• Convergence Security Journal
• /
• v.23 no.1
• /
• pp.27-33
• /
• 2023

Digitalization due to COVID-19 has rapidly developed artificial intelligence-based voice recognition technology. However, this technology causes unfair social problems, such as race and gender discrimination if datasets are biased against some groups, and degrades the reliability and security of artificial intelligence services. In this work, we compare and analyze accuracy-based unfairness in biased data environments using VGGNet (Visual Geometry Group Network), ResNet (Residual Neural Network), and MobileNet, which are representative CNN (Convolutional Neural Network) models of artificial intelligence. Experimental results show that ResNet34 showed the highest accuracy for women and men at 91% and 89.9%in Top1-accuracy, while ResNet18 showed the slightest accuracy difference between genders at 1.8%. The difference in accuracy between genders by model causes differences in service quality and unfair results between men and women when using the service.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.11
• /
• pp.455-464
• /
• 2020

This study examined the correlation between the variables related to the organizational commitment of social workers. The research results are as follows. First, the overall effect size of the variable group had an intermediate effect size. Among the four organizational commitment-related variables, the job-related negative variable had the largest effect size. Next, the group of positive variables related to the organization, the group of positive variables related to the job, and the group of individuals related to the individual were in order. Second, among the individual-related variables, experience, gender, and education showed a small effect size, and other factors were close to the small effect size. In the group of positive variables related to organization, all variables except for transactional leadership and organizational fairness, which had a medium effect size, showed a large effect size. The job-related positive variables showed a large effect size only for job satisfaction, and customer orientation and job suitability were close to the large effect size. The turnover intention, which is a group of job-related negative variables, had a large effect size and job stress showed a medium effect size.

• Journal of Distribution Science
• /
• v.13 no.2
• /
• pp.63-73
• /
• 2015

Purpose - Satisfaction with service is evaluated according to customers' subjective judgment. The expected value of customer service and its evaluations depend on the customers' position. The customer recognizes two different forms of service levels. One is satisfaction and the other is dissatisfaction. Customers who are satisfied want to receive the service in future. However, those dissatisfied try to change the service. The service provider tries to improve the service. There are two different service cycles. One is the successful cycle and the other is the failure cycle. This study aimed to empirically determine the effects of the justice and authenticity of service recovery on customer behavioral intention through an integrated approach to cognitive justice and psychological authenticity. Research design, data, and methodology - Based on a literature review, justice of service recovery was categorized into three types: distributive, procedural, and interactive. Then, authenticity was added to obtain four independent variables, along with recovery satisfaction as a parameter. Behavioral intention, as an outcome variable, was divided into the repurchase intention and positive word-of-mouth. The model and hypotheses were created and measurement items were developed. A questionnaire survey of items concerning the service recovery experience at family restaurants was conducted on college students and residents in Gwangju from September 30 to October 31, 2013. A total of 400 copies of the questionnaire were sent out and 385 were returned. Respondents answered questions about the importance of, and satisfaction with service recovery on a 5-point Likert scale. Excluding 174 copies without service failure experiences and 7 inappropriate copies, 204 copies were analyzed using SPSS 21.0 for Windows and AMOS 20.0 to determine the reliability and validity of measurements. The hypotheses were tested through a goodness-of-fit analysis. Results - First, distributive justice positively affected recovery satisfaction. Second, procedural and interactive justice had no impact. Third, authenticity positively affected recovery satisfaction. Fourth, distributive justice had relatively stronger effects on recovery satisfaction than authenticity. Fifth, recovery satisfaction significantly affected repurchase intention and positive word-of-mouth and it proved effective in mediation. Finally, additional analysis was performed for descriptive statistics of the principal variables by various demographic characteristics and significant differences were found in gender, occupation, and so on. Conclusions - This study has academic significance as the fairness and authenticity of service recovery were investigated to reveal the effects on behavior. The findings could be applied to a wide range of service recovery strategies. However, there are some limitations. First, data was collected only from the residents of Gwangju and most respondents were aged 20-30. Future studies should target a wide range of areas and age groups. Second, because the questionnaire used in this study targets only convenience family restaurants, the results of this study cannot be generalized to all services companies. Future research should be done on a wide range of industries such as hotels, airlines, and hospitals, and perform a comparison between sectors.