• Journal of Intelligence and Information Systems
• /
• v.24 no.3
• /
• pp.1-19
• /
• 2018

Large amount of data is now available for research and business sectors to extract knowledge from it. This data can be in the form of unstructured data such as audio, text, and image data and can be analyzed by deep learning methodology. Deep learning is now widely used for various estimation, classification, and prediction problems. Especially, fashion business adopts deep learning techniques for apparel recognition, apparel search and retrieval engine, and automatic product recommendation. The core model of these applications is the image classification using Convolutional Neural Networks (CNN). CNN is made up of neurons which learn parameters such as weights while inputs come through and reach outputs. CNN has layer structure which is best suited for image classification as it is comprised of convolutional layer for generating feature maps, pooling layer for reducing the dimensionality of feature maps, and fully-connected layer for classifying the extracted features. However, most of the classification models have been trained using online product image, which is taken under controlled situation such as apparel image itself or professional model wearing apparel. This image may not be an effective way to train the classification model considering the situation when one might want to classify street fashion image or walking image, which is taken in uncontrolled situation and involves people's movement and unexpected pose. Therefore, we propose to train the model with runway apparel image dataset which captures mobility. This will allow the classification model to be trained with far more variable data and enhance the adaptation with diverse query image. To achieve both convergence and generalization of the model, we apply Transfer Learning on our training network. As Transfer Learning in CNN is composed of pre-training and fine-tuning stages, we divide the training step into two. First, we pre-train our architecture with large-scale dataset, ImageNet dataset, which consists of 1.2 million images with 1000 categories including animals, plants, activities, materials, instrumentations, scenes, and foods. We use GoogLeNet for our main architecture as it has achieved great accuracy with efficiency in ImageNet Large Scale Visual Recognition Challenge (ILSVRC). Second, we fine-tune the network with our own runway image dataset. For the runway image dataset, we could not find any previously and publicly made dataset, so we collect the dataset from Google Image Search attaining 2426 images of 32 major fashion brands including Anna Molinari, Balenciaga, Balmain, Brioni, Burberry, Celine, Chanel, Chloe, Christian Dior, Cividini, Dolce and Gabbana, Emilio Pucci, Ermenegildo, Fendi, Giuliana Teso, Gucci, Issey Miyake, Kenzo, Leonard, Louis Vuitton, Marc Jacobs, Marni, Max Mara, Missoni, Moschino, Ralph Lauren, Roberto Cavalli, Sonia Rykiel, Stella McCartney, Valentino, Versace, and Yve Saint Laurent. We perform 10-folded experiments to consider the random generation of training data, and our proposed model has achieved accuracy of 67.2% on final test. Our research suggests several advantages over previous related studies as to our best knowledge, there haven't been any previous studies which trained the network for apparel image classification based on runway image dataset. We suggest the idea of training model with image capturing all the possible postures, which is denoted as mobility, by using our own runway apparel image dataset. Moreover, by applying Transfer Learning and using checkpoint and parameters provided by Tensorflow Slim, we could save time spent on training the classification model as taking 6 minutes per experiment to train the classifier. This model can be used in many business applications where the query image can be runway image, product image, or street fashion image. To be specific, runway query image can be used for mobile application service during fashion week to facilitate brand search, street style query image can be classified during fashion editorial task to classify and label the brand or style, and website query image can be processed by e-commerce multi-complex service providing item information or recommending similar item.

