• International Journal of Internet, Broadcasting and Communication
• /
• v.15 no.3
• /
• pp.58-65
• /
• 2023

Bankruptcy is a significant risk for start-up companies, but with the help of cutting-edge artificial intelligence technology, we can now predict bankruptcy with detailed explanations. In this paper, we implemented the Category Boosting algorithm following data cleaning and editing using OpenRefine. We further explained our model using the Shapash library, incorporating domain knowledge. By leveraging the 5C's credit domain knowledge, financial analysts in banks or investors can utilize the detailed results provided by our model to enhance their decision-making processes, even without extensive knowledge about AI. This empowers investors to identify potential bankruptcy risks in their business models, enabling them to make necessary improvements or reconsider their ventures before proceeding. As a result, our model serves as a "glass-box" model, allowing end-users to understand which specific financial indicators contribute to the prediction of bankruptcy. This transparency enhances trust and provides valuable insights for decision-makers in mitigating bankruptcy risks.

• Asia Marketing Journal
• /
• v.11 no.2
• /
• pp.121-147
• /
• 2009

This research attempts to investigate differences in past experiences of using internet in repurchasing. In doing so, the authors identify online consumer characteristics, particularly one-time and repeat users comparison across geographical borders of Korean and Taiwanese teenage customers. Results show that there are significant differences in online shopping typologies between Korean and Taiwanese customers. While attitude serves as a distinguishing factor for both data sets along the other two dimensions, trust does not. Since researchers have focused on comparisons between Western cultures and Asian cultures' online purchase behavior or Internet use, the current study provides a valuable comparison for this niche population of young customers at least in a Korean-Taiwanese context. The authors also make a brief argument that these findings can influence marketing practitioners and site developers in their strategies.

• ETRI Journal
• /
• v.46 no.1
• /
• pp.118-126
• /
• 2024

In human conversations, listeners often utilize brief backchannels such as "uh-huh" or "yeah." Timely backchannels are crucial to understanding and increasing trust among conversational partners. In human-machine conversation systems, users can engage in natural conversations when a conversational agent generates backchannels like a human listener. We propose a method that simultaneously predicts backchannels and recognizes speech in real time. We use a streaming transformer and adopt multitask learning for concurrent backchannel prediction and speech recognition. The experimental results demonstrate the superior performance of our method compared with previous works while maintaining a similar single-task speech recognition performance. Owing to the extremely imbalanced training data distribution, the single-task backchannel prediction model fails to predict any of the backchannel categories, and the proposed multitask approach substantially enhances the backchannel prediction performance. Notably, in the streaming prediction scenario, the performance of backchannel prediction improves by up to 18.7% compared with existing methods.

• Journal of Internet Computing and Services
• /
• v.13 no.2
• /
• pp.127-136
• /
• 2012

IT industry is rapidly changing technologies and trends are leading the field. Website also typical of these areas in the field of IT as a sector, technological advances faster, depending on the nature of the IT industry and the emergence of new services, changes in a variety of platforms, and is rapidly rising. In addition, a website containing information in accordance with the social situation changes and exposure of the agenda and vary greatly. Therefore, these changes in terms of B2C websites that can make a difference in the reliability factor looks. From this perspective, in this study to compare data before and after 10 years on the reliability factor by conducting longitudinal studies, most are not addressed in previous studies due to time differences website users to change the reliability of the analyzed factors.

• The Journal of the Korea Contents Association
• /
• v.17 no.11
• /
• pp.252-263
• /
• 2017

The study empirically explored factors influencing photo sharing behavior on Instagram based on the uses and gratification theory. An online survey was conducted 356 respondents those who have used Instagram and have shared photos within the last three months. Frequency analysis, descriptive statistics, confirmatory factor analysis, and structural model analysis were conducted using SPSS 19.0 and AMOS 21.0. The results of this study are as follows: first, communication has the greatest effect on photo-sharing behavior; second, entertainment has a positive effect on photo-sharing behavior; third, self-expression has a positive effect on photo-sharing behavior; fourth, service users' trust has a positive effect on photo-sharing behavior. This study adds to the current literature and contributes to understanding consumers' photo-sharing behavior.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.27 no.2
• /
• pp.29-50
• /
• 2016

