• Journal of Intelligence and Information Systems
• /
• v.25 no.4
• /
• pp.141-154
• /
• 2019

Rapid growth of internet technology and social media is progressing. Data mining technology has evolved to enable unstructured document representations in a variety of applications. Sentiment analysis is an important technology that can distinguish poor or high-quality content through text data of products, and it has proliferated during text mining. Sentiment analysis mainly analyzes people's opinions in text data by assigning predefined data categories as positive and negative. This has been studied in various directions in terms of accuracy from simple rule-based to dictionary-based approaches using predefined labels. In fact, sentiment analysis is one of the most active researches in natural language processing and is widely studied in text mining. When real online reviews aren't available for others, it's not only easy to openly collect information, but it also affects your business. In marketing, real-world information from customers is gathered on websites, not surveys. Depending on whether the website's posts are positive or negative, the customer response is reflected in the sales and tries to identify the information. However, many reviews on a website are not always good, and difficult to identify. The earlier studies in this research area used the reviews data of the Amazon.com shopping mal, but the research data used in the recent studies uses the data for stock market trends, blogs, news articles, weather forecasts, IMDB, and facebook etc. However, the lack of accuracy is recognized because sentiment calculations are changed according to the subject, paragraph, sentiment lexicon direction, and sentence strength. This study aims to classify the polarity analysis of sentiment analysis into positive and negative categories and increase the prediction accuracy of the polarity analysis using the pretrained IMDB review data set. First, the text classification algorithm related to sentiment analysis adopts the popular machine learning algorithms such as NB (naive bayes), SVM (support vector machines), XGboost, RF (random forests), and Gradient Boost as comparative models. Second, deep learning has demonstrated discriminative features that can extract complex features of data. Representative algorithms are CNN (convolution neural networks), RNN (recurrent neural networks), LSTM (long-short term memory). CNN can be used similarly to BoW when processing a sentence in vector format, but does not consider sequential data attributes. RNN can handle well in order because it takes into account the time information of the data, but there is a long-term dependency on memory. To solve the problem of long-term dependence, LSTM is used. For the comparison, CNN and LSTM were chosen as simple deep learning models. In addition to classical machine learning algorithms, CNN, LSTM, and the integrated models were analyzed. Although there are many parameters for the algorithms, we examined the relationship between numerical value and precision to find the optimal combination. And, we tried to figure out how the models work well for sentiment analysis and how these models work. This study proposes integrated CNN and LSTM algorithms to extract the positive and negative features of text analysis. The reasons for mixing these two algorithms are as follows. CNN can extract features for the classification automatically by applying convolution layer and massively parallel processing. LSTM is not capable of highly parallel processing. Like faucets, the LSTM has input, output, and forget gates that can be moved and controlled at a desired time. These gates have the advantage of placing memory blocks on hidden nodes. The memory block of the LSTM may not store all the data, but it can solve the CNN's long-term dependency problem. Furthermore, when LSTM is used in CNN's pooling layer, it has an end-to-end structure, so that spatial and temporal features can be designed simultaneously. In combination with CNN-LSTM, 90.33% accuracy was measured. This is slower than CNN, but faster than LSTM. The presented model was more accurate than other models. In addition, each word embedding layer can be improved when training the kernel step by step. CNN-LSTM can improve the weakness of each model, and there is an advantage of improving the learning by layer using the end-to-end structure of LSTM. Based on these reasons, this study tries to enhance the classification accuracy of movie reviews using the integrated CNN-LSTM model.

