• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.1-23
• /
• 2018

From the 21st century, various high-quality services have come up with the growth of the internet or 'Information and Communication Technologies'. Especially, the scale of E-commerce industry in which Amazon and E-bay are standing out is exploding in a large way. As E-commerce grows, Customers could get what they want to buy easily while comparing various products because more products have been registered at online shopping malls. However, a problem has arisen with the growth of E-commerce. As too many products have been registered, it has become difficult for customers to search what they really need in the flood of products. When customers search for desired products with a generalized keyword, too many products have come out as a result. On the contrary, few products have been searched if customers type in details of products because concrete product-attributes have been registered rarely. In this situation, recognizing texts in images automatically with a machine can be a solution. Because bulk of product details are written in catalogs as image format, most of product information are not searched with text inputs in the current text-based searching system. It means if information in images can be converted to text format, customers can search products with product-details, which make them shop more conveniently. There are various existing OCR(Optical Character Recognition) programs which can recognize texts in images. But existing OCR programs are hard to be applied to catalog because they have problems in recognizing texts in certain circumstances, like texts are not big enough or fonts are not consistent. Therefore, this research suggests the way to recognize keywords in catalog with the Deep Learning algorithm which is state of the art in image-recognition area from 2010s. Single Shot Multibox Detector(SSD), which is a credited model for object-detection performance, can be used with structures re-designed to take into account the difference of text from object. But there is an issue that SSD model needs a lot of labeled-train data to be trained, because of the characteristic of deep learning algorithms, that it should be trained by supervised-learning. To collect data, we can try labelling location and classification information to texts in catalog manually. But if data are collected manually, many problems would come up. Some keywords would be missed because human can make mistakes while labelling train data. And it becomes too time-consuming to collect train data considering the scale of data needed or costly if a lot of workers are hired to shorten the time. Furthermore, if some specific keywords are needed to be trained, searching images that have the words would be difficult, as well. To solve the data issue, this research developed a program which create train data automatically. This program can make images which have various keywords and pictures like catalog and save location-information of keywords at the same time. With this program, not only data can be collected efficiently, but also the performance of SSD model becomes better. The SSD model recorded 81.99% of recognition rate with 20,000 data created by the program. Moreover, this research had an efficiency test of SSD model according to data differences to analyze what feature of data exert influence upon the performance of recognizing texts in images. As a result, it is figured out that the number of labeled keywords, the addition of overlapped keyword label, the existence of keywords that is not labeled, the spaces among keywords and the differences of background images are related to the performance of SSD model. This test can lead performance improvement of SSD model or other text-recognizing machine based on deep learning algorithm with high-quality data. SSD model which is re-designed to recognize texts in images and the program developed for creating train data are expected to contribute to improvement of searching system in E-commerce. Suppliers can put less time to register keywords for products and customers can search products with product-details which is written on the catalog.

• Clinical and Experimental Pediatrics
• /
• v.49 no.9
• /
• pp.966-971
• /
• 2006

Purpose : 'Programming' describes the process that stimulus at a critical period of development has lifelong effects. The fact that low birth weight links to the risk of elevated blood pressures in adult life is well known. This study aims to examine whether this link is evident in the newborn by investigating the relationship of the intrauterine growth indices and neonatal blood pressure(BP). Methods : We studied 127 neonates who were born at Ewha Womans' Hospital and their mothers enrolled our cohort study during pregnancy. Data on the mothers and details of the birth records were tracked and collected from medical charts. Neonatal BP was measured within 24 hours after birth. Results : Neonatal SBP was positively correlated to intrauterine growth indices; birth weight(BW)(r=0.4), head circumference(HC)(r=0.4), and birth height(r=0.3). However, an inverse relationship existed, between HC/BW ratio and neonatal SBP(r=-0.4). After adjusting for the baby's sex, maternal BP, and gestational age, neonatal SBP still associated with intrauterine growth indices. SBP was 7 mmHg higher in the highest BW group(${\geq}90percentiles$) compared to the lowest group(<10 percentiles). On the other hand, SBP was 17 mmHg lower in the highest HC/BW group(${\geq}90percentiles$) compared in the lowest group(<10 percentiles). Conclusion : This study could not find the evidence that intrauterine growth retardation affect on elevated neonatal BP. It suggests that the initiating events of BP programming may occur during postnatal growth period. To identify the critical starting period that intrauterine growth retardation leads to elevated BP, a study tracking BP changes from birth to childhood is required.