• Journal of Nutrition and Health
• /
• v.42 no.8
• /
• pp.770-783
• /
• 2009

The school foodservice was quantitatively extended by policy of government all the while. There was carried out the survey of customer satisfaction about school foodservice by the ministry of education, science, and technology since 2006 years. Therefore, the purpose of this study was to grasp an improvement of the scores of school foodservice' quality attributes and satisfaction as compared with the preceding year by respondents and school type (elementary school, middle school, and high school). An annual survey was practiced to respondents (students, parents, and faculty) on september 2007 years and 2008 years in 16 cities and provinces. The statistics was analyzed to descriptive analysis and t-test for SPSS 12.0. The scores of school foodservice' quality attributes and overall customer satisfaction were almost increased to students, parents, and faculty and especially, big elevation in middle school. There was big increased the quality attributes such as 'providing information on foodservice', 'pleasant foodservice environment', 'kindness offered by employee' in elementary school, middle school, and high school to total respondents. An overall satisfaction in school foodservice was improved from 69.2 score to 71.9 score. On students, scores of overall satisfaction was increased from 72.9 to 74.0 as students of elementary school and from 61.5 to 65.8 as students of middle school (p < .001). Therefore, for improvement and development of school foodservice, there should be a necessary for an operator of school foodservice and an office of education to make an effort.

• Journal of radiological science and technology
• /
• v.26 no.2
• /
• pp.37-42
• /
• 2003

Purpose : Computed tomographic equipment is essential for diagnosis by means of radiation. With passage of time and development of science computed tomographic was developed time and again and in future examination by means of this equipment is expected to increase. In this connection these authors measured rate of scatter ray generation at front of lead glass for patients within control room of computed tomographic equipment room and outside of entrance door for exit and entrance of patients and attempted to ind out method for minimizing exposure to scatter ray. Material and Method : From November 2001 twenty five units of computed tomographic equipments which were already installed and operation by 13 general hospitals and university hospitals in Seoul were subjected to this study. As condition of photographing those recommended by manufacturer for measuring exposure to sauter ray was use. At the time objects used DALI CT Radiation Dose Test Phantom fot Head (${\oint}16\;cm$ Plexglas) and Phantom for Stomache(${\oint}32\;cm$ Plexglas) were used. For measurement of scatter ray Reader (Radiation Monitor Controller Model 2026) and G-M Survey were used to Survey Meter of Radical Corporation, model $20{\times}5-1800$, Electrometer/Ion Chamber, S/N 21740. Spots for measurement of scatter ray included front of lead glass for patients within control room of computed tomographic equipment room which is place where most of work by gradiographic personnel are carried out and is outside of entrance door for exit and entrance of patients and their guardians and at spot 100 cm off from isocenter at the time of scanning the object. The results : Work environment within computed tomography room which was installed and under operation by each hospital showed considerable difference depending on circumstances of pertinent hospitals and status of scatter ray was as follows. 1) From isocenter of computed tomographic equipment to lead glass for patients within control room average distance was 377 cm. At that time scatter ray showed diverse distribution from spot where no presence was detected to spot where about 100 mR/week was detected. But it met requirement of weekly tolerance $2.58{\times}10^{-5}\;C/kg$(100 mR/week). 2) From isocenter of computed tomographic equipment to outside of entrance door where patients and their guardians exit and enter was 439 cm in average, At that time scatter ray showed diverse distribution from spot where almost no presence was detected to spot with different level but in most of cases it satisfied requirement of weekly tolerance of $2.58{\times}10^{-6}\;C/kg$(100 mR/week). 3) At the time of scanning object amount of scatter ray at spot with 100 cm distance from isocenter showed considerable difference depending on equipments. Conclusion : Use of computed tomographic equipment as one for generation of radiation for diagnosis is increasing daily. Compared to other general X-ray photographing field of diagnosis is very high but there is a high possibility of exposure to radiation and scatter ray. To be free from scatter ray at computed tomographic equipment room even by slight degree it is essential to secure sufficient space and more effort should be exerted for development of variety of skills to enable maximum photographic image at minimum cost.

• Journal of agricultural medicine and community health
• /
• v.24 no.2
• /
• pp.269-289
• /
• 1999

