• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.3
• /
• pp.115-122
• /
• 2019

For Korean phoneme recognition, Hidden Markov-Gaussian Mixture model(HMM-GMM) or hybrid models which combine artificial neural network with HMM have been mainly used. However, current approach has limitations in that such models require force-aligned corpus training data that is manually annotated by experts. Recently, researchers used neural network based phoneme recognition model which combines recurrent neural network(RNN)-based structure with connectionist temporal classification(CTC) algorithm to overcome the problem of obtaining manually annotated training data. Yet, in terms of implementation, these RNN-based models have another difficulty in that the amount of data gets larger as the structure gets more sophisticated. This problem of large data size is particularly problematic in the Korean language, which lacks refined corpora. In this study, we introduce CTC algorithm that does not require force-alignment to create a Korean phoneme recognition model. Specifically, the phoneme recognition model is based on convolutional neural network(CNN) which requires relatively small amount of data and can be trained faster when compared to RNN based models. We present the results from two different experiments and a resulting best performing phoneme recognition model which distinguishes 49 Korean phonemes. The best performing phoneme recognition model combines CNN with 3hop Bidirectional LSTM with the final Phoneme Error Rate(PER) at 3.26. The PER is a considerable improvement compared to existing Korean phoneme recognition models that report PER ranging from 10 to 12.

• The Korean Journal of Applied Statistics
• /
• v.34 no.1
• /
• pp.115-126
• /
• 2021

Central banks communicate with the market through a statement on the direction of monetary policy while implementing monetary policy. The rapid contraction of the global economy due to the recent Covid-19 pandemic could be compared to the crisis situation during the 2008 global financial crisis. In this paper, we analyzed the text data from the monetary policy statements of the Bank of Korea and Fed reflecting monetary policy directions focusing on how they were affected in the face of a global crisis. For analysis, we collected the text data of the two countries' monetary policy direction reports published from October 1999 to September 2020. We examined the semantic features using word cloud and word embedding, and analyzed the trend of the similarity between two countries' documents through a piecewise regression tree model. The visualization result shows that both the Bank of Korea and the US Fed have published the statements with refined words of clear meaning for transparent and effective communication with the market. The analysis of the dissimilarity trend of documents in both countries also shows that there exists a sense of synchronization between them as the rapid changes in the global economic environment affect monetary policy.

• Journal of Digital Convergence
• /
• v.19 no.3
• /
• pp.319-329
• /
• 2021

This study aims to investigate Korean fashion-related research trends and knowledge structures on digital transformation through information-based approaches. Accordingly, we first identified the current status of the relevant research in Korean academic literature by year and journal; subsequently, we derived key research topics through network analysis, and then analyzed major research trends and knowledge structures by time. From 2010 to 2020, we collected 159 studies published on Korean academic platforms, cleansed data through Python 3.7, and measured centrality and network implementation through NodeXL 1.0.1. The results are as follows: first, related research has been actively conducted since 2016, mainly concentrated in clothing and art areas. Second, the online platform, AR/VR, appeared as the most frequently mentioned topic, and consumer psychological analysis, marketing strategy suggestion, and case analysis were used as the main research methods. Through clustering, major research contents for each sub-major of clothing were derived. Third, major subject by period was considered, which has, over time, changed from consumer-centered research to strategy suggestion, and design development research of platforms or services. This study contributes to enhancing insight into the fashion field on digital transformation, and can be used as a basic research to design research on related topics.

• Journal of the Korea Convergence Society
• /
• v.12 no.11
• /
• pp.45-51
• /
• 2021

With the development of deep learning, semantic segmentation methods are being studied in various fields. There is a problem that segmenation accuracy drops in fields that require accuracy such as medical image analysis. In this paper, we improved PSPNet, which is a deep learning based segmentation method to minimized the loss of features during semantic segmentation. Conventional deep learning based segmentation methods result in lower resolution and loss of object features during feature extraction and compression. Due to these losses, the edge and the internal information of the object are lost, and there is a problem that the accuracy at the time of object segmentation is lowered. To solve these problems, we improved PSPNet, which is a semantic segmentation model. The multi-scale attention proposed to the conventional PSPNet was added to prevent feature loss of objects. The feature purification process was performed by applying the attention method to the conventional PPM module. By suppressing unnecessary feature information, eadg and texture information was improved. The proposed method trained on the Cityscapes dataset and use the segmentation index MIoU for quantitative evaluation. As a result of the experiment, the segmentation accuracy was improved by about 1.5% compared to the conventional PSPNet.

