• Korean Journal of Heritage: History & Science
• /
• v.48 no.3
• /
• pp.4-29
• /
• 2015

To protect cultural heritages from damage and destruction, evaluating the crime prevention environments is considered extremely important. This study analyzed the crime patterns related to cultural heritages, classified the crime environments by their types, and deduced the elements of the CPTED(Crime Prevention Through Environment Design), aiming to present the indices for evaluating the crime prevention environments. The results of this study can be summarized as follows. First, the crimes related to cultural heritages that must be prevented were identified as the night time trespassing and arson. According to the results of the analysis of external environments based on crime actions, the crime prevention environments of cultural heritages were classified into 10 types. Second, the important evaluation principles of the cultural heritage CPTED were the access control, surveillance reinforcement and the surrounding environment. Third, the access control that cover the internal region, boundary, external region and surroundings were classified into 22 indices. The surveillance reinforcement covers natural, organized and mechanical surveillance with 21 indices. Fourth, the applicability of the CPTED evaluation index was presented according to the types of the cultural crime prevention environments. The results confirmed that the maximum 43 indices were applicable to the seowon(lecture hall), hyanggyo(Confucian school), and gwana(district government office), and the minimum 10 indices, to the ramparts. Finally, the 43 indices were applied to Donam Seowon to validate their applicability. The results confirmed that most of the indices were applicable with the partial supplements. The evaluation index presented in this study is likely to contribute to studies in the cultural heritage CPTED field and to the protection of cultural heritages. Furthermore, this study is considered significant because it unleashed continuous concerns on and developments of CPTED. However, as the field survey to validate the applicability of the indices was limited to only one type, it may require further objective verification such as through an expert's examination of the validity and applicability of the evaluation index. In addition, to accommodate the index in related policies and systems, more precise verifications of the indices by type are considered necessary.

• Journal of Nutrition and Health
• /
• v.52 no.4
• /
• pp.332-341
• /
• 2019

Purpose: Poor nutrition in hospitalized patients is closely linked to an increased risk of infection, which can result in complications affecting mortality, as well as increased length of hospital stay and hospital costs. Therefore, adequate nutritional support is essential to manage the nutritional risk status of patients. Nutritional support needs to be preceded by nutrition screening, in which accuracy is crucial, particularly for the initial screening. To perform initial nutrition screening of hospitalized patients, we used the Catholic Kwandong University (CKU) Nutritional Risk Screening (CKUNRS) tool, originally developed at CKU Hospital. To validate CKUNRS against the Patient-Generated Subjective Global Assessment (PG-SGA) tool, which is considered the gold standard for nutritional risk screening, results from both tools were compared. Methods: Nutritional status was evaluated in 686 adult patients admitted to CKU Hospital from May 1 to July 31, 2018 using both CKUNRS and PG-SGA. Collected data were analyzed, and the results compared, to validate CKUNRS as a nutrition screening tool. Results: The comparison of CKUNRS and PG-SGA revealed that the prevalence of nutritional risk on admission was 15.6% (n = 107) with CKUNRS and 44.6% (n = 306) with PG-SGA. The sensitivity and specificity of CKUNRS to evaluate nutritional risk status were 98.7% (96.8 ~ 99.5) and 33.3% (28.1 ~ 39.0), respectively. Thus, the sensitivity was higher, but the specificity lower compared with PG-SGA. Cohen's kappa coefficient was 0.34, indicating valid agreement between the two tools. Conclusion: This study found concordance between CKUNRS and PG-SGA. However, the prevalence of nutritional risk in hospitalized patients was higher when determined by CKUNRS, compared with that by PG-SGA. Accordingly, CKUNRS needs further modification and improvement in terms of screening criteria to promote more effective nutritional support for patients who have been admitted for inpatient care.