• Journal of agricultural medicine and community health
• /
• v.39 no.3
• /
• pp.161-175
• /
• 2014

Objectives: This study investigates the prevalence of suicide ideation and its the related factors by gender among Korean adults. Methods: National cross-sectional data from the database of the first and second year of $5^{th}$ Korean National Health and Nutritional Examination Survey (KNHANES V-1, V-2) were used for this study. 13,165 adults (${\geq}20$ years) were included in this analysis. Results: The prevalence of suicide ideation was 9.8% for men, 18.9% for women. Renal failure, depression, melancholy, stress, perceived health status, and smoking were significant risk factors among both men and women for suicide ideation. Divorced, separated or widowed of marital status and severe physical activity were related factors of suicide ideation for men. Low educational attainment, and alcohol dependency were associated factors of suicide ideation for women. Conclusions: There was gender difference of prevalence and risk factors of suicide ideation. These gender differences should be considered for planning and implementing suicide prevention program.

• Journal of The Korean Association For Science Education
• /
• v.36 no.1
• /
• pp.45-61
• /
• 2016

The purpose of this study is to explore students' epistemic goals and considerations in designing an experiment task and to investigate how a shift in the students' epistemology affected their argumentation. Four 7th grade students were selected as a focus group. According to the results, when they designed their own experiment, their epistemic goal was 'scientific sense-making' and their epistemic considerations - the perception of the nature of the knowledge product was 'this experiment should explain how something happened', the perception of the justification was 'we need to use our interpretation of the data' and the perception of the audience was 'constructor' - contributed to designing their experiment actively. When students tried to select one argument, their epistemic goal shifted to 'winning a debate', showing 'my experiment is better than the others' with the perception of the audience, 'competitor'. Consequently, students only deprecated the limits of different experiment so that they did not explore the meaning of each experiment design deeply. Eventually, student A's experiment design was selected due to time restrictions. When they elaborated upon their result, their epistemic goal shifted to 'scientific sensemaking', reviewing 'how this experiment design is scientifically valid' through scientific justification - we need justification to make members accept it - acting as 'cooperator'. Consequently, all members engaged in a productive argumentation that led to the development of the group result. This study lays the foundation for future work on understanding students' epistemic goals and considerations to prompt productive argumentation in science classrooms.

• Journal of dental hygiene science
• /
• v.11 no.2
• /
• pp.77-83
• /
• 2011

The purpose of this study is to survey aspects of factors in stages of change and decisional balance on oral examination of adults in their 20s by applying transtheoretical model. Transtheoretical model is the one that was developed in order to explain a human being's behavioral change. The stages of change consist of precontemplation stage, contemplation stage, preparation stage, action stage, and maintenance stage. Decisional balance is utilized with comprising 2 elements such as pros and cons in transtheoretical model, and was mentioned that pros and cons have unique form every behavioral stage of change. As a result of applying the regular oral examination behavior based on transtheoretical model, it was indicated to be precontemplation stage with 79 people(22.1%), contemplation stage with 156 people(43.6%), preparation stage with 38 people(10.6%), action stage with 47 people(13.1%), and maintenance stage with 38 people(10.6%). Subjects, who maintain regular oral examination, were indicated to recognize pros highly in the decision balance. Subjects, who don't have intention of receiving oral examination, were indicated to have characteristic of recognizing cons highly.