This study was conducted to examine the relationship among health concern, health practice and ADL of elderly staying at home in a rural community and their affecting factors. Data were collected through direct interviews made with 480 old people aged more than sixty-five from November 15, 1998 to December 20, 1998. Out of 189 male and 291 female, the high-level group that showed high health concern accounted for 44.4%, the medium-level group for 13.1%, and the low-level group for 42.5%, in the health practice, the high-level group accounted for 3.8%, the medium-level group for 18.8%, and the low-level group for 77.5%. In the self-rated health status, the high-level group accounted for 29.0%, the medium-level group for 31.0%, and the low-level group for 40.0%, and in the ADL, the high ADL group accounted for 91.5%, and the low-level ADL group for 8.5%. The result of the chi-square test showed that for male, there was a significant relation between the health concern and the health practice index score. In the relation between the health practice index score and the self-rated health status, there was significant positive relationship between health practice index and self-rated health status, and in the relation between the health practice Index score and the ADL, old people with higher health practices showed good ADL(but not significant). Old people with good ADL also showed good self-rated health status. In the multiple regression analysis where the health practice was used as a dependent variable, the health concern was added to the sociodemographic variables as an independent variables, a formula was formed for male old people only and ones with high concern in health showed good health practice. In the multiple logistic regression analysis where the sociodemographic variables to which the health practices was added were used as an independent variable and the ADL as a dependent variable, the ADL appeared to be not good if for male old people the living costs were born by their sons and daughters and as for female old people their ages increased, but it was good if old people had sources of health information such as hospitals or health centers. The self-rated health status was worse, for male old people, if they had short living costs or diseases and for female old people, if they had spouses, living costs born by their sons and daughters or diseases, but it was better, for male old people, if they had periodical gatherings or carried out health practices a lot, and for female old people, if they had sources of health information such as hospitals or health centers or carried out health practices a lot. In view of the results stated above, the higher the old people had health concern, the more they carried out health practices, and the more they carried out health practices, the better they had ADL and self-rated health status that served as the level of health. Further, the better ADL, the better self-rated health status.

• Pediatric Gastroenterology, Hepatology & Nutrition
• /
• v.7 no.1
• /
• pp.40-47
• /
• 2004

Purpose: A new classification of gastrointestinal food allergy was published, but the changes of terminology between previously reported terms and the new ones were in a state of disorder. This has resulted in confusion between medical communication and diagnostic and therapeutic approaches. The clinical observations of infants presenting with gastrointestinal cow milk allergy (GI-CMA) were performed, and the changes in the terminology reviewed through the published Korean literature. Methods: Between March 2003 and July 2003, data from 37 consecutive infants with GI-CMA, aged 2 weeks to 15 months, were reviewed. The challenge and elimination test of cow milk, and the endoscopic and histologic findings, were used for the seven subdivisions of GI-CMA according to a new classification on the basis of patients' ages, clinical manifestations and location of gastrointestinal lesions. Results: The 37 patients had a mean age of $5.4{\pm}4.8$ months, with those observed in 26 (70.3%) of patients being below 6 months of age. The seven final diagnoses were; cow milk protein-induced enterocolitis (CMPIE) in 12 (32.4%), cow milk protein proctitis (PROC) in 12 (32.4%), IgE-mediated (IGE) in 6 (16.2%), gastroesophageal reflux-associated cow milk allergy (GERA) in 5 (13.5%) and eosinophilic gastroenterocolitis in 2 (5.4%). CMPIE was revealed as the typical type in 7 (18.9%) and the atypical type in 5 (13.5%), and all of typical CMPIE revealed cow milk protein-induced enteropathy. The mean age at symptom onset was $4.3{\pm}0.8$ months, and for those with typical and atypical CMPIE, and PROC and GERA were $3.8{\pm}4.6$, $10.4{\pm}3.8$, $3.4{\pm}3.9$ and $7.8{\pm}5.7$ months, respectively (p<0.05). The period from onset of symptom to diagnosis was $2.4{\pm}3.3$ (0.5~12) months, with those observed in atypical CMPIE and GERA being over 3months. Although the birth weights in all patients were within the 10~90 percentile range, the body weights on diagnoses were below the 3 percentile in 48.6%; IGE 16.7%, EOS 0%, typical CMPIE 85.7%, atypical CMPIE 60.0%, PROC 25.0% and GERA 100% (p<0.05). Through the review of the Korean literature, 8 case reports and 14 original articles for GI-CMA were found. Conclusion: GI-CMA is not a rare clinical disorder and is subdivided into seven categories on the basis of the patient's age, clinical manifestations and location of the gastrointestinal lesions. The terms for GI-CMA are changing with new classifications, and careful approaches are necessary for medical communications.

