• Science of Emotion and Sensibility
• /
• v.25 no.3
• /
• pp.107-116
• /
• 2022

Purpose: Researchers have shown that aesthetic judgments of artworks depend on contexts, such as the authenticity of an artwork (Newman & Bloom, 2011) and an artwork's location of display (Kirk et al., 2009; Silveira et al., 2015). The present study aims to examine whether contextual information related to the creator, such as whether an artwork was created by a human or artificial intelligence (AI), influences viewers' preference judgments of an artwork. Methods: Images of Impressionist landscape paintings were selected as human-made artworks. AI-made artwork stimuli were created using Google's Deep Dream Generator by mimicking the Impressionist style via deep learning algorithms. Participants performed a preference rating task on each of the 108 artwork stimuli accompanied by one of the two creator labels. After this task, an art experience questionnaire (AEQ) was given to participants to examine whether individual differences in art experience influence their preference judgments. Results: Setting AEQ scores as a covariate in a two-way ANCOVA analysis, the stimuli with the human-made context were preferred over the stimuli with the AI-made context. Regarding the types of stimuli, the viewers preferred AI-made stimuli to human-made stimuli. There was no interaction effect between the two factors. Conclusion: These results suggest that preferences for visual artworks are influenced by the contextual information of the creator when the individual differences in art experience are controlled.

• Knowledge Management Research
• /
• v.25 no.2
• /
• pp.155-176
• /
• 2024

This study explores variations in preferences for human and AI agents within the medical and financial services. Study 1 investigates whether there are preferential disparities between human and AI agents across these service domains. It finds that human agents are favored over AI agents in medical services, while AI agents receive greater preference in the financial services. Study 2 delves into the underlying reasons for the preference differentials between human and AI agents by assessing the significance of certain capabilities as perceived by users in each domain. The findings reveal a mediating role of perceived empathy importance in the effect of service domains on human-AI preference. Furthermore, perceived empathy is deemed a more critical capability by users for preferring human over AI agents across both service domains compared to other capabilities such as experience and agency. This research is noteworthy for elucidating the variances in preferences for human and AI agents across medical and financial services and the rationale behind these differences. It enhances our theoretical comprehension of the pivotal factors influencing preferences for human and AI agents, underscoring the significance of human experiential capabilities like empathy.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.17 no.2
• /
• pp.167-177
• /
• 2022

Unlike previous studies on AI chatbot preference that focused mostly on satisfaction, this study considered both satisfaction and dissatisfaction. This study established that (1) AI chatbot preference is driven by attractive, must-be, and one-dimensional qualities, (2) AI chatbot need to develop service strategies by taking into account users' satisfaction and dissatisfaction in accordance with preference drivers, and (3) users view interaction as a requisite and thus, if they are not satisfied with services of a AI chatbot, they don't tend to appeal their opinion and leave the service with AI chatbot. This study emphasizes that a AI chatbot that desires to be a dominant market player must provide differentiated services according to the preference drivers and must continuously encourage user participation in order to improve service quality.

• Journal of Internet Computing and Services
• /
• v.23 no.3
• /
• pp.77-85
• /
• 2022

Unlike previous studies that focused on the diversity of YouTube content, this study tried to identify factors affecting users' satisfaction with the YouTube recommendation system. Specifically, by adding content preference suitability and privacy concerns to the technology acceptance model, we empirically analyzed how these variables affect user's satisfaction of the YouTube AI recommendation system. For this purpose, asurvey was conducted on college students in their 20s and 30s, and the main research results are as follows. First, in the respondents of this study, playfulness and usefulness, which are major variables of the technology acceptance model, appeared as significant factors affecting the satisfaction of the YouTube AI recommendation system, whereas the effect of ease to use was not found. Second, content preference suitability was found to affect the satisfaction with AI recommendation system, but privacy concerns did not affect the satisfaction with YouTube AI recommendation system. Based on these research results, the implications of the study and the directions for future studies were suggested.

• Journal of The Korean Association of Information Education
• /
• v.25 no.3
• /
• pp.471-481
• /
• 2021

The purpose of this study is to analyze the effectiveness of AI convergence education program. To this end, the "Elementary Science AI Convergence Education Program for Machine Learning" developed in previous research were taught to elementary school students in the fourth to sixth grades in eight times. The quantitative changes of each factor were analyzed by R program, and the effectiveness of education was analyzed by Pearson correlation and paired samples t-test. As a result, there is a deep correlation between "Attitude to AI technology, Scientific preference and STEAM Literacy" and technical average has improved in many factors. Therefore, AI convergence education program is meaningful in terms of education, and if AI education and AI convergence education are implemented into the primary formal education curriculum, they will have a positive effect.

