• Journal of the Korean Society for information Management
• /
• v.40 no.2
• /
• pp.115-135
• /
• 2023

The purpose of this study is to assess the effectiveness of using deep learning language models to extract references automatically and create a reference database for research reports in an efficient manner. Unlike academic journals, research reports present difficulties in automatically extracting references due to variations in formatting across institutions. In this study, we addressed this issue by introducing the task of separating references from non-reference phrases, in addition to the commonly used metadata extraction task for reference extraction. The study employed datasets that included various types of references, such as those from research reports of a particular institution, academic journals, and a combination of academic journal references and non-reference texts. Two deep learning language models, namely RoBERTa+CRF and ChatGPT, were compared to evaluate their performance in automatic extraction. They were used to extract metadata, categorize data types, and separate original text. The research findings showed that the deep learning language models were highly effective, achieving maximum F1-scores of 95.41% for metadata extraction and 98.91% for categorization of data types and separation of the original text. These results provide valuable insights into the use of deep learning language models and different types of datasets for constructing reference databases for research reports including both reference and non-reference texts.

• Analytical Science and Technology
• /
• v.21 no.4
• /
• pp.259-271
• /
• 2008

The analytical method of extracting compounds from human blood to examine accumulated organochlorine pesticides (OCPs) has been widely used the traditional liquid-liquid extraction (LLE) method and solid-phase extraction (SPE) method, yet these methods have certain limitations in purification and usafe of a large amount of sample. In order to overcome the se problems reside in these, solid-phase microextraction (SPME), known as a highly efficient extration method with less samples and relatively simple, was employed to collect 18 different kinds of OCPs in blood as extraction method in this study. To optimize extraction method, we examine various experimental SPME-parameters such as adsorption (fiber type, adsorption time, adsorption temperature, salting out effect), and desorption (desorption time, desorption temperature etc.). From the experimental results, the optimal conditions are as follows: fiber was polyacrylate with $85{\mu}m$, adsorption time was for 5 min, adsorption optimum temperature was at $280^{\circ}C$, and salting out effect was NaCl with 0.1 g. MDL, precision and accuracy was in the ranges of 0.05~0.20 ng/mL, 5.59~13.39%, respedively, and accuracy was -0.5% ~24.5% for all OCPs.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.5
• /
• pp.323-330
• /
• 2023

The development of an analysis model that reflects the microstructure characteristics of polyvinyl alcohol (PVA) fiber-reinforced cementitious composites, which have a highly complex microstructure, enables synergy between efficient material design and real experiments. PVA fiber orientations are an important factor that influences the mechanical behavior of PVA fiber-reinforced cementitious composites. Owing to the difficulty in distinguishing the gray level value obtained from micro-CT images of PVA fibers from adjacent phases, fiber segmentation is time-consuming work. In this study, a micro-CT test with a voxel size of 0.65 ㎛³ was performed to investigate the three-dimensional distribution of fibers. To segment the fibers and generate training data, histogram, morphology, and gradient-based phase-segmentation methods were used. A U-net model was proposed to segment fibers from micro-CT images of PVA fiber-reinforced cementitious composites. Data augmentation was applied to increase the accuracy of the training, using a total of 1024 images as training data. The performance of the model was evaluated using accuracy, precision, recall, and F1 score. The trained model achieved a high fiber segmentation performance and efficiency, and the approach can be applied to other specimens as well.

• Journal of the Korean Society of Radiology
• /
• v.17 no.7
• /
• pp.1099-1108
• /
• 2023

The mortise view radiography procedure is an ankle joint examination and observes the presence of trauma, sprain, or dislocation suspected in the ankle joint. The auxiliary equipment used during the mortise view radiography procedure can generate artifacts in the radiograph images and is not diverse enough to be custom-made for each patient; not cost-efficient. The purpose of this study is to create a custom assistive device to support mortise view radiography procedure. This study utilized 3D printing technology to create the mortise view radiography procedure assistive device (ShinHan Device; SHD). The lengths of the tibiotalar joint (TTJ), talar calcaneal joint (TCJ), and medial joint (MJ) were measured and evaluated by five researchers using both SHD and the prototype Hologic tool. The mean ranges were found to be 39.42-39.47 mm for TTJ, 31.41-31.57 mm for TCJ, and 21.21-21.23 mm for MJ while using SHD device. On the other hand, the measurements showed mean ranges of 39.73-39.79 mm for TTJ, 31.46-31.50 mm for TCJ, and 21.31-21.35 mm for MJ while using the Hologic tool. Based on this study results, the error ranges at all positions decreased by 24% for TTJ, 17% for TCJ, and 36% for MJ when using SHD device compared to the Hologic tool. Moreover, when SHD was used, it allowed for a highly reproducible examination posture (ICC = 0.99), and it enabled the acquisition of radiograph images without artifacts, which were present in the Hologic tool.

