SNS 온라인 리뷰를 활용한 TV프로그램 품질평가 연구

윤상혁(Sang-Hyeak Yoon),손지현(Ji-Hyun Shon),고민삼(Min-Sam Ko),김영걸(Young-Gul Kim) 한국방송학회 2015 방송통신연구 Vol.- No.90

Joint Sentiment 토픽모델링 기반 국내 여행 불만족 요인 연구

최윤진(Yoon-Jin Choi),이소현(So-Hyun Lee),윤상혁(Sang-Hyeak Yoon),김희웅(Hee-Woong Kim) 한국경영학회 2020 Korea Business Review Vol.24 No.2

최근 개인의 삶의 질, 일과 삶의 균형을 중요하게 생각하게 되면서 여행 활동이 증가하고 있다. 이에 따라 국내여행 산업도 성장하고 있으나 동시에 여행지에 따른 불만족에 대한 개선은 과제로 남아있다. 본 연구는 국내 여행에 대한 실제 여행자들의 주요 여행지별 불만족 요인을 도출하고 이에 대한 개선방안을 제시하는 것을 목적으로 한다. 소비자의 만족과 불만족의 개념을 설명하는 기대 불일치 이론을 기반으로 텍스트마이닝 기법과 Joint Sentiment 토픽모델링을 활용하여 실제 여행자들이 사용하는 여행 애플리케이션에서 국내 주요 여행지별 불만족 사항을 도출 및 분석하고, 이를 기반으로 여행지별 불만족 사항의 해결 및 개선방안을 제시하였다. 연구 결과, 첫째, 국내 요인 전반에 대한 불만족 요인은 ‘물가, 교통, 바가지요금, 위생, 상업화’에 대한 토픽이었다. 둘째, 주요 여행지별 토픽으로 ‘제주도’는 ‘교통, 외국인 관광객, 물가, 특색 없는 명소’, ‘전주’는 ‘먹거리 중심, 바가지요금, 상업화’, ‘부산’은 ‘호객행위, 위생, 바가지요금, 교통’, ‘경주’는 ‘주차, 바가지요금, 관광객, 교통’이 각각 도출되었다. 텍스트마이닝 기법의 토픽모델링에 감성 차원을 더한 Joint Sentiment 토픽모델링을 사용하여 국내 주요 여행지별 불만족 요인을 도출하였고 토픽과 키워드 간의 관계를 네트워크 다이어그램으로 시각화하여 직관적인 이해와 차별적인 개선방안을 도출한 것에 학술적 의의가 있다. 실무적으로는 여행지별 불만족 사항에 대한 구체적인 개선방안을 제안함으로써 국내 여행의 활성화를 통한 관광 산업의 수익 및 경쟁력 향상에 이바지할 수 있다. In recent years, travel has increased due to the interest in the quality of life and work-life balance of individuals. As a result, the domestic travel industry is growing, but at the same time, improving domestic travel dissatisfaction remains a challenge. The purpose of this study is to derive the factors of dissatisfaction of major domestic tourists for domestic travel and to suggest ways to improve them. Based on the expectancy disconfirmation theory to explain the concept of consumer satisfaction and dissatisfaction by collecting the dissatisfaction of actual travelers, we analyzed using text mining technique and joint sentiment topic modeling. This study draws out and analyzes dissatisfaction points of major domestic travel destinations and suggests ways to resolve and improve dissatisfaction factors by travel destinations. First, dissatisfaction with the overall domestic travel factors was the topic of ‘price, transportation, rip off, sanitation, commercialization’. Second, the topics by major destinations are as follows. ‘Jeju Island’: ‘traffic, foreign tourists, prices, uncharacteristic spots’, ‘Jeon Ju’: ‘parking, rip off, food-oriented spots, commercialization’. ‘Busan’: ‘tout, sanitation, rip off, transportation’, ‘Gyeon Ju’: ‘parking, rip off, tourist, transportation’. Text Mining Technique Using Joint Sentiment Topic Modelling, which adds emotional dimension to topical modelling, we derived dissatisfaction factors, by major travel destinations in Korea and visualized the relationship between topics and keywords in a network diagram to enable intuitive understanding and differentiated improvement plans. This is the significance of this study. In practice, this study proposes a concrete improvement plan for the dissatisfaction point of each destination. This can contribute to improving the profits and competitiveness of the tourism industry by activating domestic travel.