• Journal of the Korean Society of School Health
• /
• v.9 no.1
• /
• pp.77-98
• /
• 1996

This studies designed to work out a theoretical framework on the experience of sexual abuse from the perspective of grounded theory in an effort to provide more practical and efficient nursing intervention for female victims. The subcategories identified were "sexual abuse", "threatening", "absent mindness", "embarrassment", "horripilation", "dizziness", "wondrousness", "filthiness", "sexual curiosity", "violence level", "victim's age", "neighbors response", "victims personality", "common experience", "sexual abuse information", "family relations", "level of familiarity", "hiding", "suppression", "self-torture", "self-protection", "avoidance", "asking aid", "withdrawal", "hatred", "confusion", "dodging, "remmant", and "pursuing". The 29 subcategories given above were further integrated into 16 categories such as "victimizedness", "being astounded", "filthiness", "degree", "developmental stage", "response pattern", "personality", "rarity", "information availability", "family support", "cover-up", "escaping", "informing", "negative internalization", and "positive pursuit of change". The core categories linked to all the other categories turned out to be "being taken aback" and "filthiness" incorporating the relevant subcategories. A total of 23 theoretical hypothesis emerged in the process of analyzing data. 1. the grater sexual curiosity, the weaker the senses of being taken aback and filthiness. 2. The weaker sexual curiosity, the stronger the senses of being taken aback and filthiness. 3. The stronger the level of violence, The more violent the senses of being taken aback and filthiness. 4. The lower the level of violence, the weaker the senses of being taken aback and filthiness. 5. The younger the victims, the stronger the senses of being taken aback and filthiness. 6. The older the victims, The weaker the senses of being taken aback and filthiness. 7. 'Escaping' will transpire regardless of the given circumstances. 8. The weaker the senses of being taken aback and filthiness, the more probable 'informing' and 'escaping' transpire. 9. The stronger the senses of being taken aback and filthiness, the more probable 'informing' and 'escaping' transpire. 10. The more protective the response from 'informing' and 'escaping' transpire around, the more likely the response to being taken aback' and 'filthiness' will be 'informing' and 'escaping'. 11. The more repelling the response from around, the more likely the response to 'being taken aback' and 'filthiness' will be 'covering-up' and 'escaping'. 12. The more open minded the personality of the subject, the more likely the response to 'being taken aback' and 'filthiness' will be 'informing' and 'escaping'. 13. The more closed the personality of tile subject, the more likely the response to 'being taken aback' and 'filthiness' will be 'covering-up' and 'escaping'. 14. The more frequent the experience of sexual abuse, the more likely the response to 'being taken aback' and 'filthiness' will be 'informing' and 'escaping'. 15. The less frequent the experience of sexual abuse, the more lilely the response to 'being taken aback' and 'filthiness' will be 'covering-up' and 'escaping'. 16. The more available information concerning sexual abuses, the more likely response to 'being taken aback' and 'filthiness' will be 'informing' and 'escaping. 17. The less available information concerning sexual abuses, the more likely the response to 'being taken aback' and 'filthiness' will be 'covering-up' and 'escaping'. 18. The more cohesive the family of the subject, the more likely the response to 'being taken aback' and 'filthiness' will be 'informing' and 'escaping'. 19. The less cohesive the family of the subject, the more likely the response to 'being taken aback' and 'filthiness' will be 'covering-up' and 'escaping'. 20. The less familiar the subject is with the abuser, the more likely the response to 'being taken aback' and 'filthiness' will be 'informing' and 'escaping'. 21. The less familiar the subject is with the abuser, the more likely the response to 'being taken aback' and 'filthiness' will be 'covering-up' and 'escaping. 22. The more likely the response to 'being taken aback' and 'filthiness' is 'informing and 'escaping', the more positive changes the subject will pursue. 23. The more likely the response to 'being taken aback' and 'filthiness' is 'covering-up' and 'escaping', the more negative changes the subject will pursue. The following four hypotheses were conformed in the process of data analysis. 1) In case the level of violence is strong but 'being taken aback' and 'filthiness' in weak because of strong sexual curiosity and also if information concerning sexual abuse is not readily available and the frequency is low, negative internationalization marked by 'covering-up' and 'escaping' will take place despite the fact the subject is open-minded, the family is cohesive and the abuser is unfamiliar. 2) In case the level of violence is weak but 'being taken aback' and 'filthiness' is weak combined with weak sexual curiosity and also if information concerning sexual abuse is readily available and the response from around is protective and the frequency is high, the subject will pursue positive changes to 'being taken aback' and 'filthiness', further aided by the fact that the subject is open-minded, the family is cohesive and the abuser is unfamiliar. 3) In case the level of violence is strong and 'being taken abuse' and 'filthiness' is strong because of weak sexual curiosity and also if information concerning sexual abuse is reading available and the response from around is readily available and the response from around is protective and the frequency is low, the subject will persue positive changes marked by 'informing' and 'escaping' despite the fact that the family cohesion is weak and the abuser is familiar. 4) In case the level of violence is strong and 'being taken aback' and 'filthiness' is strong because of weak sexual curiosity and also if information concerning sexual abuse is not readily available and the response from around is respelling and the frequency is low negative internalization like 'covering-up' and 'escaping' will take place, further aggravated by the fact that the subject's personality is closed, family cohesion is weak, and subject is familiar. On the basis of the above finding, it is recommended that nursing intervention should focus on promoting the milieu conductive to the victims pursuing positive changes along with the adequate aids from protection facilities as well as from the people around them.

