• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.251-273
• /
• 2022

Recently, research on applying text analysis to deep learning has steadily continued. In particular, researches have been actively conducted to understand the meaning of words and perform tasks such as summarization and sentiment classification through a pre-trained language model that learns large datasets. However, existing pre-trained language models show limitations in that they do not understand specific domains well. Therefore, in recent years, the flow of research has shifted toward creating a language model specialized for a particular domain. Domain-specific pre-trained language models allow the model to understand the knowledge of a particular domain better and reveal performance improvements on various tasks in the field. However, domain-specific further pre-training is expensive to acquire corpus data of the target domain. Furthermore, many cases have reported that performance improvement after further pre-training is insignificant in some domains. As such, it is difficult to decide to develop a domain-specific pre-trained language model, while it is not clear whether the performance will be improved dramatically. In this paper, we present a way to proactively check the expected performance improvement by further pre-training in a domain before actually performing further pre-training. Specifically, after selecting three domains, we measured the increase in classification accuracy through further pre-training in each domain. We also developed and presented new indicators to estimate the specificity of the domain based on the normalized frequency of the keywords used in each domain. Finally, we conducted classification using a pre-trained language model and a domain-specific pre-trained language model of three domains. As a result, we confirmed that the higher the domain specificity index, the higher the performance improvement through further pre-training.

• Journal of Bio-Environment Control
• /
• v.31 no.4
• /
• pp.485-496
• /
• 2022

Modern agriculture is being transformed into smart agriculture to maximize production efficiency along with changes in the 4th industrial revolution. However, rural areas in Korea are facing challenges of aging, low fertility, and population outflow, making it difficult to transition to smart agriculture. Among ICT technologies, simulation allows users to observe or experience the results of their choices through imitation or reproduction of reality. The combination of the three-dimension (3D) model and the greenhouse simulator enable a 3D experience by virtual greenhouse for fruits and vegetable cultivation. At the same time, it is possible to visualize the greenhouse under various cultivation or climate conditions. The objective of this study is to apply the greenhouse climate management model for simulation development that can visually see the state of the greenhouse environment under various micrometeorological properties. The numerical solution with the mathematical model provided a dynamic change in the greenhouse environment for a particular greenhouse design. Light intensity, crop transpiration, heating load, ventilation rate, the optimal amount of CO₂ enrichment, and daily light integral were calculated with the simulation. The results of this study are being built so that users can be linked through a web page, and software will be designed to reflect the characteristics of cladding materials and greenhouses, cultivation types, and the condition of environmental control facilities for customized environmental control. In addition, environmental information obtained from external meteorological data, as well as recommended standards and set points for each growth stage based on experiments and research, will be provided as optimal environmental factors. This simulation can help growers, students, and researchers to understand the ICT technologies and the changes in the greenhouse microclimate according to the growing conditions.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.27 no.3
• /
• pp.127-143
• /
• 2022

Recently, many attempts to run numerical ocean models in cloud computing environments have been tried actively. A cloud computing environment can be an effective means to implement numerical ocean models requiring a large-scale resource or quickly preparing modeling environment for global or large-scale grids. Many commercial and private cloud computing systems provide technologies such as virtualization, high-performance CPUs and instances, ether-net based high-performance-networking, and remote direct memory access for High Performance Computing (HPC). These new features facilitate ocean modeling experimentation on commercial cloud computing systems. Many scientists and engineers expect cloud computing to become mainstream in the near future. Analysis of the performance and features of commercial cloud services for numerical modeling is essential in order to select appropriate systems as this can help to minimize execution time and the amount of resources utilized. The effect of cache memory is large in the processing structure of the ocean numerical model, which processes input/output of data in a multidimensional array structure, and the speed of the network is important due to the communication characteristics through which a large amount of data moves. In this study, the performance of the Regional Ocean Modeling System (ROMS), the High Performance Linpack (HPL) benchmarking software package, and STREAM, the memory benchmark were evaluated and compared on commercial cloud systems to provide information for the transition of other ocean models into cloud computing. Through analysis of actual performance data and configuration settings obtained from virtualization-based commercial clouds, we evaluated the efficiency of the computer resources for the various model grid sizes in the virtualization-based cloud systems. We found that cache hierarchy and capacity are crucial in the performance of ROMS using huge memory. The memory latency time is also important in the performance. Increasing the number of cores to reduce the running time for numerical modeling is more effective with large grid sizes than with small grid sizes. Our analysis results will be helpful as a reference for constructing the best computing system in the cloud to minimize time and cost for numerical ocean modeling.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.28 no.5
• /
• pp.721-734
• /
• 2022