브이튜버(Vtuber) 개인방송의 기술적 특성과 가상 크리에이터 특성이 즐거움, 시청만족도 및 유료후원의도에 미치는 영향: S-O-R 모델을 기반으로

김성군,양성병,윤상혁,Jin, Chengjun,Yang, Sung-Byung,Yoon, Sang-Hyeak 한국IT서비스학회 2022 한국IT서비스학회지 Vol.21 No.5

Personal broadcasting utilizing Vtuber, a virtual creator made of 2D or 3D avatars, has recently appeared and is growing in popularity. Vtuber is a virtual person who broadcasts on the Internet using 2D or 3D avatars with real-time motion capture and computer graphics technologies. While the personal broadcasting industry utilizing Vtuber is proliferating, related studies have mainly concentrated on technical issues. Therefore, in this study, the antecedent factors that form the technical characteristics and virtual creator characteristics of Vtuber personal broadcasting are derived using the Stimulus-Organism-Response (S-O-R) model. Then the effect of these factors on viewer pleasure and satisfaction, which lead to increased paid sponsorship is to be examined. Furthermore, we investigate how this influencing mechanism fluctuates based on the avatar type (2D vs. 3D). This study contributes to empirical examinations of viewers' paid sponsorship intention in Vtuber personal broadcasting through the S-O-R model. It also offers insights that technological or virtual creator characteristics could improve viewers' pleasure, satisfaction, and even paid sponsorship.

콘텐츠 창작자들의 NFT 시장 참여에 대한 긍·부정 요인 연구

양지훈(Ji Hoon Yang),윤상혁(Sang-Hyeak Yoon) 한국IT서비스학회 2022 한국IT서비스학회지 Vol.21 No.4

NFTs, which guarantee ownership of digital files using blockchain technology, are the new field for the content industry. The NFT provided new opportunities for content creators to trade digital contents without going through mediation freely. Additionally, collectors and investors can safely and easily own their works without the threat of illegal copies. However, since only a limited number of content creators are participating in the NFT market, there needs to be an influx of various content creators and a process of popularization for this market to grow and develop into the main stage. Furthermore, research on NFT has been limited, and understanding the drivers of creators choosing to participate in NFT is insufficient. Thus, this study aims to identify the factors affecting content creators participating in NFT by applying a mixed-methods approach and presenting practical implications. Using topic modeling and in-depth interviews, this study derives the positive and negative factors and suggests strategies to activate content creators' participation in the NFT market. Through this, we can guide that management implication to reduce the risks and costs of participating in NFTs is needed to encourage the participation of creators. It will also provide insight into ways to develop the NFT content market.

머신러닝 기반의 보상형 크라우드펀딩 성공 예측 모델링

문동지(Dong-Ji Moon),윤상혁(Sang-Hyeak Yoon),최수빈(Soobin Choi),김희웅(Hee-Woong Kim) 한국경영학회 2020 Korea Business Review Vol.24 No.3

