• 산경연구논집
• /
• 제11권12호
• /
• pp.29-38
• /
• 2020

Purpose: The purpose of this study is to empirically analyze the effect of chatbot service quality, chatbot trust, and chatbot satisfaction on chatbot reuse intention and store reuse intention. Research design, data, and methodology: We reviewed the literature on domestic and international chatbots, established hypotheses, and analyzed them. We empirically analyzed the process model in which chatbot service quality (interaction quality, information quality) has a positive effect on chatbot trust and chatbot satisfaction, and that chatbot trust and satisfaction positively affect chatbot reuse intention and store reuse intention. A survey was conducted on 212 people who had used shopping mall chatbots and financial service chatbots after demonstrating the shopping mall chatbot video. Structural equation modeling was conducted by using AMOS 24.0 to test the proposed relationships. Results: As a result of the empirical analysis, the effects of interaction quality on chatbot trust and information quality on chatbot satisfaction were not supported, but the rest of the hypotheses were statistically significant. It was found that the information quality of chatbot service had a positive effect on chatbot trust, but did not significantly affect chatbot satisfaction. In addition, the interaction quality of the chatbot positively affects the satisfaction of the chatbot, but it does not significantly affect the trust of the chatbot. Chatbot trust was found to have a positive effect on chatbot satisfaction. Chatbot trust and chatbot satisfaction were found to have a positive influence on the intention to reuse the chatbot. And, chatbot trust and chatbot satisfaction were found to have a positive influence on store reuse intention. Conclusions: The findings of this study offer significant theoretical and managerial contributions in the context of chatbot. Chatbots should enhance customer contact quality management from the perspective of total customer experience management rather than partial function. When providing a chatbot service, it is more desirable to give priority to providing accurate information to increase trust, and at the same time to improve customer satisfaction by increasing the quality of interaction. And in order to increase the competitive advantage of companies, the purpose of introducing chatbots should be clarified and approached strategically.

• 반도체디스플레이기술학회지
• /
• 제21권4호
• /
• pp.14-17
• /
• 2022

Due to the development of IT technology and the on-going Coronavirus disease, non-face-to-face services have been activated. To overcome the inconvenience of non-face-to-face service, service providers have adopted chatbots as a way to feel like a human being. As the increasing chatbot services, chatbot builders have emerged, which can help non-developers to build them. Although its popularity has increased, its performance evaluation has not been conducted on such chatbot builders. In this paper, we implement a prototype chatbot that classifies hospital departments in the medical field using Dialogflow and Rasa, which are popular chatbot builders. By measuring the accuracy of the chatbot's classification of medical subjects, we evaluated the level of accuracy that the most used chatbot builder can have when they are used to build a chatbot service. The simulation results showed that Dialogflow had 87%, 65%, and 60%, and Rasa did 64%, 70%, and 63% in surgery dermatology, and otolaryngology, respectively.

• 한국컴퓨터정보학회논문지
• /
• 제26권12호
• /
• pp.53-59
• /
• 2021

본 연구에서는 '레플리카'와 같은 텍스트 입력 기반의 부정적 감정 완화가 가능한 국내 인공지능 챗봇인 BERGPT-chatbot을 제안하고자 한다. BERGPT-chatbot은 KR-BERT와 KoGPT2-chatbot을 파이프라인으로 만들어 감정 완화 챗봇을 모델링하였다. KR-BERT를 통해 정제되지 않은 일상 데이터셋에 감정을 부여하고, 추가 데이터셋을 KoGPT2-chatbot을 통해 학습하는 방식이다. BERGPT-chatbot의 개발 배경은 다음과 같다. 현재 전 세계적으로 우울증 환자가 증가하고 있으며, 이는 COVID-19로 인해 장기적 실내 생활이나 대인 관계 제한으로 더욱 심각한 문제로 대두되었다. 그로 인해 부정적 감정 완화나 정신 건강 케어에 목적을 둔 국외의 인공지능 챗봇이 팬데믹 사태로 사용량이 증가하였다. 국내에서도 국외의 챗봇과 비슷한 심리 진단 챗봇이 서비스 되고 있으나, 국내의 챗봇은 텍스트 입력 기반 답변이 아닌 버튼형 답변 중심으로 국외 챗봇과 비교하였을 때 심리 진단 수준에 그쳐 아쉬운 실정이다. 따라서, BERGPT-chatbot을 통해 감정 완화에 도움을 주는 챗봇을 제안하였으며, BERGPT-chatbot과 KoGPT2-chatbot을 언어 모델의 내부 평가 지표인 '퍼플렉서티'를 통해 비교 분석하여 BERGPT-chatbot의 우수함을 보여주고자 한다.