• Korean Journal of Heritage: History & Science
• /
• v.56 no.4
• /
• pp.230-246
• /
• 2023

The wooden fence(木柵), which began to appear in the Bronze Age and is presumed to be the oldest defense facility in human history, was used as a fortress for the purpose of further strengthening military defense functions until after the Japanese Invasion of Korea in 1592 in the Joseon Dynasty(壬辰倭亂). As it was established as the concept of a fortress or a fence installed outside a fence castle(城柵) or barracks fence(營柵), its importance as an essential facility for defense was further highlighted. This study is the result of exploring wooden fence that were used as official facilities during the Joseon Dynasty, focusing on literature surveys such as 『Annals of the Joseon Dynasty』 and 『New Jeungdonggukyeojiseungram』 In this study, in particular, the conclusion of this study is as follows, focusing on the use and function of Nokgakseong(鹿角城), underwater wooden fence, installation methods, and materials of wooden fences, is as follows. The conclusions of this study, which focused on the materials of the wooden fence, are as follows. First, as invasions by foreign enemies became more frequent in the late Goryeo and early Joseon Dynasty, wooden fences played a major role as a major out-of-castle defense facility((防禦施設). In addition, wooden fences were modified and installed into various types such as wooden fences(木柵城), Nokgakseong, a fence made up of large branches in the shape of a deer antler, and underwater wooden fences(水中木柵) according to the circumstances of the times, government policy, and location environment. Second, wooden fences were installed in strategic locations in defense facilities for military purposes, such as mountain fortress(山城), fortresses(營), camps(鎭), forts(堡), and castles(邑城) in strategic locations, and were used for defense in case of emergency. According to the urgency of farming, it was installed in accordance with the non-farming season, when it is easy to mobilize manpower to avoid the busy farming season. The size of the wooden fence of the Joseon Dynasty, which are confirmed through literature records, was converted into Pobaekchuk(布帛尺), and the circumference was very diverse from 4,428chuk(2,066m) to 55chuk(25m). Third, Nokgakseong is an efficient combat support facility that is more aggressive than a general wooden fence, and the records of Nokgakseong in the Annals of the Joseon Dynasty appeared during the King Sejong period the record was 20 times, the most. By region, it was found that it was mainly installed in coastal rugged areas such as Pyeongan and Hamgildo(12), which are the 6-jin areas of the 4th Army. Fourth, in the early 15th century, as the royal court established a maritime defense strategy for the coastal area of the southern coast, after the Sampo Invasion(三浦倭亂), riots by Japanese settlers in Sampo in 1510, major military posts including eupseong(邑城), camps, and forts were established. The installation of underwater barriers around various government facilities rapidly increased as a defense facility to block the warships of Japanese pirates around various government facilities. Fifth, between the 15th and 17th centuries before and after the Japanese Invasion of Korea in Sampo, underwater fences were installed in the Southern coast and Ganghwa Island. In particular, in the 15th century, underwater fences were intensively installed in coastal areas of Gyeongsangnam-do, such as Jepo. Pine trees and Oaks are the main materials used for underwater fences, but other materials such as Oldham's meliosma, Loose-flower hornbeam and The vines of arrowroots were also used as materials for wooden fences.