크라우드펀딩은 최근 자금 조달 경로로 이용되며 소셜미디어와의 접목을 통해 빠르게 성장하고 있다. 2018년 기준 세계 크라우드펀딩 규모는 93억 7천만 달러, 한국 크라우드펀딩 시장은 1.1억 달러로 추정된다. 그러나 국내 크라우드펀딩 실패 확률은 2019년 기준으로 38%에 달하며, 펀딩 프로젝트가 실패할 경우 참여자(창설자, 투자자, 플랫폼) 모두에게 큰 부담이 된다. 만약 프로젝트 초기에 펀딩 성공 여부를 예측할 수 있다면 시간적 금전적 손해를 예방할 수 있다. 이에 본 연구는 국내 크라우드펀딩을 대상으로 성공 여부를 예측하는 모델을 구축하고자 한다. 기존 연구들 대부분은 펀딩 프로젝트가 끝난 후의 데이터를 사용했지만, 본 연구에서는 크라우드펀딩 사이트 와디즈의 펀딩 초기인 7일 이내의 댓글 데이터와 펀딩 참여 건수을 수집하여 예측 변수로 사용하였다. 예측 모델링 기법은 Decision Tree, SVM, Naive Bayes, AdaBoost, Gradient Boosting, Random Forest, MLP와 같은 머신러닝 알고리즘을 활용하였다. 예측 결과 Gradient Boosting이 90% 넘는 정확도를 보였고, Support Vector Machine이 가장 높은 정밀도(Precision, 0.95)를 보였다. 본 연구는 머신러닝 기반의 예측 모델을 개발함으로써, 크라우드펀딩 초기 단계에서 펀딩 성공 여부를 예측할 수 있다는 실무적 의의가 있다. Crowdfunding has been recently rising as financing channel and showed rapid growth by integrating with social media. As of 2018, global crowdfunding market size was estimated as $ 9.37 billion, and Korea crowdfunding market size was about $ 110 million. However, the probability of crowdfunding failure showed more than 38%, which gives huge burden for participants (i.e., makers, investors, platforms). So, to prevent the failure and protect participants from their loss of time and money, predicting the success of the funding in the early step is crucial. Therefore, this study aims to build a model to predict whether the crowdfunding project will success or fail. Compare to the previous studies that they used data after the end of crowdfunding, we collected data seven days before the project ends. We used data from crowdfunding site ‘Wadiz’, by collecting comment data and funding information as predict variable. Then we applied machine learning methods such as Decision Tree, Support Vector Machine, Naive Bayes, AdaBoost, Gradient Boosting, Random Forest, and MLP. As a result, Gradient Boosting showed more than 90% accuracy, and Support Vector Machine showed the highest precision score (0.95). Also, this study has a practical implication of predicting funding success in the early stage of crowdfunding by developing a prediction model based on machine learning.

Heaven or Hell for Artists in NFT : A Social Exchange Theory Perspective

Yang, Ji-Hoon (양지훈),Yoon, Sang-Hyeak(윤상혁) 한국지능정보시스템학회 2022 한국지능정보시스템학회 학술대회논문집 Vol.2022 No.6

NFTs, which guarantee ownership of digital files by using blockchain technology, are the new field for digital art in the art world. The NFT provided new opportunities for artists to freely trade digital artworks without going through auctions or galleries. Additionally, collectors and investors can safely and easily own their works without the threat of illegal copy. Since only a limited number of artists are participating in the NFT market, there needs to be an influx of various artists and a process of popularization for this market to grow and develop into the main stage. However, research on NFT has been limited, and the understanding of the drivers of artists choosing to participate in NFT is insufficient. Thus, we developed a research model of the antecedents of artists participating in NFT that builds on social exchange theory. By applying a mixed methodology, we derive the benefits and costs of artists participating in NFT through a theoretical approach and interviews. We then conduct a survey on artists to verify the influence of each factor. Through this, we can suggest that management implications to reduce the risks and costs of participating in NFTs are needed to encourage the participation of artists. It will also provide insight into ways to develop the NFT art market.

텍스트마이닝 기법을 이용한 모바일 피트니스 애플리케이션 주요 요인 분석 : 사용자 경험 관점

이소현(So-Hyun Lee),김진솔(Jinsol Kim),윤상혁(Sang-Hyeak Yoon),김희웅(Hee-Woong Kim) 한국IT서비스학회 2020 한국IT서비스학회지 Vol.19 No.3

The development of information technology leads to changes in various industries. In particular, the health care industry is more influenced so that it is focused on. With the widening of the health care market, the market of smart device based personal health care also draws attention. Since a variety of fitness applications for smartphone based exercise were introduced, more interest has been in the health care industry. But although an amount of use of mobile fitness applications increase, it fails to lead to a sustained use. It is necessary to find and understand what matters for mobile fitness application users. Therefore, this study analyze the reviews of mobile fitness application users, to draw key factors, and thereby to propose detailed strategies for promoting mobile fitness applications. We utilize text mining techniques - LDA topic modeling, term frequency analysis, and keyword extraction - to draw and analyze the issues related to mobile fitness applications. In particular, the key factors drawn by text mining techniques are explained through the concept of user experience. This study is academically meaningful in the point that the key factors of mobile fitness applications are drawn by the user experience based text mining techniques, and practically this study proposes detailed strategies for promoting mobile fitness applications in the health care area.