• The Korean Society of Law and Medicine
• /
• v.17 no.1
• /
• pp.299-346
• /
• 2016

There were also various decisions made in medical area in 2015. In the case that an inmate in a sanatorium was injured due to the reason which can be attributable to the sanatorium and the social welfare foundation that operates the sanatorium request treatment of the patient, the court set the standard of fixation of a party in medical contract. In the case that the family of the patient who was declared brain dead required withdrawal of meaningless life sustaining treatment but the hospital rejected and continued the treatment, the court made a decision regarding chargeable fee for such treatment. When it comes to the eye brightening operation which received measure of suspension from the Ministry of Health and Welfare for the first time in February, 2011, because of uncertainty of its safety, the court did not accept the illegality of such operation itself, however, ordered compensation of the whole damage based on the violation of liability for explanation, which is the omission of explanation about the fact that the cost-effectiveness is not sure as it is still in clinical test stage. There were numerous cases that courts actively acknowledged malpractices; in the cases of paresis syndrome after back surgery, quite a few malpractices during the surgery were acknowledged by the court and in the case of nosocomial infection, hospital's negligence to cause such nosocomial infection was acknowledged by the court. There was a decision which acknowledged malpractice by distinguishing the duty of installation of emergency equipment according to the Emergency Medical Service Act and duty of emergency measure in emergency situations, and a decision which acknowledged negligence of a hospital if the hospital did not take appropriate measures, although it was a very rare disease. In connection with the scope of compensation for damage, there were decisions which comply with substantive truth such as; a court applied different labor ability loss rate as the labor ability loss rate decreased after result of reappraisal of physical ability in appeal compared to the one in the first trial, and a court acknowledged lower labor ability loss rate than the result of appraisal of physical ability considering the condition of a patient, etc. In the event of any damage caused by malpractice, in regard to whether there is a limitation on liability in fee charge after such medical malpractice, the court rejected the hospital's claim for setoff saying that if the hospital only continued treatments to cure the patient or prevent aggravation of disease, the hospital cannot charge Medical bills to the patient. In regard to the provision of the Medical Law that prohibit medical advertisement which was not reviewed preliminarily and punish the violation of such, a decision of unconstitutionality was made as it is a precensorship by an administrative agency as the deliberative bodies such as Korean Medical Association, etc. cannot be denied to be considered as administrative bodies. When it comes to the issue whether PRP treatment, which is commonly performed clinically, should be considered as legally determined uninsured treatment, the court made it clear that legally determined uninsured treatment should not be decided by theoretical possibility or actual implementation but should be acknowledged its medical safety and effectiveness and included in medical care or legally determined uninsured treatment. Moreover, court acknowledged the illegality of investigation method or process in the administrative litigation regarding evaluation of suitability of sanatorium, however, denied the compensation liability or restitution of unjust enrichment of the Health Insurance Review & Assessment Service and the National Health Insurance Corporation as the evaluation agents did not cause such violation intentionally or negligently. We hope there will be more decisions which are closer to substantive truth through clear legal principles in respect of variously arisen issues in the future.

