• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.11
• /
• pp.1009-1019
• /
• 2020

This study presents an exploratory investigation of SMEs in Vietnam to understand the impact of personal-level factors on middle managers' work motivation and the moderating role of work environment. A survey of 450 middle managers (MMs) in 150 Hanoi's SMEs was conducted. The findings of this research showed a significant positive impact of Achievement (ACHV), Recognition (RECOG), and Responsibility (RESP) on work motivation of MMs under the investigation. Furthermore, the result indicated that the work environment affects the relationship between personal-level factors and work motivation of participants. Consequently, both work environment improvement and strategies related to personal-level factors need to be taken into consideration. Especially, Recognition and transparency in Responsibility are appreciated in organizations with a low level of work environment satisfaction. However, there were no indications that Participation (PAR) and Communication (CMM) have a considerable impact on work motivation of respondents, being neither low level nor high level of work environment satisfaction. Based on the findings, recommendations are suggested for Vietnam's SMEs to improve work motivation of MMs, by (i) developing standards with emphasis on their achievement, (ii) paying attention to organizational culture focusing on the responsibility of this managerial level, and (iii) building an adequate incentive system, especially non-financial incentives.

• Nuclear Engineering and Technology
• /
• v.50 no.8
• /
• pp.1234-1245
• /
• 2018

The risk of multi-unit nuclear power plants (NPPs) at a site has received considerable critical attention recently. However, current probabilistic safety assessment (PSA) procedures and computer code do not support multi-unit PSA because the traditional PSA structure is mostly used for the quantification of single-unit NPP risk. In this study, the main purpose is to develop a multi-unit Level 2 PSA method and apply it to full-power operating six-unit OPR1000. Multi-unit Level 2 PSA method consists of three steps: (1) development of single-unit Level 2 PSA; (2) extracting the mapping data from plant damage state to source term category; and (3) combining multi-unit Level 1 PSA results and mapping fractions. By applying developed multi-unit Level 2 PSA method into six-unit OPR1000, site containment failure probabilities in case of loss of ultimate heat sink, loss of off-site power, tsunami, and seismic event were quantified.

• The Journal of Asian Finance, Economics and Business
• /
• v.9 no.5
• /
• pp.135-145
• /
• 2022

To improve audit quality, certain Chinese auditing firms have added a third-level review by an additional signing auditor to the general evaluation by a signing auditor team consisting of an engagement auditor and a partner. Nonetheless, our research-based on 36,033 firm-year observations from 2004 to 2019 reveals that compared to the general review system, auditor teams under the three-level review system are less likely to issue modified audit opinions when abnormal financial conditions arise. This finding suggests that, while larger auditor teams' knowledge, experience, and information advantages can theoretically sharpen their judgment, their performance is more susceptible to interference from divergent opinions, the diffusion of responsibility, and lower energy invested by individual auditors, ultimately impairing their judgment regarding the audited enterprises' abnormal financial conditions. That is, the three-level review system, which aims to improve audit quality, actually worsens audit quality. This conclusion remains valid after the problems of heteroscedasticity and endogeneity are addressed by using firm-level cluster robust standard errors and two-stage regression. We hope that our research will draw the attention of auditing firms, prompting them to reconsider the rationality of the three-level review system.

• IEIE Transactions on Smart Processing and Computing
• /
• v.1 no.3
• /
• pp.133-142
• /
• 2012

This paper presents a salient region detection algorithm for auto-focusing based on the characteristics of a human's visual attention. To describe the saliency at the local, regional, and global levels, this paper proposes a set of novel features including multi-scale local contrast, variance, center-surround entropy, and closeness to the center. Those features are then prioritized to produce a saliency map. The major advantage of the proposed approach is twofold; i) robustness to changes in focus and ii) low computational complexity. The experimental results showed that the proposed method outperforms the existing low-level feature-based methods in the sense of both robustness and accuracy for auto-focusing.