This study analyzed the spatio-temporal variation characteristics of major environmental factors such as primary productivity (PP), chlorophyll a, nutrients, sinking particle matters, and organic contamination and biochemical composition of surface sediment on a monthly basis for approximately 2 years around shellfish mariculture in Jaran Bay, Korea. In addition, PP in Jaran Bay was compared with that in other coastal areas and related policy plans were proposed. The average PP of the study area was high in summer and autumn with 6.43~115.43 mgC m^-2 hr^-1 range. This was lower than that in Gamak Bay and Masan Bay, whereas higher than that in Garorim Bay and the West Sea. The PP in coastal waters, where many aquaculture farms were distributed, significantly fluctuated. The different size compositions of phytoplanktons constituting chlorophyll a slightly varied by month, and little restriction existed on the productivity of phytoplanktons owing to the depletion of nutrients. Typically, the Redfield ratio was less than 16, indicating that nitrogen was the limiting factor for the growth of phytoplanktons. The biochemical composition of particulate organic matters in the water column showed the highest carbohydrates, but lipids and protein contents were high in surface sediments. The concentration of TOC and AVS of the surface sediments was high at inside of bay, and sometimes, exceeded the environmental criteria of fishing grounds. The organic C:N ratio of sediments ranged from 8.1 to 10.4 on average. PP had the highest correlation with chlorophyll a, nitrogen and protein of particle organic materials. Recently, chlorophyll a, DIN, and DIP of water column trends tended to decrease, however, the contamination of sediments increased. Considering the annual PP of 125.9 gC m^-2 yr^-1 and mariculture area (oyster) of 4.97 km², the annual carbon production from phytoplanktons was estimated to be about 625 tons, and the annual total wet weight of shellfish (oyster) was estimated to be about 6,250 tons.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.50 no.5
• /
• pp.69-79
• /
• 2022

This study aims to assess the current understanding of the landscape of wind power facilities as renewable energy sources that supply sightseeing, tourism, and other opportunities. Therefore, social media data related to the landscape of wind power facilities experienced by visitors from different regions was analyzed. The analysis results showed that the common characteristics of the landscape of wind power facilities are based on the scale of wind power facilities, the distance between overlook points of wind power facilities, the visual openness of the wind power facilities from the overlook points, and the terrain where the wind power facilities are located. In addition, the preference for wind power facilities is higher in places where the shape of wind power facilities and the surrounding landscape can be clearly seen- flat ground or the sea are considered better landscapes. Negative keywords about the landscape appear on Gade Mountain in Taibai, Meifeng Mountain in Taibai, Taiqi Mountain, and Gyeongju Wind Power Generation Facilities on Gyeongshang Road in Gangwon. The keyword 'negation' occurs when looking at wind power facilities at close range. Because of the high angle of the view, viewers can feel overwhelmed seeing the size of the facility and the ridge simultaneously, feeling psychological pressure. On the contrary, positive landscape adjectives are obtained from wind power facilities on flat ground or the sea. Visitors think that the visual volume of the landscape is fully ensured on flat ground or the sea, and it is a symbolic element that can represent the site. This study analyzes landscape awareness based on the opinions of visitors who have experienced wind power facilities. However, wind power facilities are built in different areas. Therefore, landscape characteristics are different, and there are many variables, such as viewpoints and observers, so the research results are difficult to popularize and have limitations. In recent years, landscape damage due to the construction of wind power facilities has become a hot issue, and the domestic methods of landscape evaluation of wind power facilities are unsatisfactory. Therefore, when evaluating the landscape of wind power facilities, the scale of wind power facilities, the inherent natural characteristics of the area where wind power facilities are set up, and the distance between wind power facilities and overlook points are important elements to consider. In addition, wind power facilities are set in the natural environment, which needs to be protected. Therefore, from the landscape perspective, it is necessary to study the landscape of wind power facilities and the surrounding environment.

