• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.6
• /
• pp.1137-1144
• /
• 2022

In this study, an artificial intelligence(AI) was developed to help with facial expression practice in order to express emotions. The developed AI used multimodal inputs consisting of sentences and facial images for deep neural networks (DNNs). The DNNs calculated similarities between the emotions predicted by the sentences and the emotions predicted by facial images. The user practiced facial expressions based on the situation given by sentences, and the AI provided the user with numerical feedback based on the similarity between the emotion predicted by sentence and the emotion predicted by facial expression. ResNet34 structure was trained on FER2013 public data to predict emotions from facial images. To predict emotions in sentences, KoBERT model was trained in transfer learning manner using the conversational speech dataset for emotion classification opened to the public by AIHub. The DNN that predicts emotions from the facial images demonstrated 65% accuracy, which is comparable to human emotional classification ability. The DNN that predicts emotions from the sentences achieved 90% accuracy. The performance of the developed AI was evaluated through experiments with changing facial expressions in which an ordinary person was participated.

• Journal of The Korean Association of Information Education
• /
• v.26 no.1
• /
• pp.65-73
• /
• 2022

In this study, by using Entry text model learning, educational contents for artificial intelligence education of elementary school students are developed and applied to actual classes. Based on the elementary and secondary artificial intelligence content table, the achievement standards of practical software education and artificial intelligence education will be reconstructed.. Among text, images, and sounds capable of machine learning, "production of emotion recognition programs using text model learning" will be selected as the educational content, which can be easily understood while reducing data preparation time for elementary school students. Entry artificial intelligence is selected as an education platform to develop artificial intelligence education contents that create emotion recognition programs using text model learning and apply them to actual elementary school classes. Based on the contents of this study, As a result of class application, students showed positive responses and interest in the entry AI class. it is suggested that quantitative research on the effectiveness of classes for elementary school students is necessary as a follow-up study.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.10
• /
• pp.3498-3512
• /
• 2021

This article describes how events that make up text stories can be represented and extracted. We also address the results from our simple experiment on extracting and clustering events in terms of emotions, under the assumption that different emotional events can be associated with the classified clusters. Each emotion cluster is based on Plutchik's eight basic emotion model, and the attributes of the NLTK-VADER are used for the classification criterion. While comparisons of the results with human raters show less accuracy for certain emotion types, emotion types such as joy and sadness show relatively high accuracy. The evaluation results with NRC Word Emotion Association Lexicon (aka EmoLex) show high accuracy values (more than 90% accuracy in anger, disgust, fear, and surprise), though precision and recall values are relatively low.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.367-368
• /
• 2023

최근 챗봇 시스템은 급격한 발전과 함께 사용자와 자연스러운 대화를 할 수 있는 인공지능 기술의 필요성이 대두되고 있다. 기존의 챗봇 시스템은 대화 상황을 충분히 이해하지 못하거나, 학습된 데이터를 벗어나는 문장에 대한 일관성 있는 응답을 제공하지 못하는 한계가 있다. 본 논문에서는 GPT-3와 KoBERT를 활용하여 사용자의 감정 상태를 파악하고 해당 감정을 고려한 일관성 있는 대화를 제공하는 감정 분석 기반 챗봇 시스템을 제안한다. 이를 바탕으로 긍정적인 대화를 이어 나가는데 초점을 두어 자연스러운 대화가 가능할 것으로 기대된다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1204-1205
• /
• 2023

본 논문은 소비자의 리뷰 데이터를 기반으로 한 새로운 감성 분석 방법을 제안한다. 긍정, 부정, 중립으로 분류하는 전통적 감성 분석방법은 텍스트에 나타난 감정의 섬세한 차이를 파악하기 어렵다. 이에 본 연구에서는 GPT 모델을 사용하여 텍스트에서 사용자의 감정을 8 가지의 카테고리로 세분화한다. 부정적 정서를 가진 리뷰에서 분노, 혐오, 실망과 같은 구체적인 감정들을 직관적으로 파악할 수 있었고, 감정의 강도까지 파악할 수 있었다. 제안된 방법을 통해 기업은 고객의 요구 사항을 정확하게 인지할 수 있으며, 고객 맞춤형 서비스 개선에 기여할 수 있다는 점이 기대된다.

