• Journal of Environmental Health Sciences
• /
• v.47 no.6
• /
• pp.521-529
• /
• 2021

Background: The coronavirus disease (COVID-19) has caused changes in human activity, and these changes may possibly increase or decrease exposure to fine dust (PM_2.5). Therefore, it is necessary to evaluate the exposure to PM_2.5 in relation to the outbreak of COVID-19. Objectives: The purpose of this study was to compare and evaluate the exposure to PM_2.5 concentrations by the variation of dynamic populations before and after the outbreak of COVID-19. Methods: This study evaluated exposure to PM_2.5 concentrations by changes in the dynamic population distribution in Guro-gu, Seoul, before and after the outbreak of COVID-19 between Jan and Feb, 2020. Gurogu was divided into 2,204 scale standard grids of 100 m×100 m. Hourly PM_2.5 concentrations were modeled by the inverse distance weight method using 24 sensor-based air monitoring instruments. Hourly dynamic population distribution was evaluated according to gender and age using mobile phone network data and time-activity patterns. Results: Compared to before, the population exposure to PM_2.5 decreased after the outbreak of COVID-19. The concentration of PM_2.5 after the outbreak of COVID-19 decreased by about 41% on average. The variation of dynamic population before and after the outbreak of COVID-19 decreased by about 18% on average. Conclusions: Comparing before and after the outbreak of COVID-19, the population exposures to PM_2.5 decreased by about 40%. This can be explained to suggest that changes in people's activity patterns due to the outbreak of COVID-19 resulted in a decrease in exposure to PM_2.5.

• Journal of Environmental Health Sciences
• /
• v.48 no.2
• /
• pp.66-74
• /
• 2022

Background: The health effects of particulate matter (PM_2.5) bonded with various harmful chemicals differ based on their composition, so investigating and managing their concentrations and composition is vital for long-term management. As industrial complexes emit considerable quantities of pollutants, higher PM_2.5 concentrations and chemical component effects are expected than in other places. Objectives: We investigated the concentration distribution ratios of PM_2.5 chemical components to provide basic data to inform future major emissions control and PM_2.5 reduction measures in industrial complexes. Methods: We monitored five sites near the Ansan and Siheung industrial complexes from August 2020 to July 2021. Samples were collected and analyzed twice per week in spring/winter and once per week in summer/autumn according to the National Institute of Environmental Research in the Ministry of Environments' Air Pollution Monitoring Network Installation and Operation Guidelines. We investigated and compared composition ratios of 29 ions, carbon, and elemental components in PM_2.5. Results: The analysis of PM_2.5 components at the five sites revealed that ion components accounted for the greatest total mass at approximately 50% while carbon components and elemental components contributed 23~28% and 8~10%, respectively. Among the ionic components, NO₃^- occupies the greatest proportion. OC occupies the greatest proportion of the carbon components and sulphur occupies the greatest proportion of elemental components. Conclusions: This study investigated the concentration distribution ratios of PM_2.5 chemical components in industrial complexes. We believe these results provide basic chemical component concentration ratio data for establishing future air management policies and plans for the Ansan and Siheung industrial complexes.

• Journal of Internet of Things and Convergence
• /
• v.8 no.2
• /
• pp.41-48
• /
• 2022

This study tried to empirically explore which issues related to the social service agency for public(as below SSA), that is, social perceptions were formed, by using mess media related to the SSA. This study is meaningful in that it identifies the overall social perception and trend of SSA through public opinion. In order to extract media trend data, the search used the big data analysis system, Textom, to collect data from the representative portals Naver News and Daum News. The collected texts were 1,299 in 2020 and 1,410 in 2021, for a total of 2,709. As a result of the analysis, first, the most derived words in relation to the frequency of text appearance were 'SSA', 'establishment', and 'operation'. Second, as a result of the N-gram analysis, the pairs of words directly related to the SSA 'SSA and public', 'SSA and opening', 'SSA and launch', and 'SSA and Department Director', 'SSA and Staff', 'SSA and Caregiver' etc. Third, in the results of TF-IDF analysis and word network analysis, similar to the word occurrence frequency and N-gram results, 'establishment', 'operation', 'public', 'launch', 'provided', 'opened', ' 'Holding' and 'Care' were derived. Based on the above analysis results, it was suggested to strengthen the emergency care support group, to commercialize it in detail, and to stabilize jobs.