• Journal of Life Science
• /
• v.29 no.7
• /
• pp.800-808
• /
• 2019

Charcot-Marie-Tooth disease (CMT) is a group of rare peripheral neuropathies characterized by progressive muscle weakness and atrophy and areflexia in the upper and lower extremities. The most common subtype of CMT is CMT1A, which is caused by a tandem duplication of the PMP22 gene in the 17p12 region. Patients with CMT1A show a loose genotype-phenotype correlation, which suggests the existence of secondary genetic or association factors. Recently, polymorphisms of rs71428439 (n.83A>G) and rs2292832 (n.86T>C) in the MIR149 have been reported to be associated with late onset and mild phenotypic CMT1A severity. The aim of this study was to examine the intrafamilial heterogeneities of clinical phenotypes according to the genotypes of these two SNPs in MIR149. For this study, we selected 6 large CMT1A families who showed a wide range of phenotypic variation. This study suggested that both SNPs were related to the onset age and severity in the dominant model. In particular, the AG+GG (n.83A>G) and TC+CC genotypes (n.86T>C) were associated to late onset and mild symptoms. Motor nerve conduction velocity (MNCV) was not related to the MIR149 genotypes. These results were consistent with the previous studies. Therefore, we suggest that the rs71428439 and rs2292832 variants in MIR149 may serve as genetic modifiers of CMT1A intrafamilial phenotypic heterogeneity, as they have a role in the unrelated patients. This is the first study to show an association using large families with variable clinical CMT1A phenotypes. The results will be helpful in the molecular diagnosis and treatment of patients with CMT1A.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.6
• /
• pp.288-296
• /
• 2019

Radiation cooling has used ceilings or floors as cooling surfaces. In such cases, to avoid moisture condensation on the surface, the surface temperature needs be higher than the dew point temperature or an additional dehumidifier is added. In this study, with a goal for residential application, intentional moisture condensation on the cooling surface was attempted, which increased the cooling capacity and improved the indoor comfortness. This method included two separate refrigeration cycles - convection-type dehumidifying cycle and the panel cooling cycle. Test results on the panel cooling cycle showed that, at the standard outdoor ($35^{\circ}C/24^{\circ}C$) and indoor ($27^{\circ}C/19.5^{\circ}C$) condition, the refrigerant flow rate was 8.8 kg/h, condensation temperature was $51^{\circ}C$, evaporation temperature was $8.8^{\circ}C$, cooling capacity was 376 W and COP was 1.75. Furthermore, the panel temperature was uniform within $1^{\circ}C$ (between $13^{\circ}C$ and $14^{\circ}C$). As the relative humidity decreased, the cooling capacity decreased. However, the power consumption remained approximately constant. In the convection-type dehumidification cycle, the refrigerant flow rate was 21.1 kg/h, condensation temperature was $61^{\circ}C$, evaporation temperature was $5.0^{\circ}C$, cooling capacity was 949 W and COP was 2.11 at the standard air condition. When both the radiation panel cooling and the dehumidification cycle operated simultaneously, the cooling capacity of the radiation panel cycle was 333 W and that of the dehumidification cycle was 894 W, and the COP was 1.89. As the fan flow rate decreased, both the cooling capacity of the radiation panel and the dehumidification cycle decreased, with that of the dehumidification cycle decreasing at a higher rate. Finally, a possible control logic depending on the change of the cooling load was proposed based on the results of the present study.