This study conducted a survey research in order to prove the influence of public libraries on the creation of social capital. The analysis showed that the use of public libraries is generally focusing on information activities on a personal level and it is not extended to establish social relationships or form social trust between people on social level. If the personal information activities in public libraries are expanded to share information and opinions with other people, however, it can establish social relationships across people mediated by public libraries. It also has the possibilities to induce library users to participate reciprocal activities on social level. As a result, the information activities in public libraries on personal level can be extended to social levels mediated by public libraries, which can affect the creation of social capital. Based on these results, public libraries need to consider the social responsibilities on creating social capital as well as supporting information activities of members of society.

• The Journal of the Korea Contents Association
• /
• v.20 no.2
• /
• pp.203-214
• /
• 2020

Nowadays, smart speakers are increasingly personalized and serve as recommendation agents for user. The aim of this study is find out effects of 'Explanation facilities' on transparency, perceived trust, user satisfaction, behavioral intentions of users to reuse, privacy risk, and quality of recommendation in the context of an interact with smart speaker's conversational agents. And we also use measurement for level of privacy concerns to see individuals's level of privacy concerns affected the assessment. The result of this study as follow; First, all measurement variable are significantly related to 'Explanation facilities' Second, perceived trust, privacy risk are significantly related to individual's level of privacy concern. This study found that 'Explanation facilities' could be applied in context of smart speaker and possibility of cognitive dissonance according to the level of privacy concerns.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.10
• /
• pp.159-165
• /
• 2020

The demand for food safety has emerged as a major social issue due to changes in diet patterns and consumers' perceptions, along with the advancements in society and the development of the food industry. Consumers are demanding more information about the food they consume, and are sensitive to food scandals. With such interest in food safety, blockchain technology is attracting attention as a means of effectively responding to poor food management resulting in food fraud or unsafe distribution. By ensuring the accuracy of, and trust in, traceability in the food supply chain, it is possible to build trust between traders and to ensure safe food distribution. This paper proposes a next-generation agri-food distribution system that can provide and manage (for suppliers, consumers, and distribution officials) a variety of agri-food information, such as the history, distribution, safety, quality, and freshness of food. Information on product status and distribution status in all processes, including production, processing, distribution, sales, and consumption, can be monitored and controlled in real time (anytime, anywhere), and users can check the safety level of each type of food in real time through an app.

• Informatization Policy
• /
• v.19 no.2
• /
• pp.3-20
• /
• 2012

This study examines the trend and issues of sociological studies of social network service(SNS). For this purpose, the studies were classified into three types. The first type of study concerns spatial characteristics of SNS. It deals with the impact of such SNS usage on individual attitudes and behaviors. The second type of the study focuses on the formation process of interrelationship among users. It is based on sociological concepts such as trust and social capital. The last type of study emphasizes the impact of SNS on society. The study includes the issues of social movement and political participation. In the future study of SNS, more diverse sociological subjects, such as social integration, exclusion, public welfare, and affective support need to be dealt with.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.7 no.3
• /
• pp.151-162
• /
• 2011

This study defined acceptance of high-tech new product as user's active using the product such as utilizing application software rather than purchasing it. As exploring factors to effect on customer's acceptance, usefulness expectancy was examined from product side and application side. When investigating the exogenous variables to influence on usefulness expectancy from the product side, customer's product knowledge and social influences are put forward to support the hypothesis. From the application side, customer's knowledge about the application and its trust are put to explain usefulness expectancy of the application. Smart phone users were good subjects for this study and most hypotheses were tested using structural equation model and the results are followings. First usefulness expectancy of the product and of application significantly effect on customer's intention to use high-tech new product and also usefulness expectancy of the product positively effects on which of the application. Second customer's perceived knowledge about the product and social influences impact usefulness expectancy of the product. But third customer's perceived trust toward application didn't any positive impact usefulness expectancy of the application. Through the result, there will be several implications. First, from both of side; product and application, customer's usefulness expectancy should be satisfied to be successful in high-tech products market. Second, customer should be Ieant about advantages of the product and knowledge about the application as well, and then trigger their usefulness expectancy.