• Economic and Environmental Geology
• /
• v.54 no.6
• /
• pp.615-627
• /
• 2021

The Baekildo fault, a dextral strike-slip fault developed in Baekil Island, Goheung-gun, controls the distribution of tuffaceous sandstone and lapilli tuff and shows a complex fracture system around it. In this study, we examined the spatial variation in the geometry and connectivity of the fracture system by using circular sampling and topological analysis based on a detailed fracture trace map. As a result, both intensity and connectivity of the fracture system are higher in tuffaceous sandstone than in lapilli tuff. Furthermore, the degree of the orientation dispersion, intensity, and average length of fracture sets vary depending on the along-strike variation in structural position in the tuffaceous sandstone. Notably, curved fractures abutting the fault at a high angle occur at a fault bend. Based on the detailed observation and analyses of the fracture system, we conclude as follows: (1) the high intensity of the fracture system in the tuffaceous sandstone is caused by the higher content of brittle minerals such as quartz and feldspar. (2) the connectivity of the fracture system gets higher with the increase in the diversity and average length of the fracture sets. Finally, (3) the fault bend with geometric irregularity is interpreted to concentrate and disturb the local stress leading to the curved fractures abutting the fault at a high angle. This contribution will provide important insight into various geologic and structural factors that control the development of fracture systems around faults.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.4
• /
• pp.169-178
• /
• 2022

Morphemes are most primitive units in a language that lose their original meaning when segmented into smaller parts. In Korean, a sentence is a sequence of eojeols (words) separated by spaces. Each eojeol comprises one or more morphemes. Korean morphological analysis (KMA) is to divide eojeols in a given Korean sentence into morpheme units. It also includes assigning appropriate part-of-speech(POS) tags to the resulting morphemes. KMA is one of the most important tasks in Korean natural language processing (NLP). Improving the performance of KMA is closely related to increasing performance of Korean NLP tasks. Recent research on KMA has begun to adopt the approach of machine translation (MT) models. MT is to convert a sequence (sentence) of units of one domain into a sequence (sentence) of units of another domain. Neural machine translation (NMT) stands for the approaches of MT that exploit neural network models. From a perspective of MT, KMA is to transform an input sequence of units belonging to the eojeol domain into a sequence of units in the morpheme domain. In this paper, we propose a deep learning model for KMA. The backbone of our model is based on the BERT-fused model which was shown to achieve high performance on NMT. The BERT-fused model utilizes Transformer, a representative model employed by NMT, and BERT which is a language representation model that has enabled a significant advance in NLP. The experimental results show that our model achieves 98.24 F1-Score.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.4
• /
• pp.303-311
• /
• 2022

Recently the telemedicine platform firms which have been temporarily permitted since COVID-19 outbreak have increasingly provided online prescription drugs delivery, causing concerns among incumbent providers of medicine, some of whom began to take aggressive actions again them. In this study, using game theoretic market entry - deterrence model, we show that although the incumbent medicine provider can effectively deter entry by the telemedicine platform firms by its preemptive action, accommodation could be a optimal action when telemedicine platform firms already have penetrated the market with their being permitted to do business due to the COVID-19. However, for the incumbent to cooperate for the successful change in the retail market for medicines, policies like placing a ceiling on the maximum number of taking prescriptions by the pharmacists a day in the telemedince platform network, providing favorable exposure of community pharmacists on the telemedicine platform user interface, and allowing community pharmacies to participate as shareholders of the telemedicine platform firms in its initial public opening of capital, are suggested.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.263-278
• /
• 2022