• Clean Technology
• /
• v.29 no.4
• /
• pp.244-248
• /
• 2023

Recently, displays play an important role in quickly delivering a lot of information. Research is underway to reproduce various colors close to natural colors. In particular, research is being conducted on the light emitting structure of displays as a method of expressing accurate and rich colors. Due to the advancement of technology and the miniaturization of devices, the need for small but high visibility displays with high efficiency in energy consumption continues to increase. Efforts are being made in various ways to improve OLED efficiency, such as improving carrier injection, structuring devices that can efficiently recombine electrons and holes in a numerical balance, and developing materials with high luminous efficiency. In this study, the electrical and optical properties of the seven-layer stacked structure rear-light emitting blue OLED device were analyzed. 4,4'-Bis(carazol-9-yl)biphenyl:Ir(difppy)2(pic), a blue light emitting material that is easy to manufacture and can be highly efficient and brightened, was used. OLED device manufacturing was performed via the in-situ method in a high vacuum state of 5×10-8 Torr or less using a Sunicel Plus 200 system. The experiment was conducted with a seven-layer structure in which an electron or hole blocking layer (EBL or HBL) was added to a five-layer structure in which an electron or hole injection layer (EIL or HIL) or an electron or hole transport layer (ETL or HTL) was added. Analysis of the electrical and optical properties showed that the device that prevented color diffusion by inserting an EBL layer and a HBL layer showed excellent color purity. The results of this study are expected to greatly contribute to the R&D foundation and practical use of blue OLED display devices.

• Journal of the Korean GEO-environmental Society
• /
• v.25 no.4
• /
• pp.13-20
• /
• 2024

In this paper, a parallel analysis algorithm for Smoothed Particle Hydrodynamics (SPH), one of the numerical methods for fluidic materials, is introduced. SPH, which is a meshless method, can represent the behavior of a continuum using a particle-based approach, but it demands substantial computational resources. Therefore, parallel analysis algorithms are essential for SPH simulations. The domain decomposition algorithm, which divides the computational domain into partitions to be independently analyzed, is the most representative method among parallel analysis algorithms. In Discrete Element Method (DEM) and Molecular Dynamics (MD), the Cartesian coordinate-based domain decomposition method is popularly used because it offers advantages in quickly and conveniently accessing particle positions. However, in SPH, it is important to share particle information among partitioned domains because SPH particles are defined based on information from nearby particles within the smoothing length. Additionally, maintaining CPU load balance is crucial. In this study, a highly parallel efficient algorithm is proposed to dynamically minimize the size of orthogonal domain partitions to prevent excess CPU utilization. The efficiency of the proposed method was validated through numerical analysis models. The parallel efficiency of the proposed method is evaluated for up to 30 CPUs for fluidic models, achieving 90% parallel efficiency for up to 28 physical cores.

• Korean Journal of Environmental Biology
• /
• v.41 no.4
• /
• pp.370-385
• /
• 2023

Cyanobacterial harmful algal blooms (Cyano-HABs) are an international environmental problem that negatively affects the ecosystem as well as the safety of water resources by discharging cyanotoxins. In particular, the discharge of microcystins (MCs), a highly toxic substance, has been studied most actively, and various water treatment methods have been proposed for this purpose. In this paper, we reviewed adsorption technology, which is recognized as the most feasible, economical, and efficient method among suggested treatment methods for removing MCs. Activated carbons (AC) are widely used adsorbents for MCs removal, and excellent MCs adsorption performance has been reported. Research on alternative adsorption materials for AC such as biochar and biosorbents has been conducted, however, their performance was lower compared to activated carbon. The impacts of adsorbent properties(characteristics of pore surface chemistry) and environmental factors (solution pH, temperature, natural organic matter, and ionic strength) on the MCs adsorption performance were also discussed. In addition, toward effective control of MCs, the possibility of the direct removal of harmful cyanobacteria as well as the removal of dissolved MCs using adsorption strategy was examined. However, to fully utilize the adsorption for the removal of MCs, the application and optimization under actual environmental conditions are still required, thereby meeting the environmental and economic standards. From this study, crucial insights could be provided for the development and selection of effective adsorbent and subsequent adsorption processes for the removal of MCs from water resources.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.4
• /
• pp.443-449
• /
• 2024