• Journal of Applied Biological Chemistry
• /
• v.65 no.3
• /
• pp.215-219
• /
• 2022

The flowers of Cosmos bipinnatus were extracted with solvent made with methanol:water (4:1) and the concentrates were partitioned into ethyl acetate (EtOAc), n-butanol (n-BuOH), and water (H₂O) fractions. The octadecyl silica gel (ODS) and silica gel (SiO₂) column chromatographies were repeated for the EtOAc fraction to isolated of two phenolic compounds. The chemical structure of the isolated compounds were identified as benzyl O-β-ᴅ-glucopyranoside (1), and 2-phenylethyl O-β-ᴅ-glucopyranoside (2) through spectroscopic datas such as nuclear magnetic resornance, infrarad spectroscopy, and mass spectroscopy. These two compounds were first isolated from C. bipinnatus flowers through this study. To evaluate the anti-atopic activity of the two isolated compounds using a HaCaT cell line induced by ultraviolet light, several experiments were conducted and neither both compounds showed toxicity in the concentration range of 1 to 1,000 ㎍/mL. In the results of anti-atopic activity through Thymus and activation regualted chemokine (TARC) assay, both compounds showed dose-dependent TARC inhibitory activity. In particular, compound 1 showed significant activity even in a low concentration range of 10 ㎍/mL, and in different concentration ranges. Also compound 1 showed higher inhibitory activity than other compound, confirming that the anti-atopic activity was the most excellent. Based on these results, it is considered that it can be used as a functional cosmetic material.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.12
• /
• pp.59-68
• /
• 2022

This paper presents a method for 1:1 verification by comparing the similarity between the given real product image and the drawing image. The proposed method combines two existing CNN-based deep learning models to construct a Siamese Network. After extracting the feature vector of the image through the FC (Fully Connected) Layer of each network and comparing the similarity, if the real product image and the drawing image (front view, left and right side view, top view, etc) are the same product, the similarity is set to 1 for learning and, if it is a different product, the similarity is set to 0. The test (inference) model is a deep learning model that queries the real product image and the drawing image in pairs to determine whether the pair is the same product or not. In the proposed model, through a comparison of the similarity between the real product image and the drawing image, if the similarity is greater than or equal to a threshold value (Threshold: 0.5), it is determined that the product is the same, and if it is less than or equal to, it is determined that the product is a different product. The proposed model showed an accuracy of about 71.8% for a query to a product (positive: positive) with the same drawing as the real product, and an accuracy of about 83.1% for a query to a different product (positive: negative). In the future, we plan to conduct a study to improve the matching accuracy between the real product image and the drawing image by combining the parameter optimization study with the proposed model and adding processes such as data purification.

• Clean Technology
• /
• v.28 no.3
• /
• pp.203-209
• /
• 2022

An environmental evaluation was conducted by employing LCA methodology for a mechanical seal manufacturing process that uses recycled silicon recovered from end-of-cycle PV modules. The recycled silicon was purified and reacted with carbon to synthesize β-SiC particles. Then the particles underwent compression molding, calcination and heat treatment to produce a product. Field data were collected and the potential environmental impacts of each stage were calculated using the LCI DB of the Ministry of Environment. The assessment was based on 6 categories, which were abiotic resource depletion, acidification, eutrophication, global warming, ozone depletion and photochemical oxidant creation. The environmental impacts by category were 45 kg CO₂ for global warming and 2.23 kg C₂H₄ for photochemical oxide creation, and the overall environmental impact by photochemical oxide creation, resource depletion and global warming had a high contribution of 98.7% based on weighted analysis. The wet process of fine grinding and mixing the raw silicon and carbon, and SiC granulation were major factors that caused the environmental impacts. These impacts need to be reduced by converting to a dry process and using a system to recover and reuse the solvent emitted to the atmosphere. It was analyzed that the environmental impacts of resource depletion and global warming decreased by 53.9% and 60.7%, respectively, by recycling silicon from end-of-cycle PV modules. Weighted analysis showed that the overall environmental impact decreased by 27%, and the LCA analysis confirmed that recycling waste modules could be a major means of resource saving and realizing carbon neutrality.