• Journal of Service Research and Studies
• /
• v.14 no.1
• /
• pp.77-90
• /
• 2024

This research aims to investigate methods of leveraging generative artificial intelligence in service sectors where consumer sentiment and experience are paramount, focusing on minimizing hallucination phenomena during usage and developing strategic services tailored to consumer sentiment and experiences. To this end, the study examined both mechanical approaches and user-generated prompts, experimenting with factors such as business item definition, provision of persona characteristics, examples and context-specific imperative verbs, and the specification of output formats and tone concepts. The research explores how generative AI can contribute to enhancing the accuracy of personalized content and user satisfaction. Moreover, these approaches play a crucial role in addressing issues related to hallucination phenomena that may arise when applying generative AI in real services, contributing to consumer service innovation through generative AI. The findings demonstrate the significant role generative AI can play in richly interpreting consumer sentiment and experiences, broadening the potential for application across various industry sectors and suggesting new directions for consumer sentiment and experience strategies beyond technological advancements. However, as this research is based on the relatively novel field of generative AI technology, there are many areas where it falls short. Future studies need to explore the generalizability of research factors and the conditional effects in more diverse industrial settings. Additionally, with the rapid advancement of AI technology, continuous research into new forms of hallucination symptoms and the development of new strategies to address them will be necessary.

• International Journal of Advanced Culture Technology
• /
• v.6 no.3
• /
• pp.117-128
• /
• 2018

This study attempts to derive the possibility of literary therapy through addition and subtraction of emotional codons. It is presumed that the remnants of the emotions formed by the addition and subtraction of emotions will remain in the human body and cause chemical reactions. When this research is activated, cluster of emotional codons will be created by a combination of literary emotions. This is expected to accelerate therapeutic action of sentences by encoding certain emotional codons in AI.

• Science of Emotion and Sensibility
• /
• v.22 no.2
• /
• pp.49-58
• /
• 2019

In the era of the 4th Industrial Revolution, artificial intelligence (AI) has become one of the core technologies in terms of the business strategy among information technology companies. Both international and domestic major portal companies are launching AI search services. These AI search services utilize voice, images, and other unstructured data to provide different experiences from existing text-based search services. An unfamiliar experience is a factor that can hinder the usability of the service. Therefore, the usability testing of the AI search services is necessary. This study examines the usability of the AI search service on the Naver App 8.9.3 beta version by comparing it with the search services of the current Naver App and targets 30 people in their 20s and 30s, who have experience using Naver apps. The usability of Smart Lens, Smart Voice, Smart Around, and AiRS, which are the Naver App beta versions of their artificial intelligence search service, is evaluated and statistically significant usability changes are revealed. Smart Lens, Smart Voice, and Smart Around exhibited positive changes, whereas AiRS exhibited negative changes in terms of usability. This study evaluates the change in usability according to the application of the artificial intelligence search services and investigates the correlation between the evaluation factors. The obtained data are expected to be useful for the usability evaluation of services that use AI.

• Journal of Digital Convergence
• /
• v.18 no.7
• /
• pp.101-110
• /
• 2020

Devices that add emotion-specific services or various functions are appearing in AI speakers and related devices. To this end, this study performed topic modeling analysis on the topics of post-purchase texts written by AI speaker users, and compared them with the data collected via survey questionnaires. Furthermore, data on the emotional intelligence of AI speakers and relationship quality were collected from 600 users and analyzed using structural equation modeling. The findings of the study are as follows: First, the analysis results of topic modeling showed that most of the articles mainly mention the functional aspects of AI speakers. Second, emotional intelligence of AI speaker perceived by consumer affected relationship quality, and relationship quality had a positive effect on customer satisfaction. Therefore, this study expands the area of AI research by integrating the concept of emotional intelligence and relationship quality to provide new theoretical and practical implications.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.4
• /
• pp.107-112
• /
• 2024

One More Pumpkin won the grand prize at the 2023 Dubai International AI Film Festival, and new possibilities were also opened through the International AI and Metaverse Film Festival (GAMFF), which was held for the first time in Korea. Generative works began to stand out in earnest, with 527 diverse works from 42 countries at home and abroad using AI and metaverse technology submitted to this contest. AI is being used in a variety of fields, including the creation and implementation of digital characters through combination with VFX, improving the efficiency of video production, and managing the overall video production process. This contributes to saving human and material resources required for production and significantly improving the quality of produced videos. However, generative AI also has ambiguity in copyright attribution, ethical issues inherent in the learned dataset, and technical limitations that fall short of the level of human emotion and creativity. Accordingly, this study suggests implications at the level of production, screening, and use, as generative AI may have an impact in more areas in the future.