• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2008년도 제39회 하계학술대회
• /
• pp.1545-1546
• /
• 2008

In this study, an image processing method with FRANSAC(Fuzzy RANSAC) is presented and discussed for the development of a microscopic gap measuring algorithm. Many problems in edge detection processing are mainly occurred by the illumination system. A serious problem is that the edge set of gap could include the error elements that have relatively larger error than normal. This problem leads to a incorrect measurement of gap. We present a gap measuring algorithm using FRANSAC[1] that is a representative robust estimation algorithm. FRANSAC is peformed by first categorizing all data into good sample set, bad sample set and vague sample set using a fuzzy classification and then sampling in only good sample set. Experimental results show that the presented gap measuring algorithm gives a higher accurate value of gap especially for the more noisy image data.

• 대한기계학회논문집A
• /
• 제31권1호
• /
• pp.55-61
• /
• 2007

Recently simulation model becomes an essential tool for analysis and design of a system but it is often expensive and time consuming as it becomes complicate to achieve reliable results. Therefore, high-fidelity simulation model needs to be replaced by an approximate model, the so-called metamodel. Metamodeling techniques include 3 components of sampling, metamodel and validation. Cross-validation approach has been proposed to provide sequnatially new sample point based on cross-validation error but it is very expensive because cross-validation must be evaluated at each stage. To enhance the cross-validation of metamodel, sequential sampling method using candidate points and representative cross-validation is proposed in this paper. The candidate and representative cross-validation approach of sequential sampling is illustrated for two-dimensional domain. To verify the performance of the suggested sampling technique, we compare the accuracy of the metamodels for various mathematical functions with that obtained by conventional sequential sampling strategies such as maximum distance, mean squared error, and maximum entropy sequential samplings. Through this research we team that the proposed approach is computationally inexpensive and provides good prediction performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권7호
• /
• pp.3714-3732
• /
• 2019

With the rapid development of network, Intrusion Detection System(IDS) plays a more and more important role in network applications. Many data mining algorithms are used to build IDS. However, due to the advent of big data era, massive data are generated. When dealing with large-scale data sets, most data mining algorithms suffer from a high computational burden which makes IDS much less efficient. To build an efficient IDS over big data, we propose a classification algorithm based on data clustering and data reduction. In the training stage, the training data are divided into clusters with similar size by Mini Batch K-Means algorithm, meanwhile, the center of each cluster is used as its index. Then, we select representative instances for each cluster to perform the task of data reduction and use the clusters that consist of representative instances to build a K-Nearest Neighbor(KNN) detection model. In the detection stage, we sort clusters according to the distances between the test sample and cluster indexes, and obtain k nearest clusters where we find k nearest neighbors. Experimental results show that searching neighbors by cluster indexes reduces the computational complexity significantly, and classification with reduced data of representative instances not only improves the efficiency, but also maintains high accuracy.

• Journal of Preventive Medicine and Public Health
• /
• 제53권2호
• /
• pp.135-142
• /
• 2020

Objectives: In recent years, multiple studies have investigated the issue of intimate partner violence (IPV) in Korea. However, most of those studies have focused on IPV against women, while overlooking the problem of men IPV victimization. Considering this, the current study identified risk and protective factors for IPV and examined their influence on IPV victimization among Korean men. Methods: We used a nationally representative sample of 1668 Korean men from the 2013 Korea National Survey on Domestic Violence. The associations between potential IPV risk factors and different types of IPV were investigated using univariate and multivariate logistic regression. Specifically, separate analyses were conducted of 5 types of IPV (neglect, controlling behaviors, emotional violence, economic violence, and physical violence). Results: The prevalence of IPV among Korean men and women showed only marginal gender differences. Controlling behaviors (men, 23.3%; women, 23.9%) and emotional violence (men, 16.5%; women, 18.8%) were the most common types of IPV reported, followed by neglect (men, 11.2%; women, 11.7%). Separate logistic regression analyses for the 5 subtypes of IPV revealed that mutual IPV was a strong predictor of IPV. Men who abused their wives were more likely to experience neglect (odds ratio [OR], 29.24; p<0.01), controlling behaviors (OR, 36.61; p<0.01), emotional violence (OR, 58.07; p<0.01), economic violence (OR, 18.78; p<0.01), and physical violence (OR, 38.09; p<0.01). Conclusions: The findings of this study suggest that IPV intervention strategies should particularly focus on couples whose relationship is characterized by patterns of bidirectional violence.