With the recent advent of IoT technology, automatic pet feeders are being distributed so that owners can feed their companion animals while they are out. However, due to behaviors of pets, the method of measuring weight, which is important in automatic feeding, can be easily damaged and broken when using the scale. The 3D camera method has disadvantages due to its cost, and the 2D camera method has relatively poor accuracy when compared to 3D camera method. Hence, the purpose of this study is to propose a deep learning approach that can accurately estimate weight while simply using a 2D camera. For this, various convolutional neural networks were used, and among them, the ResNet101-based model showed the best performance: an average absolute error of 3.06 grams and an average absolute ratio error of 3.40%, which could be used commercially in terms of technical and financial viability. The result of this study can be useful for the practitioners to predict the weight of a standardized object such as feed only through an easy 2D image.

• Tunnel and Underground Space
• /
• v.31 no.6
• /
• pp.593-609
• /
• 2021

We proposed a numerical method for the thermo-mechanical behavior of rock fracture using a grain-based distinct element model (GBDEM) and simulated thermally induced fracture slip. The present study is the benchmark simulation performed as part of DECOVALEX-2023 Task G, which aims to develop a numerical method to estimate the coupled thermo-hydro-mechanical processes within the crystalline rock fracture network. We represented the rock sample as an assembly of Voronoi grains and calculated the interaction of the grains (blocks) and their interfaces (contacts) using a distinct element code, 3DEC. Based on an equivalent continuum approach, the micro-parameters of grains and contacts were determined to reproduce rock as an elastic material. Then, the behavior of the fracture embedded in the rock was characterized by the contacts with Coulomb shear strength and tensile strength. In the benchmark simulation, we quantitatively examined the effects of the boundary stress and thermal stress due to heat conduction on fracture behavior, focusing on the mechanism of thermally induced fracture slip. The simulation results showed that the developed numerical model reasonably reproduced the thermal expansion and thermal stress increment, the fracture stress and displacement and the effect of boundary condition. We expect the numerical model to be enhanced by continuing collaboration and interaction with other research teams of DECOVALEX-2023 Task G and validated in further study experiments.

• Korean Journal of Cognitive Science
• /
• v.33 no.1
• /
• pp.23-50
• /
• 2022

Source memory is a cognitive process that combines the representation of the origin of the episodic experience with an item. By studying this daily process, researchers have made fundamental discoveries that make up the foundation of brain and behavior research, such as executive function and binding. In this paper, we review and conduct a bibliometric analysis on source memory papers published from 1989 to 2020. This review is based on keyword co-occurrence networks and author citation networks, providing an in-depth overview of the development of source memory research and future directions. This bibliometric analysis discovers a change in the research trends: while research prior to 2010 focused on individuality of source memory as a cognitive function, more recent papers focus more on the implication of source memory as it pertains to connectivity between disparate brain regions and to social neuroscience. Keyword network analysis shows that aging and executive function are continued topics of interest, although frameworks in which they are viewed have shifted to include developmental psychology and meta memory. The use of theories and models provided by source memory research seem essential for the future development of cognitive enhancement tools within and outside of the field of Psychology.

• Journal of Broadcast Engineering
• /
• v.27 no.2
• /
• pp.185-197
• /
• 2022

Recently, deep learning networks with high performance for object recognition are emerging. In the case of object recognition using deep learning, it is important to build a training data set to improve performance. To build a data set, we need to collect and label the images. This process requires a lot of time and manpower. For this reason, open data sets are used. However, there are objects that do not have large open data sets. One of them is data required for license plate detection and recognition. Therefore, in this paper, we propose an artificial license plate generator system that can create large data sets by minimizing images. In addition, the detection rate according to the artificial license plate arrangement structure was analyzed. As a result of the analysis, the best layout structure was FVC_III and B, and the most suitable network was D2Det. Although the artificial data set performance was 2-3% lower than that of the actual data set, the time to build the artificial data was about 11 times faster than the time to build the actual data set, proving that it is a time-efficient data set building system.