• 한국의류산업학회지
• /
• 제24권2호
• /
• pp.229-240
• /
• 2022

This study explores the effect of anthropomorphism on fashion chatbot reliability, mediated by perceived intelligence and cognitive evaluation. The moderating effects of individuals' need for human interaction between chatbot anthropomorphism and perceived intelligence, cognitive evaluation, and chatbot reliability are also explored. Participants, who were recruited through the online research firm, responded to questions after watching a video clip showing a conversation with a fashion chatbot on a mobile screen. The data were collected through Mturk, a crowdsourcing platform with an online research panel. All responses (N = 212) were analyzed using SPSS 26.0 for the descriptive statistics, frequency analysis, reliability analysis, exploratory factor analysis, and PROCESS procedure. The results demonstrate that chatbot anthropomorphism increases chatbot reliability, and this is mediated by chatbot intelligence. Although chatbot anthropomorphism increases cognitive evaluation, the effect of cognitive evaluation on chatbot reliability is not significant; thereby, the effect of chatbot anthropomorphism on chatbot reliability is not mediated by the cognitive evaluation. The direct effect of anthropomorphism on chatbot reliability is also moderated by individuals' need for human interaction. For participants with a high need for human interaction, chatbot anthropomorphism increases chatbot reliability; however, anthropomorphism does not significantly affect chatbot reliability for participants with a low need for human interaction. The study's findings contribute to expanding the literature on consumers' new technology acceptance by testing the antecedents affecting service reliability.

• 패션비즈니스
• /
• 제25권6호
• /
• pp.119-130
• /
• 2021

This study investigated the effects of the chatbot's level of anthropomorphism - closeness to the human form - and its self-disclosure - delivery of emotional exchange with the chatbot through its facial expressions and chatting message on the user's intention to accept the service. A 2 (anthropomorphism: High vs. Low) × 2 (self-disclosure through facial expressions: High vs. Low) × 2 (self-disclosure through conversation: High vs. Low) between-subject factorial design was employed for this study. An online survey was conducted and a total of 234 questionnaires were used in the analysis. The results showed that consumers used chatbot service more when emotions were disclosed through facial expressions, than when it disclosed fewer facial expressions. There was statistically significant interaction effect, indicating the relationship between chatbot's self-disclosure through facial expression and the consumers' intention to use chatbot service differs depending on the extent of anthropomorphism. In the case of "robot chatbots" with low anthropomorphism levels, there was no difference in intention to use chatbot service depending on the level of self-disclosure through facial expression. When the "human-like chatbot" with high anthropomorphism levels discloses itself more through facial expressions, consumer's intention to use the chatbot service increased much more than when the human-like chatbot disclosed fewer facial expressions. The findings suggest that chatbots' self-disclosure plays an important role in the formation of consumer perception.

• 산경연구논집
• /
• 제13권8호
• /
• pp.29-50
• /
• 2022

Purpose - This study aims to investigate factors affecting Chatbot service acceptance attitude. For wide use of Chatbot service, firms need to find barriers or obstacles for customers, if any, not to use Chatbot service. Research design, data, and methodology - We apply value-based accept model to investigate the quality of Chatbot, to verify the meaning of service value of Chatbot and to find the relationship among variables. To test hypotheses, we conducted survey. We collected 300 questionnaires. SPSS version 2.0 is used. Regression analysis, moderating effect test is conducted. Results - 4 Qualities of Chatbot, Ease of use, Usefulness, Enjoyment, Interaction are affecting acceptance attitude, and 5 service values, only interaction does not affect emotion. Trust, Specialty, Necessity, Social, Emotion moderating Chatbot service to accepting attitude. Regarding moderating effects by personal characteristics and personal tendency, innovation resistance, innovativeness, and social effects are turned to have influence while regulatory focus, construal level does not have moderating force. Also, the auxiliary service like Chatbot service affects customers' evaluation on the main service quality. Conclusions - Service firms adopt Chatbot service for various purposes. The results imply that customers are generally recognize the merits of Chatbot, but there are some barriers such as innovation resistance characteristic especially uncomfortable.