• 대한통합의학회지
• /
• 제7권1호
• /
• pp.57-64
• /
• 2019

Purpose : This study was implemented to identify the risk conditions influencing suicidal ideation in a nationally representative sample of elderly South Koreans. Methods : Data from 1,152 men and 1,581 women aged 65 years or older were gathered from the 2013 and 2014 Korea National Health and Nutrition Examination Survey VI. All analyses were performed using SPSS. To determine significant correlations between risk condition and suicidal ideation, a t-test was used. Results : There were differences in suicidal ideation according to the following individual factors: age, educational background, marital status, economic activity, recognition of stress, experience of depression, and economic status. There were differences in suicidal ideation according to the following health-related factors: subjective health status, EQ-5D (EuroQoL-5 Dimensions), hours of sleep, and BMI. There were also differences in suicidal ideation according to the following disease-related factors: HTN, COPD, asthma, stroke, depression and osteoarthritis. Conclusion : The findings indicate that broad intervention programs should be distributed to prevent suicidal ideation. It is also recommended that programs be developed in a way that can help manage the variables identified in this study. Furthermore, follow-up studies should be conducted to verify the program.

• 한국항행학회논문지
• /
• 제16권2호
• /
• pp.386-393
• /
• 2012

본 논문에서는 객체의 관심점(interest points)에 대한 지역 특징 기술자를 이용하여 이미지나 동영상에서 다수의 관심 객체를 효과적으로 인식하고 추적하기 위한 기법을 제안한다. 이를 위해 먼저 대상이 되는 객체를 포함하는 다양한 이미지를 수집하고 SURF 알고리즘을 적용하여 객체의 관심점과 그들에 대한 지역 특징 기술자를 생성한다. 지역 특징에 대한 통계적인 분석을 통하여 관심점들 중에서 해당 객체의 특성을 가장 잘 표현하는 대표점(representative points)을 선택하고 이를 바탕으로 이미지에 존재하는 객체를 인식한다. 또한, 지역 특징 기술자의 정합을 응용하여 각 SURF 지점들의 움직임 벡터를 생성하고 이를 기반으로 실시간으로 객체를 추적한다. 제안하는 기법은 모든 객체를 독립적으로 다루기 때문에, 여러 개의 객체를 동시에 인식하고 추적할 수 있다. 다양한 실험을 통해, 동영상에서 객체의 존재 여부 및 종류를 신속하게 판별하고 관심 객체의 추적을 효과적으로 수행할 수 있음을 보인다.

