• East Asian mathematical journal
• /
• v.37 no.3
• /
• pp.333-354
• /
• 2021

A support vector machine (SVM) is a state-of-the-art machine learning model rooted in structural risk minimization. SVM is underestimated with regards to its application to real world problems because of the difficulties associated with its use. We aim at showing that the performance of SVM highly depends on which kernel function to use. To achieve these, after providing a summary of support vector machines and kernel function, we constructed experiments with various benchmark datasets to compare the performance of various kernel functions. For evaluating the performance of SVM, the F1-score and its Standard Deviation with 10-cross validation was used. Furthermore, we used taylor diagrams to reveal the difference between kernels. Finally, we provided Python codes for all our experiments to enable re-implementation of the experiments.

• Journal of the Korean Home Economics Association
• /
• v.45 no.3
• /
• pp.109-120
• /
• 2007

The purposes of this study were to provide evidence concerning the effects of Emotional Leadership and examine the impacts of Emotional Leadership on employee-related variables which were 'job satisfaction', 'organizational commitment', 'organizational performance', 'turnover intention'. A survey was conducted from August 23 to November 3, 2005 to collect data from kitchen staff(N=611). Statistical analyses were completed using SPSS Win(12.0) for descriptive analysis, reliability analysis, factor analysis, t-test, correlation analysis, cluster analysis and AMOS(5.0) for structural equation modeling. Kitchen staffs gave high point to their leader in the Emotional Leadership competence 'Optimism: seeing the upside in events' and 'Adaptability: flexibility in handing change' and gave lower point in the Emotional Leadership competence 'Inspirational leadership: guiding and motivating with compelling vision'. Employees' job satisfaction on 'coworker' were relatively high. However, the extents of satisfaction on 'payroll', 'promotion', 'work environment' were relatively low. The organizational commitment score was higher at 'loyalty' factor than 'commitment' factor. the test of hypothesis using structural equation modeling found that Emotional Leadership produced positive effects on job attitude and job performance. In conclusion, this study has identified that the Emotional Leadership effects on their organizational performance and attitudes toward their job.

• Earthquakes and Structures
• /
• v.14 no.6
• /
• pp.615-626
• /
• 2018

The seismic vulnerability of Turkey is relatively high due to its active fault systems with potential to create destructive earthquakes. Thus, reducing the loss of life and property, the number of the earthquake-prone buildings and their retrofit requirements are considerably significant key issues under the scenario earthquakes. The street survey based rapid assessment (SSRA) method can be considered as a powerful tool to determine the seismic vulnerability of building stock of an earthquake-prone city/state. In this study, the seismic vulnerability of the building stock of the Kirikkale province in Turkey is aimed to be estimated adopting the street survey based rapid assessment method (SSRA). For this purpose, central 2074 existing reinforced concrete (R/C) buildings were structurally surveyed with rapid visual site screening and disadvantages such as, the existence of short-column, soft-story, heavy overhangs, pounding effect and local soil conditions were determined for obtaining the structural performance score of each. The results obtained from the study demonstrate that 11-25% of the surveyed buildings in the study region needs to be investigated through more advanced assessment methods. Besides, higher correlation between increasing story number and unsafe/safe building ratio is obtained for the buildings with soft-story parameter than that for those with heavy overhangs and short-column parameters. The conformity of the results of the current study with the previous documented cases of rapid assessment efforts in the recent earthquakes in Turkey shows that the SSRA method for the Kirikkale province performed well, and thus this methodology can be reliably used for similar settlement areas.

