• KIISE Transactions on Computing Practices
• /
• v.21 no.11
• /
• pp.681-688
• /
• 2015

The k nearest neighbor (k-NN) graph construction is an important operation with many web-related applications, including collaborative filtering, similarity search, and many others in data mining and machine learning. Despite its many elegant properties, the brute force k-NN graph construction method has a computational complexity of $O(n^2)$, which is prohibitive for large scale data sets. Thus, (Key, Value)-based distributed framework, MapReduce, is gaining increasingly widespread use in Locality Sensitive Hashing which is efficient for high-dimension and sparse data. Based on the two-stage strategy, we engage the locality sensitive hashing technique to divide users into small subsets, and then calculate similarity between pairs in the small subsets using a brute force method on MapReduce. Specifically, generating a candidate group stage is important since brute-force calculation is performed in the following step. However, existing methods do not prevent large candidate groups. In this paper, we proposed an efficient algorithm for approximate k-NN graph construction by regrouping candidate groups. Experimental results show that our approach is more effective than existing methods in terms of graph accuracy and scan rate.

• Journal of Korean Home Economics Education Association
• /
• v.23 no.3
• /
• pp.139-160
• /
• 2011

The purpose of this study is twofold: (i) to develop a teaching and learning plan and practical performance assessment tools for the improvement of teenager-parent communication and relationships as well as explore their effects on the communication in the everyday family life; and (ii) to find the underlying problems of teenager- parent communication through conversation analysis and to provide a improved dialogue model. We provided the experimental group with a performance task of communication training between teenagers and their parents in the real family situation while the control group practiced communication skills in a learning situation. However for both classes, before and after performance tasks were equally provided. The experimental group exhibited a longer conversation time with their parents, better communication skills, and higher degrees of relational satisfaction than the control group. Conversation analysis revealed that the experimental group reduced the use of blocking techniques in the teenager-parent conversations more than the control group, and all so raised the frequency of functional communications more than the control group. In both areas of communication in the experimental group was significantly improved, Most notably, a problem-solving case through no-lose conflict resolution methods was effective, succeeding by 70% in the e experimental group and 43.3% in the control group. Parents use blocking techniques like admonition, lecturing, blaming. sarcastic remarking, ordering and so forth, while teenagers use dispute, avoidance, blaming, and teasing in this order. The communication problems during the conversation process, teenagers' evasive and rebellious way of speaking instigates adverse communication responses from parents, so their conversation tends to unfold as ambiguous evasion opposed to: inquiring or evasion by short answers vs. ordering-preaching, or disputing vs. criticizing-making sarcastic, disputing vs. disputing-teaching, and criticizing vs. criticizing.

• Journal of Intelligence and Information Systems
• /
• v.22 no.1
• /
• pp.139-157
• /
• 2016