• Annual Conference on Human and Language Technology
• /
• 2024.10a
• /
• pp.89-93
• /
• 2024

본 논문에서는 사전학습된 대규모언어모델인 Llama-2 모델을 한국어, 영어 교차 언어 (Cross-lingual) 모델로 효율적으로 전이하는 선호도 최적화 (Preference Optimization) 기법인 X-Opt(Cross-Optimization)를 제안한다. 제안된 기법은 영어에서 1600개의 시드 데이터만을 사용하여 교차 언어전이 환경에서 한국어의 성능을 큰 폭으로 향상시키며, Supervised Fine-tuning 방법과 DPO, ORPO로 학습된 모델을 큰 차이로 능가하는 것으로 나타났다. 또한, 자연어 처리 벤치마크에서도 SFT 모델 대비 정렬 세금(alignment tax)의 영향을 거의 받지 않는 것으로 나타났다. 제안하는 방법은 참조 모델 없이 보상을 최적화하고 소수의 영어 데이터를 사용한다는 점에서, 낮은 컴퓨팅 환경과 데이터 구축 비용 없이 English-centic 모델을 cross-lingual 모델로 전이할 수 있는 효율적인 방법이다.

• Environmental and Resource Economics Review
• /
• v.34 no.1
• /
• pp.27-54
• /
• 2025

This paper analyzes the impact of representing visual alternatives on survey respondents' choices and preferences in valuing non-market goods. The survey was divided into two groups for the ecological river restoration project: a text-only group and a virtual photo and text group. The estimation results show that respondents express stronger preferences for environmental attributes when they are presented with virtual images and text. In particular, the estimated willingness to pay for water quality, biodiversity, and recreational uses was significantly higher than in the text-only group. These results suggest that traditional text-only stated preference methods may lead to an underestimation of willingness to pay. We also explored the potential application of generative AI in creating visual representations. This paper demonstrates that providing visual aids can mitigate the limitations of the stated preference approach and enhance the reliability of data used to estimate the value of environmental goods.

• Journal of the Korean Home Economics Association
• /
• v.27 no.3
• /
• pp.71-78
• /
• 1989

As the level of life improves, the eating habit is changing from rice meal to bread meal and at the time, eat more strawberry jam than before. We tried to study to select the good cooking method and the proper strawberry variety for the jam through the sensory evaluation We made four kinds jam of Bogyo-Joseoung and Ai-berry by different cooking methods, the result6s of the sensory evaluation are as follow: The jam of Ai-berry is better than that of Bogyo-Joseoung by the paired comparison test but the difference between those, if we add some lemon to the jam of Bogyo-Joseoung and Ai-berry, is very little. The multiple comparison test proves the sourness, if added some lemon and citric acid, become better and the color and viscosity, if added pectin, became better. The overall preference about jam, if pectin and citric acid were added together, was best. In addition, we evaluated the quality of jammed bread by multiple comparison test. The result is like this: The jam with lemon is very good in color, flavor, sourness and texture, but the jam with pectin and citric acid was the best in overall preference.

• Journal of Korean Elementary Science Education
• /
• v.39 no.1
• /
• pp.117-132
• /
• 2020

The purpose of this study is to investigate elementary school teachers' awareness of Artificial Intelligence (AI) and find out how to apply it in elementary science education. The survey was conducted online and involved 95 teachers working in the metropolitan area. The results of this study are as follows. First, teachers need to learn about the general characteristics of AI and how to apply it to education. Second, science classes had the highest preference for AI among elementary school subjects. Third, the preference for AI application by elementary science field was 68.4% for earth and space, 54.7% for exercise and energy, 32.6% for matter, 27.4% for life. Fourth, AI-based Science Education (AISE) teaching- learning strategies were developed based on AI characteristics and the changing perspective of elementary science education, AISE's teaching-learning strategies are five: 'automation', 'individualization', 'diversification', 'cooperation' and 'creativity' and teachers can use them in teaching design, class practice and evaluation stages. Finally, the creative problem-solving Doing Thinking Making Sharing (DTMS) model was devised to implement the creativity strategy in AISE. This model consists of four-steps teaching courses: Doing, Thinking, Making and Sharing based on the empirical learning theory. In the future, follow-up research is needed to verify the effectiveness of this model by applying it to elementary science education.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.57 no.5
• /
• pp.607-611
• /
• 2024

In this study, the AI-based photo-identification program IBEIS was used to efficiently analyze the photo-identification of spotted seals Phoca largha, around Baengnyeongdo Island, Korea. A total of 3,640 photographs collected during field surveys from 2016 to 2021 were analyzed, leading to the identification of 1,234 seals. The spotted seal population around Baengnyeongdo Island was estimated to be at least 1,200 individuals. The photo-identification method relies on the unique black spot patterns on the left cheek of the seals, which remain consistent over time. This method allows for long-term ecological studies and tracking of individual seals. In addition, site preferences of the spotted seals were analyzed, revealing that more than 70% of individuals were re-identified at the site where they were first observed and indicating site preference. The results suggest that environmental and ecological factors may influence habitat preferences. Future studies, including collaborative efforts with neighboring countries such as North Korea and China, will provide more comprehensive insights into the habitat usage patterns of the species. This study offers a foundation for estimating the population size of spotted seals and key data for habitat management and conservation strategies by using AI-based photo identification methods.