• The Journal of Society for e-Business Studies
• /
• v.23 no.2
• /
• pp.83-96
• /
• 2018

Extracting keywords representing documents is very important because it can be used for automated services such as document search, classification, recommendation system as well as quickly transmitting document information. However, when extracting keywords based on the frequency of words appearing in a web site documents and graph algorithms based on the co-occurrence of words, the problem of containing various words that are not related to the topic potentially in the web page structure, There is a difficulty in extracting the semantic keyword due to the limit of the performance of the Korean tokenizer. In this paper, we propose a method to select candidate keywords based on semantic similarity, and solve the problem that semantic keyword can not be extracted and the accuracy of Korean tokenizer analysis is poor. Finally, we use the technique of extracting final semantic keywords through filtering process to remove inconsistent keywords. Experimental results through real web pages of small business show that the performance of the proposed method is improved by 34.52% over the statistical similarity based keyword selection technique. Therefore, it is confirmed that the performance of extracting keywords from documents is improved by considering semantic similarity between words and removing inconsistent keywords.

• Research in Plant Disease
• /
• v.10 no.2
• /
• pp.145-149
• /
• 2004

'Hongro' is one of the most important apple cultivars whose growing area is increasing because of its good quality. Recently fruit shrinking symptom causing decrease of fruit size, juice, and quality, appears in some commercial 'Hongro' orchards. The average frequency of occurrence of fruit shrinking symptom was 12 % of total trees investigated and Apple chlorotic leaf spot virus (ACLSV) was detected from all the trees showing fruit shrinking symptom by ELISA. A typical virus infection symptom of leaf epinasty and stem necrosis appeared on woody indicators, Spy227 and Virginia crab grafted with infected trees and all the grafted trees showed positive reaction to ACLSV antiserum by ELISA. It was proved that ACLSV can be easily transmitted by grafting. ACLSV was also isolated from the leaves of C. quinoa inoculated with sprouting leaf sap of infected trees. To prove that the fruit shrinking symptom was caused by ACLSV infection, ACLSV-infected scion was grafted on virus-free 'Hongro/M9' and the fruit characteristics were investigated. Consequently the same symptoms of fruit size and juice decreasing were observed from the trees grafted with ACLSV-infected scion. Therefore, it is suggested that the fruit shrinking symptom is caused by ACLSV infection and 'Hongro' can be classified as sensitive cultivar to ACLSV.

• Korean Journal of Environmental Biology
• /
• v.37 no.4
• /
• pp.690-706
• /
• 2019

Paddy fields play an important role in the biodiversity conservation of agroecosystems as man-made wetlands. This study focused on the floristic composition and characteristics of paddy fields in South Korea. The flora surveys were conducted in 33 areas in nine provinces at two times (June and August) in 2016. The results showed that the vascular plants of the paddy fields in South Korea included 389 taxa, composed of 51 families, 219 genera, 350 species, four subspecies, 30 varieties, and five forms. Among the 51 families, Gramineae was the most diverse in species (76 taxa), followed by Compositae (61 taxa), Leguminosae (32 taxa), Cyperaceae (27 taxa), Polygonaceae (15 taxa), and Scrophulariaceae (12 taxa). Based on the occurrence frequency of each species, Digitaria ciliaris (Retz.) Koel. (100%), and Echinochloa crus-galli (L.) P.Beauv. (100%) were the highest, followed by Acalypha australis L. (97.0%), Bidens frondosa L. (97.0%), Alopecurus aequalis Sobol. (97.0%), Rorippa palustris (Leyss.) Besser (93.9%), Ludwigia prostrata Roxb. (93.9%), Lemna perpusilla Torr. (93.9%), Mazus pumilus (Burm. f.) Steenis(90.9%), Rorippa palustris(Leyss.) Besser(87.9%), Eclipta prostrata(L.) L.(87.9%), Commelina communis L.(87.9%), Artemisia princeps Pamp.(84.8%), Erigeron annuus (L.) Pers.(84.8%), Cyperus microiria Steud.(84.8%), and Aneilema keisak(Hassk.) Hand.-Mazz. (81.8%). The biological type of paddy fields in South Korea was determined to be Th-R₅-D₄-t type. Rare plants were found in six taxa: Aristolochina contorta Bunge, Penthorum chinense Pursh, Sagittaria sagittifolia subsp. leucopetala (Miq.) Hartog, Ottelia alismoides (L.) Pers., Pseudoraphis ukishiba Ohwi, and Acorus calamus L. Eighty taxa contained naturalized plants composed of 15 families, 54 genera, 76 species, three varieties, and one form. The urbanization and naturalization indices were 24.8% and 20.6%, respectively.