Bankruptcy involves considerable costs, so it can have significant effects on a country's economy. Thus, bankruptcy prediction is an important issue. Over the past several decades, many researchers have addressed topics associated with bankruptcy prediction. Early research on bankruptcy prediction employed conventional statistical methods such as univariate analysis, discriminant analysis, multiple regression, and logistic regression. Later on, many studies began utilizing artificial intelligence techniques such as inductive learning, neural networks, and case-based reasoning. Currently, ensemble models are being utilized to enhance the accuracy of bankruptcy prediction. Ensemble classification involves combining multiple classifiers to obtain more accurate predictions than those obtained using individual models. Ensemble learning techniques are known to be very useful for improving the generalization ability of the classifier. Base classifiers in the ensemble must be as accurate and diverse as possible in order to enhance the generalization ability of an ensemble model. Commonly used methods for constructing ensemble classifiers include bagging, boosting, and random subspace. The random subspace method selects a random feature subset for each classifier from the original feature space to diversify the base classifiers of an ensemble. Each ensemble member is trained by a randomly chosen feature subspace from the original feature set, and predictions from each ensemble member are combined by an aggregation method. The k-nearest neighbors (KNN) classifier is robust with respect to variations in the dataset but is very sensitive to changes in the feature space. For this reason, KNN is a good classifier for the random subspace method. The KNN random subspace ensemble model has been shown to be very effective for improving an individual KNN model. The k parameter of KNN base classifiers and selected feature subsets for base classifiers play an important role in determining the performance of the KNN ensemble model. However, few studies have focused on optimizing the k parameter and feature subsets of base classifiers in the ensemble. This study proposed a new ensemble method that improves upon the performance KNN ensemble model by optimizing both k parameters and feature subsets of base classifiers. A genetic algorithm was used to optimize the KNN ensemble model and improve the prediction accuracy of the ensemble model. The proposed model was applied to a bankruptcy prediction problem by using a real dataset from Korean companies. The research data included 1800 externally non-audited firms that filed for bankruptcy (900 cases) or non-bankruptcy (900 cases). Initially, the dataset consisted of 134 financial ratios. Prior to the experiments, 75 financial ratios were selected based on an independent sample t-test of each financial ratio as an input variable and bankruptcy or non-bankruptcy as an output variable. Of these, 24 financial ratios were selected by using a logistic regression backward feature selection method. The complete dataset was separated into two parts: training and validation. The training dataset was further divided into two portions: one for the training model and the other to avoid overfitting. The prediction accuracy against this dataset was used to determine the fitness value in order to avoid overfitting. The validation dataset was used to evaluate the effectiveness of the final model. A 10-fold cross-validation was implemented to compare the performances of the proposed model and other models. To evaluate the effectiveness of the proposed model, the classification accuracy of the proposed model was compared with that of other models. The Q-statistic values and average classification accuracies of base classifiers were investigated. The experimental results showed that the proposed model outperformed other models, such as the single model and random subspace ensemble model.

• Journal of the Korean earth science society
• /
• v.21 no.2
• /
• pp.103-115
• /
• 2000

The ability-grouping is the essence of the seventh educational curriculum, applied to school from year 2000, and its enriched-supplementary type will be carried out for science course. This study examines the effect of the enriched-supplementary ability-grouping within class to student's academic achievement and the attitude, related to science. Thus we developed teaching and learning methods with intellectual level about the subject of 'Water Circulation and Weather Change' in Middle-School Science 2. Then we tested 152 eighth graders who were divided into the experimental and control groups. The experimental one was taught through the ability-grouping for about six weeks, while the control through conventional lecture. The improvement of the experimental group in academic achievement was more effective than that of the control, and particularly to below-average students who ranked in lower thirty percent. The experimental one got more negative change in domain 'Science as a Subject, and in subdomain 'Anxiety in Science Lesson'. While outstanding students who ranked in upper thirty percent showed a significant positive change in subdomain 'Satisfaction in Teaching Method, the below-average were negatively changed in subdomain 'Anxiety in Science Lesson'. The current ability-grouping was suitable for the improvement of academic achievement, but not for the general attitude related to science. In order to enhance the ability-grouping effect in science education, we need to additionally consider student's interest and concern in grouping, and develop various teaching and learning methods together with proper textbook contents.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.11
• /
• pp.73-81
• /
• 2016

In general, infrared images have less sharpness and image details than visible images. So, the prior image upscaling methods are not effective in the infrared images. In order to solve this problem, this paper proposes an algorithm which initially up-scales an input infrared (IR) image by using adaptive dynamic range encoding (ADRC)-based super-resolution (SR) method, and then fuses the result with the corresponding visible images. The proposed algorithm consists of a up-scaling phase and a fusion phase. First, an input IR image is up-scaled by the proposed ADRC-based SR algorithm. In the dictionary learning stage of this up-scaling phase, so-called 'pre-emphasis' processing is applied to training-purpose high-resolution images, hence better sharpness is achieved. In the following fusion phase, high-frequency information is extracted from the visible image corresponding to the IR image, and it is adaptively weighted according to the complexity of the IR image. Finally, a up-scaled IR image is obtained by adding the processed high-frequency information to the up-scaled IR image. The experimental results show than the proposed algorithm provides better results than the state-of-the-art SR, i.e., anchored neighborhood regression (A+) algorithm. For example, in terms of just noticeable blur (JNB), the proposed algorithm shows higher value by 0.2184 than the A+. Also, the proposed algorithm outperforms the previous works even in terms of subjective visual quality.