• ETRI Journal
• /
• v.36 no.3
• /
• pp.429-438
• /
• 2014

Semantic role labeling (SRL) is a task in natural-language processing with the aim of detecting predicates in the text, choosing their correct senses, identifying their associated arguments, and predicting the semantic roles of the arguments. Developing a high-performance SRL system for a domain requires manually annotated training data of large size in the same domain. However, such SRL training data of sufficient size is available only for a few domains. Constructing SRL training data for a new domain is very expensive. Therefore, domain adaptation in SRL can be regarded as an important problem. In this paper, we show that domain adaptation for SRL systems can achieve state-of-the-art performance when based on structural learning and exploiting a prior model approach. We provide experimental results with three different target domains showing that our method is effective even if training data of small size is available for the target domains. According to experimentations, our proposed method outperforms those of other research works by about 2% to 5% in F-score.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.12
• /
• pp.223-231
• /
• 2020

This research aims to examine (1) the effect of carbon emission disclosure on firm value, (2) the effect of good corporate governance on firm value, (3) the mediating role of financial performance between carbon emission disclosure and firm value, and (4) the mediating role of financial performance between good corporate governance and firm value. The research sample includes 43 mining, agro, and manufacturing firms listed in the Indonesian Stock Exchange over the 2015-2017 period. Carbon emission disclosure is measured by an indicator of the Global Reporting Initiative Series of Environmental Aspect. Good corporate governance is measured by the corporate governance score of shareholder rights, boards of directors, outside directors, audit committee and internal auditor, and disclosure to investors. Financial performance is measured by return on assets, while firm value is measured by Tobin's Q. Data analysis uses the structural equation modeling. The result shows carbon emission disclosure and good corporate governance have no direct effect on firm value. On the other hand, financial performance mediates the effect of carbon emission disclosure and good corporate governance on firm value. It shows that higher carbon emission disclosure and good corporate governance are meaningless for the investor if they do not give any financial performance improvement.

• Korean Journal of Health Education and Promotion
• /
• v.19 no.1
• /
• pp.59-85
• /
• 2002

The purpose of this study was to examine the health promoting behavior of university students and its related factors. A self-administered questionnaire survey was carried out for 1,031(male, 477, female, 554) students from 7 universities in Taegu from March 10 to April 15 2000. Collected data were analyzed by the $X^2$-test, t-test, ANOV A and structural equation modeling analysis. The results of this study were as follows; Mean score of health promoting behavior of women was higher than mean score of men(p〈0.01). According to the subordinate fields, female students showed significantly higher scores in the fields of stress management, nutrition, personal relations and alcohol-cigarette(p〈0.05). And in male students, the only fields of exercise had meaningfully higher score(p〈0.01). The highest degree of performance was marked in alcohol-cigarette field in female, and interpersonal field in male. In Both male and female students, the lowest field was health responsibility. The mean score of health promoting behavior showed significant differences in intermediating factors such as age, health education and grade in male, and such as grade, pocket money for a month, taking health education, perceived body shape and gastrointestinal symptoms in female(p〈0.05). The mean score of health promoting behavior showed significant differences in cognitive-perceptual factors and cues to action such as perceived health status, perceived benefits, perceived barriers, and health concern of parents in both male and female(p〈0.05). By the results of structural equation modeling analysis in male students, the economic status, gastrointestinal symptoms and perceived body shape as lean had significant direct effects on cognitive-perceptual factors($\mid$T$\mid$〉2.0). Grade and health concern of parents had significant direct effects on health promotion behaviors($\mid$T$\mid$〉2.0). In female students, perceived body shape as lean and perceived body shape as fat had significant direct effects on cognitive-perceptual factors($\mid$T$\mid$〉2.0). Grade, pocket money and health concern of parents had significant direct effects on health promotion behaviors($\mid$T$\mid$〉2.0). On the consideration of above findings, it is recommended that perceived health status and perceived benefits should be raised and perceived barriers should be lowered to increase health promoting behaviors, by means such as institutional approach which can enhance the primary prevention for a disease. In addition, the health policy for improving parents' concern about health as well as university students, through various ways such as health education, consulting, and health program, should be employed.