• Journal of the Korean earth science society
• /
• v.30 no.6
• /
• pp.734-743
• /
• 2009

In order to examine how climatological condition can influence on urban scale particulate air pollutants, single and cross spectrum analysis have been performed to daily mean concentrations of particulate matters ($PM_{10}$) in Busan together with the climatological variables over the Asian dust source regions. Single power spectrum analysis of $PM_{10}$ concentrations in Busan shows that, aside from the typical and well-known periodicities, 3-4 year of peak periodicity of power spectrum density was identified. In cross spectrum analysis, this 3-4 year periodicity is found to have a strong positive correlation with the wind speed and pressure, and negative with the temperature and relative humidity, which is rather consistent with both characteristics of air mass during the Asian dust event whose periodicities have been recorded inter-annually over the Korean urban cities. Over the Asian dust source regions, $PM_{10}$ vs. precipitation shows no significant periodicity from the time series of precipitation data, but the periodicity of EDI (Effective Drought Index) shows some interannual variabilities ranging from 2 to 4 years over the various source regions, suggesting that, rather than precipitation itself, the EDI could be more closely associated with the occurrence frequency of Asian dust and interannual variability of urban particle concentrations in Korean cities.

• Journal of fish pathology
• /
• v.23 no.3
• /
• pp.303-311
• /
• 2010

Our previous finding summarizing that larval sieving process is inevitable but triggers the outbreak of scutica-like ciliates (SLCs) in the seed production of Patinopecten yessoensis urged further study to determine best suggestable sieving interval in an agreeable range of water quality. In the mass seed production of the scallop, SLC outbreak was closely related to the larval sieving in which larvae were drained on the basis of every 3-day (5T), 5-day (3T), 7-day (2T), or 9-day (1T) from culture tanks onto a mesh screen and placed back into new water in cleaned tanks. The larval performance of growth and survival was clearly dependent on the sieving intervals. It was in order of 3T, 5T, 2T, and 1T for both of growth and survival and in reverse order for SLC infection frequency, confirming that larval sieving is necessary but damageable if it overwhelms the larval resistance. Interestingly, the larval damages by the sieving persisted to their ensuing spat life in terms of nursery growth, survival, and abnormality.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.28 no.5
• /
• pp.503-516
• /
• 1995

Macrobenthic community was studied in Deukryang Bay, southwestern coast of Korea, in November 1991 and Januavy 1992. Hacrobenthic fauna consisted of 118 species with 20,767 individuals. The number of species and occurrence frequency of the major groups were as follows: 52 polychaetes species $(44.1\%)$, 45 crustaceans species $(38,1\%)$, 14 mollusks species $(11.9\%)$ and 7 other group species $(5.9\%)$. The mean density was $1432.2ind/m^2$. The density of major groups showed that mollusks was 920.4$(ind./m^2)$, polyrhaetes 275.8$(ind./m^2)$, and crustaceans 219.6$(ind./m^2)$. The dominant species were Musculista senhousia (Bivalvia), Eteone longa (Polychaeta), Nippopisella nagatai (Amphipoda) with density, 632.8$ind./m^2$, 37.8$ind./m^2$, and 35.3$ind./m^2$, respectively. This area could be divided into four areas by the cluster analysis based on the species composition.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.11
• /
• pp.1-9
• /
• 2021