• Journal of agricultural medicine and community health
• /
• v.30 no.2
• /
• pp.227-240
• /
• 2005

Health promotion has more comprehensive approaches in recent years. Nevertheless we accept the concept of health promotion differently, we are agree on that community is the most important field in health promotion which includes population at the aspect of health policy, individual skill and, environment. And there are a number of different approaches to health promotion. In them, 'population approaches' and 'high -risk group approaches' has the most different characteristics. 'Population approaches' is equally important or more important than 'individual approaches' for maintaining and promoting population health. Almost part of this article contents is the summary of the guideline and population strategy of health promotion in Korea, 1999 - 2005. Community based health promotion program should be reinforced, integrated, comprehensive, collaborative through efficiently utilizing community resources. Recent new orientation of community health program is integrated health program, we can find this orientation at Ottawa charter 1986. Comprehensive approaches with the determinant factors for health are essential task. Physical activity is a key health determinant. The population-health approach suggests that educating people about physical activity is not enough. Individual behavior changes are important too, but need to be balanced with strategies for environmental change. Population strategy with physical activity for health promotion should be developed through improving social and physical supportive environment, linking and integrating community resources between public and private sectors in national, regional and local level. Continuous public education and social marketing should be provided through collaborating with community physical activity organization, facilities, work-places and school for increasing concern of all the people of community about physical activity. Governments, agencies and citizens should held and participate to building movement. And the strategy that various 'active for life' program should be developed, delivered, maintained and reinforced continuously. Basically, adequate and sufficient financing, developing human resources, policies and legislation would be provided and supported fully too. At last, research development and knowledge exchange are required domestically and internationally. In Korea, we had classified the category of strategic priority of physical activity programs by environmental support, life-course approach, high-risk group approach and disease group approach for physical activity program based on community health center. Community based core programs for physical activity that includes infrastructure building and establishment of supporting environment, community campaign, health promotion education and public service announcement, physical activity programs for elderly and obesity, exercise prescription program.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.43 no.3
• /
• pp.77-91
• /
• 2015

This study aims to suggest adequate soil management through the analysis of physicochemical properties of soil in the planting grounds of Incheon International Airport, which was constructed on a massive land reclamation site. Study areas were 5 sites at the international business complex, the passenger terminal, the airport support complex, the free trade zone, and the access road. Soil profile analysis showed that 9 plots out of the 27 plots were hardpan and heterospere within 80cm from the soil surface. The earth laid on the ground was categorized as gravel based soil(4 plots), dredged soil from the sea bottom and mixed reclamation materials(2 plots), clay with poor permeability(3 plots) and waste construction material(1 plot). Average soil hardness was $11.5kg/cm^2$ and soil textures were sandy soil, sandy loam and loamy sand. Average soil pH was 6.7 and average organic matter content was 0.7%. Electrical conductivity was 0.0dS/m and exchangeable cation concentrations were $Ca^{2+}$ 3.4cmol/kg, $Mg^{2+}$ 1.5cmol/kg, $K^+$ 0.3cmol/kg and $Na^+$ 1.0cmol/kg. Average cation exchange capacity was 11.0cmol/kg. Although average figures in Solum mostly meet the landscape design criteria, properties of each soil layer showed various values sometimes over the limit. Base saturations were $Ca^{2+}$ 29.9%, $Mg^{2+}$ 13.3% and $K^+$ 3.7% for lower soil, $Ca^{2+}$ 33.3%, $Mg^{2+}$ 17.0% and $K^+$ 2.7% for mid-soil and $Ca^{2+}$ 32.6%, $Mg^{2+}$ 12.2% and $K^+$ 1.9% for upper soil. Exchangeable sodium percentages were 16.4% for lower soil, 7.5% for mid-soil and 4.7% upper soil. Sodium adsorption rates were 0.8 for lower soil, 0.3 for mid-soil and 0.2 for upper soil. Factors affecting to the vegetation growth were heterogeneity and poorness of solum, disturbance of dredged soils, high soil hardness including hardpan in the subsurface soil layer and shallow effective soil depth, high soil acidity, imbalance of base contents, low organic matter content and low available phosphate levels in the soil.