• Journal of Life Science
• /
• v.28 no.11
• /
• pp.1369-1378
• /
• 2018

Prunus mume Siebold & Zucc., a member of the Rosaceae family (called Maesil in Korea), has been widely distributed in East Asia, e.g. Korea, Japan and China, and its fruit has been used as a traditional drug and health food. In this study, we evaluated physicochemical properties and physiological activities of condensed Prunus mume juice treated with pectinase (PJ). The values of total acidity, pH, sugar contents, turbidity moisture content of the PJ were 35.81%, 2.73, $54.36^{\circ}Brix$, 2.75 and 51.32%, respectively. The PJ had effective DPPH radical scavenging activity, reducing power effect, $H_2O_2$ scavenging activity and ${\beta}$-carotene bleaching effect. DPPH radical scavenging activities of PJ was 46.31%; their reducing power ($OD_{700}$) was 1.80; $H_2O_2$ scavenging activity of PJ was 91.62%; and ${\beta}$-carotene bleaching effect of PJ was 73.02%. Also, PJ showed effective levels of ${\alpha}$-glucosidase inhibition activity. The cell viability was measured by SRB assay. The PJ significantly decreased the cell viability of mouse melanoma cells (B16) and human melanoma cells (SK-MEL-2 and SK-MEL-28) in a dose-dependent manner, however, there was no effect on human keratinocyte HaCaT. In morphological study, PJ-treated SK-MEL-2 cells showed distorted and shrunken cell masses. Total polyphenol contents and total flavonoid contents of PJ were 588.31 mg% (gallic acid equivalent) and 860.45 mg% (rutin equivalent). The antiproliferative effect of PJ seems to be associated with the antioxidant activity of its flavonoid and polyphenol contents. In conclusion, PJ may be beneficial in development of a functional food material.

• Journal of Bio-Environment Control
• /
• v.28 no.1
• /
• pp.38-45
• /
• 2019

Recycling of drained nutrient solution in hydroponic cultivation of horticultural crops is important in the conservation of the water resources, reduction of production costs and prevention of environmental contamination. Objective of this research was to obtain the fundamental data for the development of a recirculation system of hydroponic solution in semi-forcing cultivation of 'Bonus' tomato. To achieve the objective, tomato plants were cultivated for 110 days and the contents of inorganic elements in plant, supplied and drained nutrient solution were analyzed when crop growth were in the flowering stage of 2nd to 8th fruiting nodes. The T-N content of the plants based on above-ground tissue were 4.1% at the flowering stage of 2nd fruiting nodes (just after transplanting), and gradually get lowered to 3.9% at the flowering stage of 8th fruiting nodes. The tissue P contents were also high in very early stage of growth and development and were maintained to similar contents in the flowering stage of 3rd to 7th fruiting nodes, but were lowed in 8th node stages. The tissue Ca, Mg and Na contents in early growth stages were lower than late growth stages and the contents showed tendencies to rise as plants grew. The concentration differences of supplied nutrient solution and drained solution in $NO_3-N$, P, K, Ca, and Mg were not significant until 5 weeks after transplanting, but the concentration of those elements in drained solution rose gradually and maintained higher than those in supplied solution. The concentrations of B, Fe, and Na in drained solution were slightly higher in the early stages of growth and development and were significantly higher in the mid to late stages of growth than those in supplied solution. The above results would be used as a fundamental data for the correction in the inorganic element concentrations of drained solution for semi-forcing hydroponic cultivation of tomato.

• Journal of Science and Technology Studies
• /
• v.19 no.2
• /
• pp.169-206
• /
• 2019

This article seeks to explore new relationships and ethics of human and technology by analyzing a cultural imaginary produced by artificial intelligence. Drawing on theoretical reflections of the Feminist Scientific and Technological Studies which understand science and technology as the matter of care(Puig de la Bellacas, 2011), this paper focuses on the fact that artificial intelligence and robots materialize cultural imaginary such as autonomy. This autonomy, defined as the capacity to adapt to a new environment through self-learning, is accepted as a way to conceptualize an authentic human or an ideal subject. However, this article argues that artificial intelligence is mediated by and dependent on invisible human labor and complex material devices, suggesting that such autonomy is close to fiction. The recent growth of the so-called 'assistant technology' shows that it is differentially visualizing the care work of both machines and humans. Technology and its cultural imaginary hide the care work of human workers and actively visualize the one of the machine. And they make autonomy and agency ideal humanness, leaving disabled bodies and dependency as unworthy. Artificial intelligence and its cultural imaginary negate the value of disabled bodies while idealizing abled-bodies, and result in eliminating the real relationship between man and technology as mutually dependent beings. In conclusion, the author argues that the technology we need is not the one to exclude the non-typical bodies and care work of others, but the one to include them as they are. This technology responsibly empathizes marginalized beings and encourages solidarity between fragile beings. Inspired by an art performance of artist Sue Austin, the author finally comes up with and suggests 'artificial intelligence in wheelchair' as an alternative figuration for the currently dominant 'autonomous artificial intelligence'.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.43-62
• /
• 2019