• Journal of dental hygiene science
• /
• v.13 no.2
• /
• pp.165-173
• /
• 2013

The purpose of this study was to evaluate the surface change after 15% carbamide peroxide home bleaching to various restorative materials (composite resin [CR], resin modified glass ionomer [RMGI] and glass ionomer [GI]) and to observe the effect of surface condition of the materials on re-staining. Three esthetic restorative materials (Filtek Z250, 3M, USA; Fuji II LC, GC, Japan; Fuji II, GC, Japan) were used in this study. Twenty specimens per material group were made and divided into two groups (bleached and control). The specimens were immersed in coffee after applying bleaching agent. The color change and surface roughness were measured before and after bleaching and after immersion in coffee. The data were analyzed with SPSS 18.0. The results were as follows: 1. The color of all experiment groups was significantly changed after bleaching (p<0.05). RMGI was the greatest value of ${\Delta}E^*$ and ${\Delta}L^*$. GI and CR groups were in ordering (p<0.05). The ${\Delta}a^*$ value was decreased GI, RMGI and CR. RMGI was only significantly decreased in ${\Delta}b^*$ value (p<0.05). 2. The surface roughness before and after bleaching was significantly different on CR, RMGI and GI (p<0.05). 3. After staining with coffee, the value of ${\Delta}E^*$ was increased in GI, RMGI and CR, furthermore GI and RMGI showed significant difference in the bleaching groups (p<0.05). The ${\Delta}L^*$ value of GI and RMGI was significantly decreased. 4. The change of surface roughness after staining was not significantly different in all groups (p>0.05). The maintenance of color stability in esthetic restorations is one of the most important properties. Tooth whitening is for the aesthetic. Therefore, dental professionals should notice to patients about re-staining after tooth whitening. They should give an instruction that how to prevent and which kinds of agents could be stained.

• The korean journal of orthodontics
• /
• v.24 no.3 s.46
• /
• pp.569-586
• /
• 1994

The purpose of this study is to investigate the effect of loss of incisal function on the thickness, growth activities, ultrastructure of the condylar cartilage and on the muscle fibers of masseter superlicialis, anterior belly of digastric muscle in the growing rats. 37 day-old-rats of which incisors had been trimmed every day received soft diet from weaning and were studied by the autoradiography, electron microscopy and muscle histochemistry. The results obtained were as follows : The thickness of the fibrous, proliferative layer in superior, posterosuperior portion of the condylar cartilage was significantly(p<0.01) reduced in experimental groups and the decrease rate of fibrous layer thickness was greater in posterosuperior portion than in superior portion of cartilage and was greater than in proliferative layer. In normal group, more cells of posterosuperior portion moved more rapidly towards the medullary cavity. In experimental group, the labelling index of posterosuperior portion was decreased in proliferative layer at 2 hours, in transitional layer at 1, 2 days, in hypertrophic layer at 4 days after injection relative to posterosuperior portion of control group. But labelling index of superior portion was not different from that of control group at all time course after injection. From the muscle histochemistry, the diameter of type IIB fibers in masseter superficialis muscle, type IIA, type IIB fibers in anterior belly of digastric muscle decreased significantly(p<0.01) relative to controls in experimental group. From electron microscopic study, in the fibrous layer of the posterosuperior portion of condylar cartilage in normal group, many fibroblast like cells near the joint cavity showed extensive remodelling activities in ultrastructure. There was no morphological changes between experimental and control group in all cartilage cell layers of superior portion but cells near the joint cavity in fibrous layer of posterosuperior portion of experimental group showed morphologically inactive state relative to control group.