• Korean Journal of Environment and Ecology
• /
• v.28 no.3
• /
• pp.342-356
• /
• 2014

This study aims to provide the fundamental data for restoration through the analysis of vegetation structure characteristic for managed area with deforest and fumigation among the places damaged by oak wilt disease in Bukhansan National Park. Study areas were selected in six damaged areas with 18 plots ($20m{\times}20m$) including comparing plots between year 2006 and year 2011 after preliminary study finding the exact location with easy access. Analysis items are density and coverage of under story layer, Importance Percentage of under story layer and Importance Percentage of urban adapted species and alien species. As a result, density and coverage under the canopy layer's vegetation was higher than comparing plot in Musugol valley and Soquicheon valley which 4 years and more passed. And Wondobong, Darakwon, Doseonsa and Dobong valley which had not been passed 3 years after deforestation and fumigation decreased in those of vegetation. However, the plots in Wondobong and Doseonsa was analyzed to be higher in density of shrubs than comparing plot. In terms of distirbution of species in under canopy layer which is grown as a upper vegetation, Quercus serrata and Sorbus alnifolia dominated each other in under canopy layer except for Wondobong and Darakwon. Wondobong continuously maintained Quercus mongolica and Q. mongolica and Pinus densiflora dominated in under canopy layer of Darakwon. S. alnifolia which is Urban adapted species were distributed in most study areas. In terms of alien species, Magnolia obovata and Robinia pseudoacacia were newly intruded in the opened crown, and Pueraria thunbergiana was distributed in Soquicheon valley. As the result of soil analysis, acidification(pH 4.49) and low level of organic matter(3.23%), available phosphate(1.40mg/kg) and exchangeable cation were analyzed and it was considered that those environment poorly support the vegetation growth.

• Korean Journal of Child Education & Care
• /
• v.17 no.4
• /
• pp.1-30
• /
• 2017

The purpose of this study is to investigate the awareness of daycare teachers by the nursery school operation performance of the National Competency Standards(NCS). For this purpose, 35 public and private daycare center, home daycare centers and corporation daycare center teachers were randomly sampled in Busan Metropolitan City. We conducted a questionnaire consisting of 230 sub-elements based on 45 sub-competency unit elements extracted from 12 areas of job ability of the national competence standards. Final questionnaires were collected from 117 questionnaires of 28 national daycare teachers, 30 private daycare teachers, 30 home daycare teachers, and 29 corporation daycare teachers. As a result, there was a difference in the awareness of importance among the four groups in all 12 performance capability units. For the 12 job performance units, the importance was recognized in the order of corporate daycare center, national daycare center, home daycare center, and private daycare center. Among them, "Establishment of Child Care Center Management Policy", "Construction of Child Care Service", "Child Care Activity Management", "Infant and Child Playing Guide", "Body and Art Activity Guidance", "Language, Nutritional guidance" there were no differences in the awareness of the importance of corporate daycare teachers and national daycare teachers, and there was no difference between national daycare teachers and home daycare teachers. However, there was a difference between corporate daycare teachers and home daycare teachers.. There is also a difference in the awareness of importance between home daycare teachers and private daycare teachers. In the remaining five job competence units, "Child Care Assessment", "Support for Infant and Toddler Development", "Cooperation with the Home and Community", "Child Care Management", and "Child Care Research", corporation daycare centers, national daycare centers. There was no statistically significant difference in the awareness of importance, but there was a significant difference from the private daycare teachers. Also this tendency was consistently observed in 45 sub-capacity unit elements.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.19 no.4
• /
• pp.52-62
• /
• 2016