• International Journal of Contents
• /
• v.12 no.3
• /
• pp.1-11
• /
• 2016

This exploratory study examines the critical factors associated with news consumers' perception of information overload and news consumption patterns. An online survey was conducted with Qualtrics panels (N = 1001). The demographics and three antecedent factors of perceived information overload were considered including the frequency of news access through multiple media platforms, level of attention to news, and interest in news. Three news consumption patterns were investigated as possible consequences of perceived information overload: news avoidance, selective exposure, and willingness to pay for news. The results of hierarchical regression analyses revealed a meaningful distinction between general and news information overload. Overall, news consumers who paid more attention to news through newer media/platforms/devices perceived higher levels of information overload, were more willing to pay for the news, and often avoided news or selectively exposed themselves to certain sources of news to manage news information overload.

• Journal of Families and Better Life
• /
• v.28 no.6
• /
• pp.47-57
• /
• 2010

This study examines the effects of parental variable(parental support and supervision), temperament(activity level, attention span/persistence, and emotionality) and the internal locus of control on self-regulation of children. Data were collected from 455 5th and 6th graders and analyzed with Pearson's correlations and pathway analysis. The results were as follows : Children's temperament, internal locus of control and parental variable directly affected children's self-regulation. Parental variables mediated between children's temperament and internal locus of control and self-regulation. Internal locus of control mediated between children's temperament and self-regulation: in addition, the most important variable predicting children's self-regulation was children's attention span/persistence temperament.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2017.11a
• /
• pp.111-112
• /
• 2017

Radon has been considered the greatest source of exposure within the total radiation exposure of the human body. xposure from radon, which exists in indoor air quality, lacks public perception, Radon, which exists anywhere on earth, is not regarded as a state of attention even if it is above the average level. Indoor radon exposure situations are not intentionally introduced, and essentially the attention and responsibilities of radon exposures are assumed to be in indoor occupants. So, these are caused by common uranium and thorium scattering on Earth, and are brought into the building by fine cracks or exposed indicators of the buildings. Therefore, this study aims to reduce the risk of radon rays and reduce radon, which induces diseases caused by breathing in the body of indoor air pollutants and emitting diseases by emitting alpha rays from the radon gas.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.952-953
• /
• 2019

최근 딥러닝으로 진행되는 광학 문자 인식 분야는 대부분 단어 단위로 인식하는 것으로 글자 단위의 영역을 검출하는 데에는 적합하지 못하다. 본 연구는 각 글자의 영역을 검출하기 위해 기존의 딥러닝을 이용한 광학 문자 인식 절차인 단어 분리 과정과 단어 인식 과정을 유지하면서 어텐션 중심을 이용하여 각 글자의 영역을 보다 정확하게 검출하는 것을 목표로 한다. 제안하는 모델은 CRAFT 와 Attention Network 를 사용한 OCR 과정을 확장한 모델로 각 단어 문자열 결과물에 각 글자의 영역을 추가로 나타내게 되며 각 글자와 라벨 간의 IOU 평균은 0.671 로 나타났다.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.399-410
• /
• 2021

Aiming at the problem that complex external variables in natural scenes have a greater impact on facial expression recognition results, a facial expression recognition method based on two-stream convolutional neural network is proposed. The model introduces exponentially enhanced shared input weights before each level of convolution input, and uses soft attention mechanism modules on the space-time features of the combination of static and dynamic streams. This enables the network to autonomously find areas that are more relevant to the expression category and pay more attention to these areas. Through these means, the information of irrelevant interference areas is suppressed. In order to solve the problem of poor local robustness caused by lighting and expression changes, this paper also performs lighting preprocessing with the lighting preprocessing chain algorithm to eliminate most of the lighting effects. Experimental results on AFEW6.0 and Multi-PIE datasets show that the recognition rates of this method are 95.05% and 61.40%, respectively, which are better than other comparison methods.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.2
• /
• pp.119-128
• /
• 2022

Recently, due to the development of related technologies for autonomous vehicles, driving work is changing more safely. However, the development of support technologies for level 5 full autonomous driving is still insufficient. That is, even in the case of an autonomous vehicle, the driver needs to drive through forward attention while driving. In this paper, we propose a method to monitor driving tasks by recognizing driver behavior. The proposed method uses pre-trained deep convolutional neural network models to recognize whether the driver's face or body has unnecessary movement. The use of pre-trained Deep Convolitional Neural Network (DCNN) models enables high accuracy in relatively short time, and has the advantage of overcoming limitations in collecting a small number of driver behavior learning data. The proposed method can be applied to an intelligent vehicle safety driving support system, such as driver drowsy driving detection and abnormal driving detection.