• Journal of the Korean Academy of Child and Adolescent Psychiatry
• /
• v.16 no.2
• /
• pp.219-230
• /
• 2005

Objectives : This study was designed to compare the demographic data, clinical characteristics, developmental delay, and psychological tests between childhood-onset and adolescent-onset schizophrenic in-patients. Methods Medical records of the 17 childhood-onset (very early onset) Schizophrenia and 16 adolescent-onset (early onset) Schizophrenia in-patients were reviewed. Sex, age, psychiatric past history, prodromal symptoms and period, subtype, co-morbid disease, developmental delay, prescribed drug and dosage, treatment response, intelligence quotient (IQ), and Rorschach test were evaluated. Results : The mean admission age of childhood-onset (very early onset) group and adolescent-onset (early onset) group were 12.69$({\pm}2.34)$ and 15.13$({\pm}1.04)$ years. The mean onset age of childhood-onset(very early onset) group and adolescent-onset (early onset) group were 10.79$({\pm}1.95)$ and 14.46$({\pm}0.82)$ years. The mean prodromal period of childhood-onset (very early onset) group and adolescent-onset (early onset) group were 15.94$({\pm}12.33)$ and 8.06$({\pm}6.10)$ month. The time to remission period of childhood-onset (very early onset) group and adolescent-onset (early onset) group were 50.58$({\pm}24.67)$ and 30.06$({\pm}18.04)$ days. Longer time to remission period in childhood-osnet (very early onset) group was associated with earlier age of onset. The mean of total IQ, performance IQ, verbal IQ were at an average level. Discussion : Childhood-onset (very early onset) group and adolescent-onset (early onset) group Schizophrenia had different clinical and psychological features including prodromal period, and IQ subtests.

• Pediatric Gastroenterology, Hepatology & Nutrition
• /
• v.7 no.2
• /
• pp.228-238
• /
• 2004

Purpose: Obesity is rapidly increasing in Korean children. Obesity is a risk factor for cardiovascular morbidity and is frequently associated with hypertension, diabetes mellitus and coronary artery disease. This study was designed to evaluate risk factors of the metabolic syndrome in obese children. Methods: From February 2000 to June 2004, eighty eight obese (body mass index ${\geq}95th$ percentile) children aged 4 to 15 years were included. We measured serum lipid levels (total cholesterol, triglyceride, HDL cholesterol, LDL cholesterol), fasting sugar levels and insulin levels. Insulin resistance was determined by homeostasis model assessment, fasting insulin/glucose ratio and quantitative insulin sensitivity check index. Results: Clustering of risk factors for the metabolic syndrome in obese children demonstrated that 60.2% had more than one risk factors. Hypertension (14.8%), hypertriglyceridemia (14.8%), HDL-hypocholesterolemia (14.8%), LDL-hypercholesterolemia (12.5%) and hyperinsulinemia (12.5%) were observed. As BMI increased, there was statistically significant increase in systolic blood pressure, insulin and insulin resistance values. Insulin resistance was correlated to systolic blood pressure, serum lipid and insulin levels. The more risk factors for the metabolic syndrome obese children had, the higher was their insulin resistance. Conclusion: The increase in insulin resistance and clustering of risk factors for the metabolic syndrome are already apparent in obese children. Monitoring these risk factors for the metabolic syndrome should become a part of routine medical care for obese children.

• Journal of Science Education
• /
• v.33 no.2
• /
• pp.336-345
• /
• 2009

The purpose of this study is to analysis ERF components patterns of causal questions generated during the observation of biological phenomenon. First, the system that shows pictures causing causal questions based on biological phenomenon (evoked picture system) was developed in a way of cognitive psychology. The ERF patterns of causal questions based on time-series brain processing was observed using MEG. The evoked picture system was developed by R&D method consisting of scientific education experts and researchers. Tasks were classified into animal (A), microbe (M), and plant (P) tasks according to biological species and into interaction (I), all (A), and part (P) based on the interaction between different species. According to the collaboration with MEG team in the hospital of Seoul National University, the paradigm of MEG task was developed. MEG data about the generation of scientific questions in 5 female graduate student were collected. For examining the unique characteristic of causal question, MEG ERF components were analyzed. As a result, total 100 pictures were produced by evoked picture and 4 ERF components, M1(100~130ms), M2(220~280ms), M3(320~390ms), M4(460~520ms). The present study could guide personalized teaching-learning method through the application and development of scientific question learning program.