing such neural style transfer. We can expect that the gradient operation for style transfer based on partial image, with the aid of the fixed size of receptive field, can produce the same result as the gradient operation using the entire image. Based on this idea, each component of the style transfer loss function is analyzed in this paper to obtain the necessary conditions for partitioning and padding, and to identify, among the information required for gradient calculation, the one that depends on the entire input. By structuring such information for using it as auxiliary constant input for partition-based gradient calculation, this paper develops a recursive algorithm for super high-resolution image style transfer. Since the proposed method performs style transfer by partitioning input image into the size that a GPU can handle, it can perform style transfer without the limit of the input image resolution accompanied by the GPU memory size. With the aid of such super high-resolution support, the proposed method can provide a unique style characteristics of detailed area which can only be appreciated in super high-resolution style transfer.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.6
• /
• pp.701-708
• /
• 2017

The construction cost is estimated based on the drawings at the design stage and constructor will find efficient construction methods for budgeting and budgeting appropriate to the budget. Accurate quantity estimation and budgeting are critical to determining whether the project is profitable or not. However, since this process is mostly performed depending on manpower or 2D drawings, errors are likely to occur and The BIM(Build Information Modeling) program, which can be automated, is very expensive and difficult to apply in the field. In this study, 3D architectural modeling was performed using SketchUp which is a 3D modeling software and suggest a methodology for Quantity Estimation. As a result, 3D modeling was performed effectively using 2D drawings of buildings. Based on the modeling results, it was possible to calculate the difference of the quantity estimation by 2D drawing and 3D modeling. The research suggests that the 3D modeling using the SketchUp and the calculation of the quantity can prevent the error of the conventional 2D calculation method. If the applicability of the research method is verified through continuous research, it will contribute to increase the efficiency of architectural modeling and quantity Estimation work.