• Korean Journal of Health Education and Promotion
• /
• v.20 no.3
• /
• pp.91-108
• /
• 2003

Psychological stress is a growing issue in work stress research because work stressors are closely related to depression； and depression, in turn, decreases organizational effectiveness. Considering such causal relationships of work stress, a comprehensive source to control work stress is needed for worksite mental well-being. This study was conducted to identify how social support at work controlled work stress and which characteristics of social support were effective on work stress reduction. The study participants were 240 workers employed in a public hospital in Georgia, U.S.A self-administered survey was given to employees with their pay slips, and followed by a hospital wide voice reminder for 7 days. Surveys were conducted over a 20-day period. The questionnaires asked about job demands, job control, social support at work, depression, job performance, absenteeism, and demographics. The social support construct was structured on the source of support at work and the kinds of support were provided. Statistical analyses were conducted in the structural equation modeling approach. Social support at work was directly related to high job control, low depression, and high job performance. High score of social support at work were significantly associated with high job control, low depressive symptoms, and high job performance. By source of support, only organizational support was positively related to high job control. Organizational support was more effective than supervisor and coworker support. Any stressors and their outcomes were not differenciated by the kinds of support. This result indicated that job control was influenced more by the source of support than the kinds of support provided at work； and the most efficient source of support was the organization. Organizational support was a strong factor in improving workers" perceived controllability of their jobs from a work stress reduction perspective.tive.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.22 no.6
• /
• pp.138-147
• /
• 2018

With the recent series of damage caused by earthquakes in Korea, such as Gyeongju and Pohang, we know that Korea is no longer a safe zone for earthquakes and that we need to be prepared for them. In addition, bridges built prior to the introduction of seismic design concepts remain without adequate seismic reinforcement measures, and earthquake reinforcement should be performed efficiently considering economic and structural safety. Preliminary assessment of seismic performance of existing bridges is divided into four seismic groups, taking into account seismicity, vulnerability and Impact, considering the magnitude of the existing bridge's seismic, and prioritization for further evaluation of seismic performance. In this study, unlike the existing anti-seismic reinforcement priority method, scores are calculated based on the seismic design criteria applied to bridges, importance coefficient of the bridge including the zone coefficient and the Importance, vulnerability index of the bridge including the soil condition and the elapsed years, detail coefficient of the bridge including the superstructure form, the span length, the width, the height, the design load, and the daily traffic volume. The calculated score items will be weighted and grouped according to the results. Using this, a simpler and more efficient algorithm was proposed to determine the priority of seismic reinforcement on a bridge.

• Journal of Korean Society for Quality Management
• /
• v.35 no.2
• /
• pp.63-83
• /
• 2007

As the mobile internet service market is rapidly growing, measuring and managing the service quality become a more critical issue to the service providers. Thus the purpose of this study is to compare SERVQUAL, SERVPERF, and non-difference score measures and to determine which one is superior to measure the service quality of the mobile internet service. This study also aims to analyze the structural relationship between the service quality, customer satisfaction and behavioral intentions (repurchase intention and word-of-mouth). For the purpose, survey data were gathered from the respondents who have used the mobile internet services and an empirical test was conducted. As the first research result, it was revealed that the non-difference score measure showed a much better model than other service measuring models (SERVQUAL, SERVPERF). And the second result was that two service quality factors (tangibles and empathy) among the mobile internet service quality factors significantly influenced the customer satisfaction and also the customer satisfaction had a significant effect on the repurchase and word-of-mouth.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.221-235
• /
• 2022

Manual inspection of steel box girders on long span bridges is time-consuming and labor-intensive. The quality of inspection relies on the subjective judgements of the inspectors. This study proposes an automated approach to detect and segment cracks in high-resolution images. An end-to-end cascaded framework is proposed to first detect the existence of cracks using a deep convolutional neural network (CNN) and then segment the crack using a modified U-Net encoder-decoder architecture. A Naïve Bayes data fusion scheme is proposed to reduce the false positives and false negatives effectively. To generate the binary crack mask, first, the original images are divided into 448 × 448 overlapping image patches where these image patches are classified as cracks versus non-cracks using a deep CNN. Next, a modified U-Net is trained from scratch using only the crack patches for segmentation. A customized loss function that consists of binary cross entropy loss and the Dice loss is introduced to enhance the segmentation performance. Additionally, a Naïve Bayes fusion strategy is employed to integrate the crack score maps from different overlapping crack patches and to decide whether a pixel is crack or not. Comprehensive experiments have demonstrated that the proposed approach achieves an 81.71% mean intersection over union (mIoU) score across 5 different training/test splits, which is 7.29% higher than the baseline reference implemented with the original U-Net.