At one time, the anomaly detection sector dominated the method of determining whether there was an abnormality based on the statistics derived from specific data. This methodology was possible because the dimension of the data was simple in the past, so the classical statistical method could work effectively. However, as the characteristics of data have changed complexly in the era of big data, it has become more difficult to accurately analyze and predict the data that occurs throughout the industry in the conventional way. Therefore, SVM and Decision Tree based supervised learning algorithms were used. However, there is peculiarity that supervised learning based model can only accurately predict the test data, when the number of classes is equal to the number of normal classes and most of the data generated in the industry has unbalanced data class. Therefore, the predicted results are not always valid when supervised learning model is applied. In order to overcome these drawbacks, many studies now use the unsupervised learning-based model that is not influenced by class distribution, such as autoencoder or generative adversarial networks. In this paper, we propose a method to detect anomalies using generative adversarial networks. AnoGAN, introduced in the study of Thomas et al (2017), is a classification model that performs abnormal detection of medical images. It was composed of a Convolution Neural Net and was used in the field of detection. On the other hand, sequencing data abnormality detection using generative adversarial network is a lack of research papers compared to image data. Of course, in Li et al (2018), a study by Li et al (LSTM), a type of recurrent neural network, has proposed a model to classify the abnormities of numerical sequence data, but it has not been used for categorical sequence data, as well as feature matching method applied by salans et al.(2016). So it suggests that there are a number of studies to be tried on in the ideal classification of sequence data through a generative adversarial Network. In order to learn the sequence data, the structure of the generative adversarial networks is composed of LSTM, and the 2 stacked-LSTM of the generator is composed of 32-dim hidden unit layers and 64-dim hidden unit layers. The LSTM of the discriminator consists of 64-dim hidden unit layer were used. In the process of deriving abnormal scores from existing paper of Anomaly Detection for Sequence data, entropy values of probability of actual data are used in the process of deriving abnormal scores. but in this paper, as mentioned earlier, abnormal scores have been derived by using feature matching techniques. In addition, the process of optimizing latent variables was designed with LSTM to improve model performance. The modified form of generative adversarial model was more accurate in all experiments than the autoencoder in terms of precision and was approximately 7% higher in accuracy. In terms of Robustness, Generative adversarial networks also performed better than autoencoder. Because generative adversarial networks can learn data distribution from real categorical sequence data, Unaffected by a single normal data. But autoencoder is not. Result of Robustness test showed that he accuracy of the autocoder was 92%, the accuracy of the hostile neural network was 96%, and in terms of sensitivity, the autocoder was 40% and the hostile neural network was 51%. In this paper, experiments have also been conducted to show how much performance changes due to differences in the optimization structure of potential variables. As a result, the level of 1% was improved in terms of sensitivity. These results suggest that it presented a new perspective on optimizing latent variable that were relatively insignificant.

• Korean Journal of Clinical Laboratory Science
• /
• v.51 no.3
• /
• pp.309-315
• /
• 2019