• Journal of Practical Agriculture & Fisheries Research
• /
• v.25 no.1
• /
• pp.20-35
• /
• 2023

Chemical preservatives have a good effect on antibacterial activity, but many side effects on the human body have been reported. Recently, the development of natural preservatives that are harmless to the human body and have preservative functions and self-efficacy is active. In addition, in order to increase the absorption rate of natural products by the human body, the method of fermentation using strains is also increasing. Therefore, this study selected varieties that are harmless to the human body and have good antibacterial activity. 1. The yield of origin, thickness and solvent was investigated. Scutellaria baicalensis Georgi was made in China and received a yield of 21.88% from 50% ethyl alcohol extract. Salvia miltiorrhiza Bunge was made in Korea and received a yield of 25.62% from 50% ethyl alcohol extract. Dryopteris crassirhizoma Nakai was made in China and received a yield of 6.50% from 70% ethyl alcohol extract. 2. The solid fermentation with the S. baicalensis and S. miltiorrhiza with B. Subtilis yield gained 24.40%, 39.30%, and D. crassirhizoma obtained 11.10% yield when fermented with L. casei. 3. After the liquid fermentation, a clear zone of 9mm was identified for the S. aureus strain in the S. baicalensis, and the antibacterial activity was not confirmed in S. miltiorrhiza and D. crassirhizoma. 4. When the S. baicalensis was fermented with L. Casei, it showed high antibacterial activity in C. albicans and S. aureus. S. miltiorrhiza showed antibacterial activity in S. aureus when it was solid with S. cerevisiae. When the spectators were solid with L. casei and S. cerevisiae, antibacterial activity was high in E. coli and S. aureus. Overall, the antibacterial activity after fermentation was much higher than when fermented. 5. The change in active ingredients was baicalin 101.57, baicalein 28.26, and wogonin 5.33mg/g in the S. baicalensis that did not ferment solid. When solid fermentation with S. cerevisiae, the content of baicalinin with baicalin 94.31, baicalein 30.41, and wogonin 3.57mg/g was found to have increased. S. miltiorrhiza that was not fermented, salvianolic acid A was 1.82mg/g, and when fermented with S. cerevisiae, it increased to 5.70mg/g. The active ingredients of the spectators were flavaspidic acid AP, flavaspidic acid PB, flavaspidic acid AB, and flavaspidic acid BB.

• Journal of agricultural medicine and community health
• /
• v.48 no.1
• /
• pp.28-40
• /
• 2023

Objectives: This study aimed to identify the level of self-assessment of life satisfaction and various factors related to the life satisfaction of Korean physicians. Methods: This study is a secondary data analysis using the 2016 Korean physician survey Korean Physician Survey(KPS) data collected by the Research Institute for Healthcare Policy of the Korean Medical Association. The member database(DB) of the Korean Medical Association was used for sampling and the target population was formed and surveyed by using stratified quota sampling. A questionnaire was sent by E-mail as an online survey method and was conducted for a total of 7 weeks from November 21, 2016 to January 8, 2017. The final number of respondents was 8,564 (response rate 13.8%). In this study, a total of 7,228 physicians, excluding residents and public health doctors who are currently treating patients directly, were studied. Factors affecting the life satisfaction of physicians were analyzed using ordinal logistic regression analysis. Results: The physical factors positively related to the life satisfaction of physicians were those who were in their 60s, female, and thought they had good health status. As for psychological factors, stress was low. As for economic factors, satisfaction with income was high. As for social factors, the physicians lived with their families and were satisfied with the time they could spend with them. Also, the physicians were satisfied with the social respect they received as a doctors. Conclusions: Based on the results of this study, it is thought that a multifaceted approach is needed to increase the life satisfaction of physicians.