• Journal of Applied Biological Chemistry
• /
• v.65 no.4
• /
• pp.247-252
• /
• 2022

The arial parts of Artemisia argyi were extracted with methanol : water (70:30), and the concentrates was partitioned into EtOAc (ethyl acetate), n-BuOH (normal butanol), and H₂O (water) fractions. The repeated silica gel and ODS (octadecyl silica gel) column chromatographies for EtOAc and n-BuOH fractions led to isolation of four flavonoids without any ambiguity based on intensive interpretation of several spectroscopic data including nuclear magnetic resonance, and mass spectrometry. The chemical structure of the isolated compounds revealed to (2S)-naringenin (1), 3-methylkaempferol (2), 3,3'-dimethylquercetin (3), and 3,3',4'-trimethylquercetin (4). These four compounds were first isolated from A. argyi through this study. In this study, four compounds isolated from A. argyi showed an increase in glutathione mean and a decrease in glutathione heterogeneity so that the compounds uniformly raised the intracellular glutathione (GSH) level. Based on these results, it is considered that it can be used as a functional pharmacological material.

• Journal of the Korean Society for Library and Information Science
• /
• v.57 no.4
• /
• pp.333-351
• /
• 2023

This study investigates user behavior in library spaces through the lens of AI camera analytics. By leveraging the face recognition and tracking capabilities of AI cameras, we accurately identified the gender and age of visitors and meticulously collected video data to track their movements. Our findings revealed that female users slightly outnumbered male users and the dominant age group was individuals in their 30s. User visits peaked between Tuesday to Friday, with the highest footfall recorded between 14:00 and 15:00 pm, while visits decreased over the weekend. Most visitors utilized one or two specific spaces, frequently consulting the information desk for inquiries, checking out/returning items, or using the rest area for relaxation. The library stacks were used approximately twice as much as they were avoided. The most frequented subject areas were Philosophy(100), Religion(200), Social Sciences(300), Science(400), Technology(500), and Literature(800), with Literature(800) and Religion(200) displaying the most intersections with other areas. By categorizing users into five clusters based on space utilization patterns, we discerned varying objectives and subject interests, providing insights for future library service enhancements. Moreover, the study underscores the need to address the associated costs and privacy concerns when considering the broader application of AI camera analytics in library settings.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.18 no.2
• /
• pp.21-34
• /
• 2023

The purpose of this study is to examine the extent of Sustainable Development Goals (SDGs)-related research in the field of entrepreneurship globally since the adoption of the SDGs at the UN General Assembly, and to compare international and domestic research trends in order to determine the direction of SDGs-related research in entrepreneurship in Korea. Utilizing three databases-Web of Science (WoS), KCI, and DBpia- SDGs-related studies in entrepreneurship were extracted by employing specific search terms. After data purification, a total of 356 studies abroad and 4 studies in Korea were used for analysis. After data purification, a total of 356 international studies and 4 Korean studies were analyzed. Due to the limited number of domestic studies, the research trends were examined by conducting frequency analysis and keyword network analysis on international studies alone. Frequency analysis revealed that SDGs research in entrepreneurship primarily focused on sustainability-related terms and was conducted in conjunction with business models, innovation, entrepreneurship education, and strategies. Furthermore, yearly frequency analysis demonstrated an expansion of topics to encompass research on entrepreneurship and SDGs policies, the roles and capabilities of female entrepreneurs in SDGs implementation, energy start-ups and SDGs, directions for implementing SDGs in business schools and SDGs education, indicators for SDGs implementation and evaluation, and technologies for sustainability. The keyword network analysis identified central topics such as business, sustainability, SDGs, innovation, entrepreneurship, business models, and education, with research areas extending to entrepreneurship ecosystems, change and strategy, ethics, and climate. This study holds significance in establishing a foundation for SDGs research in entrepreneurship, which is currently an underexplored area in Korea, by presenting emerging research trends related to SDGs in entrepreneurship.