• 유통과학연구
• /
• 제14권6호
• /
• pp.53-63
• /
• 2016

Purpose - This study examines the effects of the influence strategies on sales representative and headquarter trust, and investigates how sales representative trust plays a mediating role in the relationship between influence strategies and headquarter trust. For these purposes, a structural model which consists of several constructs was developed. In this model, influence strategies that consist of relational influence strategies (information exchange, recommend, promise) and mandatory influence strategies (legal plea, request, threat) were proposed to affect the sales representative trust and in turn, increase the headquarter trust. Thus, this study proposed that sale representative trust plays a core mediating role in the relationship between relational and mandatory influence strategies and headquarter trust in B2B food materials distribution context. Research design, data, and methodology - For these purposes, the authors collected the data from 208 B2B specialized complex agents. We used the 2,200 B2B specialized complex agents which trade with CJ, Ottogi, and Daesang firms and supply food materials to restaurant, school cafeteria, supermarket and traditional market as a sample frame. Once we identified 330 B2B specialized complex agent owners, CEOs, and/or Directors who had agreed to participate in this study, we dropped off a questionnaire at each B2B specialized complex agent and explained the purpose of this study. The survey was conducted from October 1, 2015 to December 15, 2015. A total of 230 questionnaires were collected. Of these collected questionnaires, 28 questionnaires excluded since they had not been fully completed. The data were analyzed using frequency test, reliability test, measurement model analysis, and structural equation modeling with SPSS and SmartPLS 2. Results - First, information exchange, recommendation, and promise of relational influence strategies had positive effects on sales representative trust. The threat of mandatory influence strategies had a negative effect on sales representative trust, but legal plea and request did not have a significant effect on sales representative trust. Second, information exchange and recommendation of relational influence strategies had positive effects on headquarter trust, but promise did not. Also, legal plea, request, and threat of mandatory influence strategies did not have a significant effect on headquarter trust. Third, this findings show that sales representative trust plays a partial mediator between information exchange and headquarter trust, and threat and headquarter trust, and a full mediator between promise and headquarter trust, and recommendation and headquarter trust. Conclusions - The aim of this study was to examine the effects how diverse dimensions of relational and mandatory influence strategies relate to sales representative trust and headquarter trust. To do so, we integrated the influence strategies and the trust transfer theory to hypothesize that various influence strategies increase sales representative and headquarter trust. The findings of this study suggest that headquarter firms should establish and enforce proper influence strategies guidelines to make clear what proper actions sales representatives should implement in relationship with B2B specialized complex agents. Also, relational and mandatory influence strategies must be regarded as a long-term and ongoing strategy that eventually build a long-term orientation with B2B specialized complex agents and guarantee a company's sustainable growth and success.

• 한국데이터정보과학회:학술대회논문집
• /
• 한국데이터정보과학회 2003년도 춘계학술대회
• /
• pp.77-88
• /
• 2003

Cluster analysis has been widely used in many applications, such that data analysis, pattern recognition, image processing, etc. But clustering requires many hours to get clusters that we want, because it is more primitive, explorative and we make many data an object of cluster analysis. In this paper we propose a new clustering method, 'Clustering algorithm using a center of gravity for grid-based sample'. It is more fast than any traditional clustering method and maintains accuracy. It reduces running time by using grid-based sample and keeps accuracy by using representative point, a center of gravity.

• Journal of the Korean Data and Information Science Society
• /
• 제16권2호
• /
• pp.217-226
• /
• 2005

Cluster analysis has been widely used in many applications, such as data analysis, pattern recognition, image processing, etc. But clustering requires many hours to get clusters that we want, because it is more primitive, explorative and we make many data an object of cluster analysis. In this paper we propose a new clustering method, 'Clustering algorithm using a center of gravity for grid-based sample'. It reduces running time by using grid-based sample and keeps accuracy by using representative point, a center of gravity.

• 산업식품공학
• /
• 제21권3호
• /
• pp.191-200
• /
• 2017

There have been great efforts to develop a rapid and sensitive detection method to monitor the presence of pathogenic bacteria in food. While a number of methods have been reported for bacterial detection with a detection limit to a single digit, most of them are suitable only for the bacteria in pure culture or buffered solution. On the other hand, foods are composed of highly complicated matrices containing carbohydrate, fat, protein, fibers, and many other components whose composition varies from one food to the other. Furthermore, many components in food interfere with the downstream detection process, which significantly affect the sensitivity and selectivity of the detection. Therefore, isolating and concentrating the target pathogenic bacteria from food matrices are of importance to enhance the detection power of the system. The present review provides an introduction to the representative sample preparation strategies to isolate target pathogenic bacteria from food sample. We further describe the nucleic acid-based detection methods, such as PCR, real-time PCR, NASBA, RCA, LCR, and LAMP. Nucleic acid-based methods are by far the most sensitive and effective for the detection of a low number of target pathogens whose performance is greatly improved by combining with the sample preparation methods.