• Korean Journal of Poultry Science
• /
• v.50 no.2
• /
• pp.91-99
• /
• 2023

This study aimed to develop a high-productivity breed of Korean native Samgye chicken. We evaluated the production performance of six Korean native chicken combinations (KNC-SCYC, SCYD, SDYC, SDYD, SYYC, SYYD) and GSP-Hanhyup Korean native chickens, with Baeksemi chickens used as a control group. The performance test was conducted from hatching to 7 weeks of age on 756 chickens, and we measured survival rate, body weight, shank length, feed utility, and carcass yield. The overall survival rate was nearly 100% for all strains. However, body weight showed significant differences between strains at all ages (P<0.01), with Baeksemi weighing 863.8±76.9 g, GSP-Hanhyup weighing 804.7±72.5 g, and KNC-combinations weighing 543.0±61.8 g at 5 weeks of age. The duration needed to reach 850 g was estimated to be 34.5 days for Baeksemi, 37.5 days for GSP-Hanhyup, and 45.8-48.8 days for KNC-combinations. Carcass yield percentage was highest for KNC-SYYD combination at 63.3%, followed by Baeksemi at 60.4%, and GSP-Hanhyup at 56.1%. Shank length at 850 g body weight was 7.6 cm for KNC-SYYD combination, 7.8 cm for Baeksemi, and 8.0 cm for GSP-Hanhyup. The feed conversion ratio at 850 g body weight was 1.81 for Baeksemi, 2.17 for GSP-Hanhyup, and 2.27 for KNC-SCYC combination. Our results suggest that the KNC-SYYD combination and GSP-Hanhyup breed have the potential to be used in Samgye production due to their moderate growth performance, higher carcass yield, and shorter shank length, despite their lower growth productivity and feed efficiency when compared to Baeksemi.

• Korean Journal of Heritage: History & Science
• /
• v.56 no.3
• /
• pp.158-171
• /
• 2023

Pottery filled with organic materials was excavated from the G-2 building site of Yongjangseong Fortress, Jingo, a relic of the Goryeo Dynasty. In this study, the characteristics of organic material were confirmed by a scientific analysis of organic material in pottery found at the palace in Yongjangseong, Jindo. In addition, it was intended to review the analysis method to identify the natural resin and to secure characteristic components(biomarkers) for each natural resin and use them as basic data in the future. The organic materials in the pottery were analyzed using attenuated total reflectance Fourier-transformed infrared spectroscopy(ATR-FTIR) and gas chromatography mass spectrometry(GC-MS). The infrared spectral characteristics were estimated to be natural resin, and biomarkers of organic materials were identified as sesquiterpene-based compounds(C₁₅H₂₄, MW 204) and derivatives. The lacquer(T.vemicifluum) is composed mainly of alkenes, alkanes, and catechol. Pine resin(P.densiflora), on the other hand, is primarily composed of diterpenoid(abietic acid, pimaric acid) and Whangchil(yellow lacquer) is identified to have sesquiterpenes(such as selinene, muurolene, calamenene) as its main components. So, the organic material in the pottery can be identified as Whangchil by comparing their compounds with modern resin materials from Dendropanax. morbifera that correspond with the results. Whangchil, which is exuded from the Dendropanax. morbifera, has been used as a natural coating materials since ancient times, and it has been confirmed that the characteristic components are well preserved even 700 years later. It can be assumed that the interior Whangchil was stored not for use as a coating, but rather for ritual purposes when the building was constructed, because the pottery was found near the cornerstone. Furthermore, based on simplified sample preparation using pyrolysis-gas chromatography mass spectrometry(Py-GC-MS), the thermal decomposition products were found to be similar to the characteristic components, suggesting that this method can be applied to the identification of natural resins used in historic artifacts.