• Korean Journal of Weed Science
• /
• v.6 no.2
• /
• pp.174-190
• /
• 1986

A series of laboratory and greenhouse experiments were conducted to find out the possibility of tissue culture and cell culture methods as a tool to study herbicidal behaviour and herbicide tolerance screening from 1985 to 1986 at the Yeongnam Crop Experiment Station. For dehulled-rice culture, pure agar medium was the most appropriate in rice growth campared to other media used for plant tissue culture method. All the media but the pure agar medium resulted in growth retardance by approximately 50% and this effect was more pronounced to root growth than shoot growth. Herbicidal phytotoxicity was enhanced under light condition for butachlor, 2.4-D, and propanil while this effect was reversed for DPX F-5384 and CGA 142464, respectively. And also, herbicides of butachlor, chlornitrofen, oxadiazon, and BAS-514 resulted in more phytotoxic effect when shoot and root of rice were exposed to herbicide than root exposure only while other used herbicides exhibited no significant difference between two exposure regimes. Similar response was obtained from Echinochloa crusgalli even though the degree of growth retardance was much greater. Particularly, butachlor, 2.4-D, chlornitrofen, oxadiaxon, pyrazolate and BAS-514 totally inhibited chlorophyll biosynthesis even at the single contact of root. Apparent cultivar differences to herbicide were observed at the young seedling culture method and dehulled rice cultivars were more tolerant in DPX F-5384, NC-311, pyrazolate and pyrazoxyfen, respectively. For derant than other types or rice cultivar in butachlor, pretilachlor, perfluidone and oxadiazon while Tongil-type rice cultivars were more tolerant in DPXF-5384, NC-311, Pyrazolate and Pyrazoxyfen, respectively. For dehulled rice culture, on the other hand, Japonica-type rice cultivar was less tolerant to herbicides of butachlor, propanil, chlornitrofen and oxadiazon that was reversed trend to young seedling culture test. Cultivar differences were also exhibited within same cultivar type. In general, relatively higher tolerant cultivars were Milyang 42, Cheongcheongbyeo, Samgangbyeo, Chilseoungbyeo for Tongil-type, Somjinbyeo for Japonica-type and IR50 for Indica-type, respectively. The response of callus growth showed similar to dehulled rice culture method in all herbicides regardless of property variables. However, concentration response was much sensitive in callus response. The concentration ranges of $10^{-9}M-10^(-8)M$ were appropriate to distinguish the difference between herbicides for E. crusgalli callus growth. Among used herbicides, BAS-514 was the most effective to E. crusgalli callus growth. Based on the above results, tissue culture method could be successfully used as a tool for studying herbicidal behaviour and tolerance screening to herbicide.

• Journal of Oral Medicine and Pain
• /
• v.32 no.2
• /
• pp.187-200
• /
• 2007

This study attempted to contribute to the clinical application of implant operation by making a quantitative nerve examination using a neurometer for the evaluation of sensory disturbances that could be incurred after the implantation in the dental clinics, and it intended to establish an objective guideline in the evaluation of sensory nerve after the operation of implant. An inspection was performed with the frequencies of 2000Hz, 250 Hz and 5 Hz before and after the operations of tooth implant using $Neurometer^{(R)}$ CPT/C (Neurotron, Inc. Baltimore, Maryland, USA) for 44 patients who had performed an implant operation among the patients coming to Daejeon Sun Dental Hospital in 2006 and 30 people for control group. The measuring sites were maxillary nerve ending and mandibular nerve ending of trigeminal nerve according to the implant operating regions. The current perception threshold (CPT) by each nerve fiber was specifically responded under the electric stimulation of 2000 Hz in case of $A{\beta}$ fiber and of 250 Hz in case of $A{\delta}$ fiber and of 5Hz in case of C fiber. The CPT test could be performed to assess the damages of peripheral nerve in the trigeminal nerve area and it stimulated selective nerve fibers by generating the electricity of specific frequency in the peripheral nerve area. The nerve fibers with varied thickness were responsive selectively to the electric stimulation with different frequencies; accordingly, they applied the electric stimulation with different frequencies and the reaction threshold of $A{\beta},\;A{\delta}$ and C fibers selectively responsive to each electric current could be individually evaluated. In the assessment through the CPT, the increase and decrease of the CPT could be measured so that sensory disturbances such as hyperaesthesia or hypoaesthesia could be diagnosed. This study could obtain the following results after the assessment of the CPT before and after the implant operation. 1. In the assessment before and after the implant operation, the CPT in the frequencies of 2000 Hz, 250 Hz, 5 Hz for maxillary branch increased on the whole after the operation and the CPT for mandibular branch in the $A{\beta}$-fiber(2000 Hz) and C-fiber(5 Hz )after the operation increased statistically significantly. 2. For the groups of patients with medically compromised or its subsequent medicinal prescription, there were no significant differences before and after the implant operation and for the control groups, significantly high CPT was shown after the implant operation in the left $A{\beta}$-fiber(2000 Hz) and C-fiber(5 Hz). 3. In the comparison of the measured value of the CPT before the operation between the control group and the implant operation group, the latter group had a significantly high measured value of the CPT in the right $A{\beta}$-fiber(2000 Hz) and C-fiber(5 Hz) and there were significant differences in $A{\beta}$-fiber(2000 Hz) in the CPT assessment after the implant operation for the control group. 4. Male participants had higher CPT than female counterparts; however, there were no statistic significances. In the CPT evaluation before and after implant operation, there were no statistical differences in the male group while the right C-fiber(5 Hz) and left $A{\beta}$-fiber(2000Hz) were significantly high in the female group. 5. In the comparison between the group who complain sensory disturbance and the other group, the CPT increased on the whole in the former group, but there were no statistical significances. In the groups, whom there was an increase in VAS, the CPT after the implant operation in the right C-fiber(5 Hz) increased significantly; meanwhile, in case that the VAS mark was '0' before and after the operation, the CPT after the operation in the left $A{\beta}$-fiber(2000 Hz) increased significantly. This study suggested that the CPT measurements using $Neurometer^{(R)}$ CPT/C, provide useful information of objective and quantitative sensory disturbances for tooth implantation.