3D-NAND flash memory provides high capacity per unit area by stacking 2D-NAND cells having a planar structure. However, due to the nature of the lamination process, there is a problem that the frequency of error occurrence may vary depending on each layer or physical cell location. This phenomenon becomes more pronounced as the number of write/erase(P/E) operations of the flash memory increases. Most flash-based storage devices such as SSDs use ECC for error correction. Since this method provides a fixed strength of data protection for all flash memory pages, it has limitations in 3D NAND flash memory, where the error rate varies depending on the physical location. Therefore, in this paper, pages and layers with different error rates are classified into clusters through the K-means machine learning algorithm, and differentiated data protection strength is applied to each cluster. We classify pages and layers based on the number of errors measured after endurance test, where the error rate varies significantly for each page and layer, and add parity data to stripes for areas vulnerable to errors to provides differentiate data protection strength. We show the possibility that this differentiated data protection policy can contribute to the improvement of reliability and lifespan of 3D NAND flash memory compared to the protection techniques using RAID-like or ECC alone.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.3
• /
• pp.583-589
• /
• 2021

This study is a study on the development of virtual reality to prepare for the increase in disasters of households living with companion animals due to the increase of companion animals. The increase in single-person households and DINKs(Double Income, No Kid) along with the low birth rate and aging population is raising the risk of disasters caused by companion animals in particular. Among these disasters, there is an increase in the occurrence of fires primarily due to the raising of companion animals. Electric stove fires caused by pets are the most common fires. In particular, the frequency of electric stove fires caused by cats is the highest. Careful precautions by the owner are necessary to reduce fires caused by pets. Parenting of companion animals causes pet loss syndrome due to emotional exchange. There are injuries to pets in escalators and injuries to owners in elevators due to disasters caused by the owner's negligence. In order to reduce injuries on escalators and elevators, basic etiquette for using escalators and elevators with pets is required as basic etiquette. It is necessary to utilize virtual reality to reduce disasters caused by such companion animals. Virtual reality can be experienced without a physical space in advance training to overcome disasters, so real disaster cases can be experienced immersively. Therefore, learning how to reduce fires caused by companion animals, disasters caused by owner's negligence, and petloss syndrome through virtual reality will greatly contribute to disaster prevention and reduction of social costs.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_2
• /
• pp.669-683
• /
• 2022

The frequency and damage of forest fires have tended to increase over the past 20 years. In order to effectively respond to forest fires, information on forest fire damage should be well managed. However, information on the extent of forest fire damage is not well managed. This study attempted to present a method that extracting information on the area of forest fire in real time and quasi-real-time using visible infrared imaging radiometer suite (VIIRS) images. VIIRS data observing the Korean Peninsula were obtained and visualized at the time of the East Coast forest fire in March 2022. VIIRS images were classified without supervision using iterative self-organizing data analysis (ISODATA) algorithm. The results were reclassified using the relationship between the burned area and the location of the flame to extract the extent of forest fire. The final results were compared with verification and comparison data. As a result of the comparison, in the case of large forest fires, it was found that classifying and extracting VIIRS images was more accurate than estimating them through forest fire occurrence data. This method can be used to create spatial data for forest fire management. Furthermore, if this research method is automated, it is expected that daily forest fire damage monitoring based on VIIRS will be possible.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.34 no.4
• /
• pp.29-54
• /
• 2023

This study aimed to analyze the research trends of journal articles published by medical schools representing Korean modern. A total of 682 were selected from two journals published by Medical College in Keijo and Keijo Imperial University Medical Faculty. In results, the affiliations of authors who participated in Acta Medicinalia in Keijo included various schools and hospitals, and the authors' major was found to be similar in basic medicine and clinical medicine. In The Keijo Journal of Medicine, only school-affiliated authors participated, and 96.33% of the authors were majors in basic medicine. Co-occurrence network analysis was conducted on MeSH terms from the title of the article using MeSH on Demand, and the keyword that derived in both journals was 'erythrocytes', which analyzed the condition of red blood cells according to organs and diseases. In frequency analysis, a common area of research in both journals was the study focusing on blood and blood cells, and the study of anemia and tuberculosis, which were prevalent diseases at the time. As for comparing each journal, Acta Medicinalia in Keijo has focused on inflammatory diseases and clinical pathological studies in humans, and The Keijo Journal of Medicine has focused on anatomical studies on animals and pharmacological studies on medicines. Through this study, it was possible to identify the research topics and major keywords in two medical schools with different founding goals.