Cervical cancer is the fourth most common malignant neoplasm in women worldwide. Most cases of cervical cancer are caused by an infection by the human papillomavirus. Molecular diagnostic methods have emerged to detect the HPV for sensitivity, specificity, and objectivity. In particular, real-time PCR has been introduced to acquire a more sensitive target DNA or RNA. RNA extraction and complementary DNA synthesis are proceeded before performing real-time PCR targeting RNA. To identify an adequate and sensitive cDNA synthesis kit, this study evaluated the two commonly used kits for cDNA synthesis. The results show that the $R^2$ and efficiency (%) of the two cDNA synthesis kits were similar in the cervical cancer cell lines. On the other hand, the Takara kit compared to Invitrogen kit showed P<0.001 in the $10^2$ and $10^3$ SiHa cell count. The Takara kit compared to the Invitrogen kit showed P<0.001 in the $10^1$ and $10^2$ HeLa cell count. Furthermore, 8, 4, 2, 1, and 0.5 ml of forty exfoliated cell samples were used to compare the cDNA synthesis kits. The Takara kit compared to the Invitrogen kit showed P<0.01 in 8, 4, and 1 ml and P<0.05 in 0.5 mL. The study was performed to identify the most appropriate cDNA synthesis kit and suggests that a cDNA synthesis kit could affect the real-time PCR results.

• Journal of Intelligence and Information Systems
• /
• v.25 no.1
• /
• pp.43-61
• /
• 2019

Development of technologies in artificial intelligence has been rapidly increasing with the Fourth Industrial Revolution, and researches related to AI have been actively conducted in a variety of fields such as autonomous vehicles, natural language processing, and robotics. These researches have been focused on solving cognitive problems such as learning and problem solving related to human intelligence from the 1950s. The field of artificial intelligence has achieved more technological advance than ever, due to recent interest in technology and research on various algorithms. The knowledge-based system is a sub-domain of artificial intelligence, and it aims to enable artificial intelligence agents to make decisions by using machine-readable and processible knowledge constructed from complex and informal human knowledge and rules in various fields. A knowledge base is used to optimize information collection, organization, and retrieval, and recently it is used with statistical artificial intelligence such as machine learning. Recently, the purpose of the knowledge base is to express, publish, and share knowledge on the web by describing and connecting web resources such as pages and data. These knowledge bases are used for intelligent processing in various fields of artificial intelligence such as question answering system of the smart speaker. However, building a useful knowledge base is a time-consuming task and still requires a lot of effort of the experts. In recent years, many kinds of research and technologies of knowledge based artificial intelligence use DBpedia that is one of the biggest knowledge base aiming to extract structured content from the various information of Wikipedia. DBpedia contains various information extracted from Wikipedia such as a title, categories, and links, but the most useful knowledge is from infobox of Wikipedia that presents a summary of some unifying aspect created by users. These knowledge are created by the mapping rule between infobox structures and DBpedia ontology schema defined in DBpedia Extraction Framework. In this way, DBpedia can expect high reliability in terms of accuracy of knowledge by using the method of generating knowledge from semi-structured infobox data created by users. However, since only about 50% of all wiki pages contain infobox in Korean Wikipedia, DBpedia has limitations in term of knowledge scalability. This paper proposes a method to extract knowledge from text documents according to the ontology schema using machine learning. In order to demonstrate the appropriateness of this method, we explain a knowledge extraction model according to the DBpedia ontology schema by learning Wikipedia infoboxes. Our knowledge extraction model consists of three steps, document classification as ontology classes, proper sentence classification to extract triples, and value selection and transformation into RDF triple structure. The structure of Wikipedia infobox are defined as infobox templates that provide standardized information across related articles, and DBpedia ontology schema can be mapped these infobox templates. Based on these mapping relations, we classify the input document according to infobox categories which means ontology classes. After determining the classification of the input document, we classify the appropriate sentence according to attributes belonging to the classification. Finally, we extract knowledge from sentences that are classified as appropriate, and we convert knowledge into a form of triples. In order to train models, we generated training data set from Wikipedia dump using a method to add BIO tags to sentences, so we trained about 200 classes and about 2,500 relations for extracting knowledge. Furthermore, we evaluated comparative experiments of CRF and Bi-LSTM-CRF for the knowledge extraction process. Through this proposed process, it is possible to utilize structured knowledge by extracting knowledge according to the ontology schema from text documents. In addition, this methodology can significantly reduce the effort of the experts to construct instances according to the ontology schema.