This study aims to analyze the complex impact of technological changes in the Fourth Industrial Revolution era and demographic shifts in Korea on military personnel management, and to explore innovative strategies for the Korean military's workforce operations. The research findings indicate that changes in future battlefield environments and the introduction of advanced technologies necessitate a fundamental restructuring of military personnel, emphasizing a shift towards a highly specialized and elite workforce. Key research findings are as follows: First, the military application of cutting-edge technologies, such as unmanned systems, autonomous weapon systems, and AI-based decision support systems, is expanding. Second, this technological advancement requires a restructuring of personnel to foster a technology-intensive elite force, including optimizing troop size, reorganizing unit structures, and increasing the utilization of civilian expertise. Third, strategies for securing high-tech talent include strengthening internal technology talent development programs, establishing systems to attract civilian experts, and building a talent development system through industry-academia-research cooperation. The significance of this study lies in providing a theoretical and practical foundation for building a future-oriented and efficient Korean military organization by presenting innovative measures for military human resource management systems suitable for the Fourth Industrial Revolution era. For these changes to be successfully implemented, cooperation among relevant stakeholders, including the military, government, academia, and industry, is essential, supported by comprehensive national-level planning and support.

• Microbiology and Biotechnology Letters
• /
• v.37 no.4
• /
• pp.377-382
• /
• 2009

Viral safety is an important prerequisite for clinical preparations of all biopharmaceuticals derived from plasma, cell lines, or tissues of human or animal origin. To ensure the safety, implementation of multiple viral clearance (inactivation and/or removal) steps has been highly recommended for manufacturing of biopharmaceuticals. Of the possible viral clearance strategies, Ultraviolet-C (UVC) irradiation has been known as an effective viral inactivating method. However it has been dismissed by biopharmaceutical industry as a result of the potential for protein damage and the difficulty in delivering uniform doses. Recently a continuous flow UVC reactor (UVivatec) was developed to provide highly efficient mixing and maximize virus exposure to the UV light. In order to investigate the effectiveness of UVivatec to inactivate viruses without causing significant protein damage, the feasibility of the UVC irradiation process was studied with a commercial therapeutic protein. Recovery yield in the optimized condition of $3,000\;J/m^2$ irradiation was more than 98%. The efficacy and robustness of the UVC reactor was evaluated with regard to the inactivation of human immunodeficiency virus (HIV), hepatitis A virus (HAV), bovine herpes virus (BHV), bovine viral diarrhea virus (BVDV), porcine parvovirus (PPV), bovine parvovirus (BPV), minute virus of mice (MVM), reovirus type 3 (REO), and bovine parainfluenza virus type 3 (BPIV). Non enveloped viruses (HAV, PPV, BPV, MVM, and REO) were completely inactivated to undetectable levels by $3,000\;J/m^2$ irradiation. Enveloped viruses such as HIV, BVDV, and BPIV were completely inactivated to undetectable levels. However BHV was incompletely inactivated with slight residual infectivity remaining even after $3,000\;J/m^2$ irradiation. The log reduction factors achieved by UVC irradiation were ${\geq}3.89$ for HIV, ${\geq}5.27$ for HAV, 5.29 for BHV, ${\geq}5.96$ for BVDV, ${\geq}4.37$ for PPV, ${\geq}3.55$ for BPV, ${\geq}3.51$ for MVM, ${\geq}4.20$ for REO, and ${\geq}4.15$ for BPIV. These results indicate that UVC irradiation using UVivatec was very effective and robust in inactivating all the viruses tested.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.43 no.1
• /
• pp.162-171
• /
• 2014

As the foodservice industry is labor intensive, the efficient management of human resources is an essential element for improving operational efficiency. This study was designed to investigate the performance level of human resource management in self-operated school foodservices and to examine the relationship between human resource management, organizational commitment, and productivity. Whether these factors rated differently by employment and job types was also examined. Data were collected through questionnaires distributed to 60 dieticians and 240 cooks in Seongnam and Yongin. Frequency analysis, factor analysis, the t-test and multiple regression analysis were tested using Minitab. Human resource management practices rated highly in the dimension of employee training (3.51), followed by working environment (3.39), and turnover management (3.37). In contrast, reward management (1.73) and working condition (1.56) received the lowest score. Organizational commitment scored 3.65 points, and showed that dedication (3.82) was the highest score; however, self-esteem (3.50) was the lowest score. For human resource management, there were significant differences between employment types as full time workers (3.85) rated management more highly than part-time worker (3.43). Overall productivity was fairly high, especially in high school foodservices. Only the performance appraisal had an influence on organizational commitment (${\beta}$=0.292, P<0.05). Productivity was positively correlated with human resource management (r=0.432, P<0.001) and organizational commitment (r=0.36, P<0.01). In conclusion, school foodservices need to establish objective performance standards, and increase employee morale by enhancing reward systems and working conditions. In particular, irregular employees require training with job performance standards and given a proper reward program, depending on their performance, to improve organizational commitment.