• Journal of Intelligence and Information Systems
• /
• v.20 no.2
• /
• pp.137-148
• /
• 2014

Recommender system has become one of the most important technologies in e-commerce in these days. The ultimate reason to shop online, for many consumers, is to reduce the efforts for information search and purchase. Recommender system is a key technology to serve these needs. Many of the past studies about recommender systems have been devoted to developing and improving recommendation algorithms and collaborative filtering (CF) is known to be the most successful one. Despite its success, however, CF has several shortcomings such as cold-start, sparsity, gray sheep problems. In order to be able to generate recommendations, ordinary CF algorithms require evaluations or preference information directly from users. For new users who do not have any evaluations or preference information, therefore, CF cannot come up with recommendations (Cold-star problem). As the numbers of products and customers increase, the scale of the data increases exponentially and most of the data cells are empty. This sparse dataset makes computation for recommendation extremely hard (Sparsity problem). Since CF is based on the assumption that there are groups of users sharing common preferences or tastes, CF becomes inaccurate if there are many users with rare and unique tastes (Gray sheep problem). This study proposes a new algorithm that utilizes Social Network Analysis (SNA) techniques to resolve the gray sheep problem. We utilize 'degree centrality' in SNA to identify users with unique preferences (gray sheep). Degree centrality in SNA refers to the number of direct links to and from a node. In a network of users who are connected through common preferences or tastes, those with unique tastes have fewer links to other users (nodes) and they are isolated from other users. Therefore, gray sheep can be identified by calculating degree centrality of each node. We divide the dataset into two, gray sheep and others, based on the degree centrality of the users. Then, different similarity measures and recommendation methods are applied to these two datasets. More detail algorithm is as follows: Step 1: Convert the initial data which is a two-mode network (user to item) into an one-mode network (user to user). Step 2: Calculate degree centrality of each node and separate those nodes having degree centrality values lower than the pre-set threshold. The threshold value is determined by simulations such that the accuracy of CF for the remaining dataset is maximized. Step 3: Ordinary CF algorithm is applied to the remaining dataset. Step 4: Since the separated dataset consist of users with unique tastes, an ordinary CF algorithm cannot generate recommendations for them. A 'popular item' method is used to generate recommendations for these users. The F measures of the two datasets are weighted by the numbers of nodes and summed to be used as the final performance metric. In order to test performance improvement by this new algorithm, an empirical study was conducted using a publically available dataset - the MovieLens data by GroupLens research team. We used 100,000 evaluations by 943 users on 1,682 movies. The proposed algorithm was compared with an ordinary CF algorithm utilizing 'Best-N-neighbors' and 'Cosine' similarity method. The empirical results show that F measure was improved about 11% on average when the proposed algorithm was used