• Culinary science and hospitality research
• /
• v.19 no.1
• /
• pp.230-243
• /
• 2013

The purpose of this study is to offer the basic materials for the development of nutrition education programs for youth and help domestic science teachers in schools implement effective dietary education by examining youth's purchase behavior of processed foods. As a result of figuring out youth's purchasing behavior of processed food and the difference in accordance with social, demographic variables, they considered taste and price mainly when choosing foods. The results showed that what they consider important when checking food display information was shelf life and price. It was observed that 56% of them check additives display information in food when purchasing processed food. In terms of demographic factors, the more likely they are a girl student, the lower grader they are, and the lower price they purchase processed food at, the better they used the nutritional knowledge learned in school. Based upon these results, it is necessary to offer the consumer's level of education and training for their demands by accurately figuring out youth's purchasing behavior of processed foods. For this, home economics education must allow youth to lead healthy diet by implementing a systematic and professional training on food additives on a basis of the research and utilization of a variety of educational media and teaching and learning methods.

• Journal of the Korean Society of Food Culture
• /
• v.27 no.6
• /
• pp.666-676
• /
• 2012

The purpose of this study was to propose effective marketing communication strategies for overseas Korean restaurants through a multilateral comparison analysis of American, Chinese, and Japanese consumers' attitudes towards communication media and Korean restaurants. The survey was written in English, Chinese, and Japanese, with guideline for surveyors, and conducted using both online and offline methods. Samples were collected from five major cities - Los Angeles, New York, Tokyo, Beijing and Shanghai, which are the foothold for the globalization of Korean food. When it comes to attitudes towards communication media, word-of-mouth showed a high mean value, indicating it as the most useful and reliable media recognized by consumers who visited Korean restaurants. Furthermore, the necessity of recognizing the importance of visual communication in the physical environment of Korean restaurants and specialized websites, featuring restaurants and gourmet food, was observed. Consumers in all three nations chose word-of-mouth as the most useful and reliable media for learning about Korean restaurants. In addition, American consumers highly depended on signage and restaurant exteriors. Chinese consumers highly recognized the usefulness and reliability of offline media, such as newspapers, magazines, and events, while Japanese consumers considered online media, such as gourmet websites, blogs and social networks, as useful and reliable sources. A significantly positive attitude and high value was observed in all who had visited Korean restaurants. American and Japanese consumers had a significantly higher rate of intention to visit Korean restaurants in the future and to tell others about their satisfaction with Korean restaurants. Meanwhile, the average rate of prior preference for Korean restaurants (when choosing restaurants) was the lowest in all three countries. This study is useful for both the Korean government and food enterprises abroad to plan and develop marketing communication strategies properly for overseas Korean restaurants.

• Journal of Preventive Medicine and Public Health
• /
• v.37 no.1
• /
• pp.80-87
• /
• 2004

Objectives: To evaluate the learning achievement and satisfaction levels for the Field Epidemiology Specialist Training Program (FESTP), on infectious disease control between March 19 and October 31, 2002. Methods : The FESTP was designed as a set of 84 hours curricula including lectures, discussions, self-studies, and field practicals, and organized both centrally and locally by the Division of Communicable Disease Control of the National Institute of Health and 11 universities. Before and after the program, a questionnaire survey on the educational need (49 items) and satisfaction (15 items) was conducted on 484 trainees, who were responsible for communicable disease control and immunization at 242 regional health centers. The data were analyzed with paired t-tests for comparison of the educational needs between the pre and post scores. Results : The average score for satisfaction was 3.06 out of 5.0; with relatively higher scores for sincerity (4.10) and professionalism (4.01) of the tutors, adequacy (3.54) and clearness (3.51) of the evaluation criteria, usefulness (3.54) and fitness (3.52) of the contents, but with relatively lower satisfaction for schedule (2.96) and self-studies (2.91). The average for requirement for education improved, as shown by the decrease from 2.72 to 2.22 (p<.0001) with the biggest decrease in the outbreak investigation from 2.60 to 2.08. Conclusion : The FESTP was evaluated as being effective, the trainees showed moderate satisfaction and decrease educational needs. However, the actual schedules and self-studies should be rearranged to improve the satisfaction level.