• 한국IT서비스학회지
• /
• 제19권4호
• /
• pp.31-47
• /
• 2020

Recently, as the COVID-19 has spread and prolonged worldwide, the 'Untact' society is becoming routinized, and various smart technologies are leading to the spread of the 'Ontact' culture. This is because the desire of consumers to purchase a product and use the service has increased while minimizing the direct contact. In order to quickly respond to this circumstance, the percentage of the companies which are adopting Chatbot in various fields such as orders, delivery, and inquiries is increasing and they are getting a positive result. However as the demand for building Chatbot increases dramatically, there are many confusions among the companies which want to introduce Chatbot to their system, due to the lack of professional technicians and difficulties in understanding AI technologies and how to build them effectively. I believe that in the post COVID-19 era, much more companies will adopt Chatbot, and this will intensify the problem. The purpose of this study was to derive the needs for a guide on the method of buiilding a Chatbot through considering the prior research on Chatbot and analysis of the recent surge in the use of Chatbot services related to COVID-19. There are implications to presenting 5 phases of universal Chatbot implementation methodology using the platform to the stakeholders who want to introduce Chatbot to their customer so that they can understand and build Chatbot more easily and use AI Chatbot actively in response to the POST COVID-19 era.

• International Journal of Contents
• /
• 제16권4호
• /
• pp.1-15
• /
• 2020

This study examined the impact of chatbot service quality (process quality, outcome quality, and servicescape quality) on user satisfaction and reliability by identifying the relationships between user satisfaction, reliability, immersion, and the paths of three variables influencing reuse intention. The survey was conducted of Korean users in their teens and 70s who had experience using chatbot services. A total of 218 convenience samples were extracted and the data analyzed. By the IS success and SERVQUAL model, the results of structural equation modeling revealed that the chatbot service quality did not affect user satisfaction and reliability. However, user satisfaction and reliability of the chatbot services were shown to lead to reuse intention, and user satisfaction was shown to affect immersion and immersion in reliability. The results showed that satisfaction, reliability, and immersion in the chatbot services were important factors in the chatbot reuse intention. Through the satisfaction and reliability gained through the service, the users wanted to reuse the chatbot services, especially the chatbot services that gained reliability, which will have a greater impact on reuse intention. We can use these results as marketing information to attract loyal customers by identifying the reuse intention of the chatbot service users.

• 한국의류산업학회지
• /
• 제24권2호
• /
• pp.183-194
• /
• 2022

This study aims to understand the consumers' negative responses to communication failure of chatbots caused by their imperfections. Specifically, this study examines 1) the relationship among chatbot's communication failure, dissatisfaction, negative behavior (complaint, negative word-of-mouth (nWOM), and inertia); 2) the moderating effect of technostress on the relationship between chatbot's communication failure and dissatisfaction; 3) the differences in the negative responses between the generation MZ and the previous generations. Data were collected via an online survey. First, the participants interacted with the chatbot developed for this survey, to experience the chatbot's communication failure. Thereafter, they responded to a questionnaire. PLS-SEM was conducted using the R software environment to test the hypotheses. This study empirically identified that chatbot's communication failure positively affected dissatisfaction. In addition, the customers who were more dissatisfied with the chatbot's communication failures were more likely to complain than engage in nWOM. Compared to the generation MZ, chatbot's communication failure caused a higher level of dissatisfaction in previous generations. The results suggest that online shopping malls should carefully introduce an improved chatbot service after minimizing its communication failure rate. The chatbot developers of online shopping malls targeting middle-aged and elderly consumers should strive to develop and implement strategies to further alleviate consumers' dissatisfaction in the situation of chatbot's communication failure.

• 한국IT서비스학회지
• /
• 제17권3호
• /
• pp.17-36
• /
• 2018

In this study, we developed a chatbot (Dialogue agent) using small Q & A data and evaluated its performance. The chatbot developed in this study was developed in the form of an FAQ chatbot that responds promptly to customer inquiries. The development of chatbot was conducted in three stages : 1. Analysis and planning, 2. Content creation, 3. API and messenger interworking. During the analysis and planning phase, we gathered and analyzed the question data of the customers and extracted the topics and details of the customers' questions. In the content creation stage, we created scenarios for each topic and sub-items, and then filled out specific answers in consultation with business owners. API and messenger interworking is KakaoTalk. The performance of the chatbot was measured by the quantitative indicators such as the accuracy that the chatbot grasped the inquiry of the customer and correctly answered, and then the questionnaire survey was conducted on the chatbot users. As a result of the survey, it was found that the chatbot not only provided useful information to the users but positively influenced the image of the pension. This study shows that it is possible to develop chatbots by using easily obtainable data and commercial API regardless of the size of business. It also implies that we have verified the validity of the development process by verifying the performance of developed chatbots as well as an explicit process of developing FAQ chatbots.