Smart City is urban development for complex problem solving that provides convenience and safety for citizens, and it is a blueprint for future cities. In 2008, the Korean government defined the construction, management, and government support of U-Cities in the legislation, Act on the Construction, Etc. of Ubiquitous Cities (Ubiquitous City Act), which included definitions of terms used in the act. In addition, the Minister of Land, Infrastructure and Transport has established a "ubiquitous city master plan" considering this legislation. The concept of U-Cities is complex, due to the mix of informatization and urban planning. Because of this complexity, the foundation of relevant regulations is inadequate, which is impeding the establishment and implementation of practical plans. Smart City intelligent service facilities are not easy to define and classify, because technology is rapidly changing and includes various devices for gathering and expressing information. The purpose of this study is to complement the legal definition of the intelligent service facility, which is necessary for integrated management and operation. The related laws and regulations on U-City were analyzed using text-mining techniques to identify insufficient legal definitions of intelligent service facilities. Using data gathered from interviews with officials responsible for constructing U-Cities, this study identified problems generated by implementing intelligent service facilities at the field level. This strategy should contribute to improved efficiency management, the foundation for building integrated utilization between departments. Efficiencies include providing a clear concept for establishing five-year renewable plans for U-Cities.

• Journal of dental hygiene science
• /
• v.8 no.2
• /
• pp.65-71
• /
• 2008

The purpose of this study was to examine people's awareness of health insurance in a bid to help improve the management of dental health insurance coverage. The subjects in this study were 1,036 people who included experts in that field and medical consumers. The findings of the study were as follows: 1. In regard to the demographic characteristics of the expert group including gender and age, the female experts outnumbered the males, as the former accounted for 84.7 percent. And the experts in their 20s made up the largest age group, followed by the 30-39 age group and those in their 40s. As to the demographic characteristics of the medical consumers, the rates of the men and women stood at 49.8 percent and 50.2 percent respectively, which were similar. By age, the largest number of the medical consumers were in their 20s, followed by in their 30s and in their 40s. 2. Concerning opinions on the procurement of finance for health insurance, many respondents in both groups agreed that the health insurance coverage should be enlarged by securing more finance, and that items involved in the coverage should be prioritized again. Regarding a means of securing financial resources, the experts placed the most emphasis on extended budgetary support from the government, followed by establishing a social security system and budget compilation by local governments. The medical consumers attached the most importance to extended budgetary backing from the government, followed by budget compilation by local governments and determining contribution according to income level. 3. As for general view of the dental health insurance, the experts believed that the amounts of copayment and self-payment by the insured were appropriate(2.47), and they found the number of benefitted items to be proper(2.29) as well. The medical consumers thought that the benefit percentage of the dental health insurance in the overall health insurance budget was appropriate(2.26), and they also considered the number of benefitted items adequate(2.16). But this group didn't give a lot of marks to the appropriateness of those things. 4. As to perception about the dental health insurance, both groups took a similar view of it. Specifically, they felt the strong need for a dental health insurance system. In terms of satisfaction level, both groups were dissatisfied with it, and there was a strong tendency for them to be discontented with the health insurance coverage.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.16 no.1
• /
• pp.79-89
• /
• 2005

Objectives : To acquire an improved understanding of oppositional defiant disorder, we evaluated the characteristics of children who have the symptoms of ODD in community sample. Methods : 1200 children from an elementary school in Bucheon (an urban community near Seoul) were recruited by randomized sampling method. By Disruptive Behavior Disorder Scale according to DSM-III-R & DSM-IV, we evaluated the symptoms of ODD and selected subjects with ODD. Psychiatric comorbidity, character trait were compared in subjects with ODD and comparison group. Also we examined the association between prenatal/perinatal risk factors, family functions and the symptoms of ODD. Data were analyzed by appropriate statistical method using SPSS 11.5 window version. Result : Children with oppositional defiant disorder were revealed to have significantly higher rates of psychiatric comorbidity and significantly greater family dysfunction compared to comparison group. Among the prenatal/perinatal risk factors, severe emotional stress during pregnancy, postpartum depression, medication during pregnancy were revealed as risk factors of ODD. In character inventory, ODD group were evaluated to have high score in novelty seeking, harm avoidance, but low in reward dependency. Conclusion : These results support that 1) prenatal/perinatal and psycho-social risk factors could be a important role in the progression of ODD, and 2) children with ODD have diverse comorbid psychiatric symptoms.