• The Journal of Korean Society for School & Community Health Education
• /
• v.8 no.2
• /
• pp.49-59
• /
• 2007

Objectives: This study was designed to understand the oral health fact of high school students, search for an education method and learning goal appropriate for the subjects, and utilize them as basic data to think of effective solutions for oral health care. Methods: The samples to achieve the purpose of this research are composed of 386 students in J industrial high school, OO city, Kyounggido, the number of male students 286, and female students 100. Data were statistically analyzed by frequency analysis, cross tabulation by using SPSS WIN Ver. 12.0. Results: First, The answers in terms of oral symptoms were in the order of 31.8% for gum hurt and bleeding, 31.4% for pain in teeth by drinking or eating hot and cold drinks or food, 27.1% for broken or cracked teeth, 25.7% for aching and throbbing teeth, 20.6% for unpleasant mouth smell, and 5.7% for pain of tongue and soft oral tissues. Second, In terms of behavior for oral health, 88.4% answered that they consume cariogenic snacks, while 29.2% of them, 2.5% of male and 6.3% of female enjoy consuming them but hardly brush their teeth after consumption. Third, In terms of timing for brushing teeth, the results show that 50.0% do so before breakfast, 47.9% after breakfast, 15.2% after lunch, 36.9% after dinner, 65.5% before going to sleep, and 3.5% after consuming snacks, whereas 64.2% answered that they are not aware of fluorine mouthwash. Fourth, In terms of oral health condition. proportion of children with one or more caries in permanent teeth was 71.0%, while 51.4% have $2{\sim}3$ cariogenic teeth, and 16.8% have more than 4 cariogenic teeth. Considering sex, female students have somewhat higher ratio showing 68.5% for male and 78.0% for female, but the ratio was 53.1% for male and 46.7% for female in case of $2{\sim}3$ cariogenic teeth, and 17.5% for male and 15.8% for female in case of more than 4 cariogenic teeth. 9.3% of students turned out to have decayed teeth. Fifth, In terms of oral hygiene condition, only 7.9% of them had excellent condition, and 38.4% need to improve. More specifically, 41.1% of male students and 30.6% of female students need to improve their oral hygiene condition. Conclusions: The study results suggest that School Dental health programs increase and need to be community-incorporated and comprehensive from elementary to high school.

• Journal of the Korean earth science society
• /
• v.31 no.2
• /
• pp.172-184
• /
• 2010

This study is to investigate the problem solving styles according to the left /right brain preference among earth science gifted students. We took the R/LCT and the test of BPI to investigate the brain preference of earth science gifted students (N=16), and took S-CPST to investigate the problem solving styles on them. In the R/LCT, the earth science gifted students were classified into 3 groups (8 left-brain preference students, 7 right-brain preference students, 1 middle-brain preference student). In the BPI, 8 students had the appearance of left-brain preference, whereas 8 students had the appearance of right-brain preference. According to the result of S-CPST, first the left brain preference students tended to resolve a problem into simple components, then they put together each simple component. They prefer to solve a problem using numbers and mathematical signs logically, but they were afraid of giving trouble to describe own idea with pictures. Whereas the right brain preference students solved a problem with 3 steps. First, they saw an overall form of problem. Second, they tried to analyze each simple component of it, and then, made up all in one. Also, the right brain preference students observed the intuitive pattern of problem first, and then suggested the various problem solving methods later, and they took a solving plan using a picture in detail. In sum, earth science gifted students are unequal in problem solving styles according to the left/right brain preference. Thus, a teaching-learning method needs to be developed based on left/right brain preference for more effective gifted education.