개인의 마이데이터 제공의도에 영향을 미치는 요인: 개인역량과 기관유형의 조절효과를 중심으로

박동근 ( Dong Keun Park ),양성병 ( Sung-byung Yang ),윤상혁 ( Sang-hyeak Yoon ) 한국지식경영학회 2023 지식경영연구 Vol.24 No.1

최근 데이터의 중요성과 개인정보보호에 관련된 이슈가 함께 주목받으면서, 마이데이터 시장이 성장하고 있다. 마이데이터는 본인 정보에 대한 개인의 권리를 보장하고, 개인의 동의에 따라 자신의 데이터를 제공 및 활용하는 개념을 의미한다. 마이데이터 사업자는 고객정보를 결합, 분석하여 개인 맞춤 서비스를 제공할 수 있다. 초기 마이데이터 사업은 민간기업, 금융산업을 중심으로 활성화되었지만, 최근에는 공공기관도 마이데이터 활용에 적극적으로 나서고 있다. 한편, 마이데이터 사업의 성공을 위한 개인의 마이데이터 제공의도 중요성은 지속적으로 증가하고 있지만, 이와 관련된 연구는 부족한 상황이다. 더욱이, 기존 연구는 마이데이터의 개인혜택 측면에서 주로 이루어졌으나, 공공혜택 및 지각된 위험 측면 요인이 모두 함께 고려된 연구는 충분하지 않다. 이에, 본 연구는 마이데이터 제공의도에 영향을 주는 요인을 프라이버시 계산모형을 통해 도출한 후, 그 영향 메커니즘을 살펴봄과 동시에, 개인역량과 수집기관 유형에 따른 조절효과를 추가로 검증해 보고자 한다. 본 연구는 마이데이터 제공의도 맥락에서 프라이버시 계산모형을 실증했다는 점에서 학술적 의의를 찾을 수 있으며, 본 연구의 결과를 통해 민간 및 공공기관이 마이데이터 사업을 진행하는 데 있어 새로운 서비스 개발 및 관리를 위한 실무적인 지침을 제공해 줄 수 있을 것으로 기대한다. Recently, the MyData market has been growing as the importance of data and issues related to personal information protection have drawn much attention together. MyData refers to the concept of guaranteeing an individual's right to personal information and providing and utilizing one's data according to individual consent. MyData service providers can combine and analyze customer information to provide personalized services. In the early days, the MyData business was activated mainly by private companies and the financial industry, but recently, public institutions are also actively taking advantage of MyData. Meanwhile, the importance of an individual's intention to provide MyData for the success of MyData businesses continues to increase, but research related to this is lacking. Moreover, existing studies have been mainly conducted on individual benefits of MyData; there are not enough studies in which both public benefit and perceived risk factors are considered at the same time. In this regard, this study intends to derive factors affecting the intention to provide MyData based on the privacy calculus model, examine their influencing mechanism, and further verify the moderating effects of individual capabilities and institutional type. This study can find academic significance in that it expanded and demonstrated the privacy calculus model in the context of MyData providing intention. In addition, the results of this study are expected to offer practical guidelines for developing and managing new services in MyData businesses.

암호화폐 가격 예측을 위한 딥러닝 앙상블 모델링 : Deep 4-LSTM Ensemble Model

최수빈(Soo-bin Choi),신동훈(Dong-hoon Shin),윤상혁(Sang-Hyeak Yoon),김희웅(Hee-Woong Kim) 한국IT서비스학회 2020 한국IT서비스학회지 Vol.19 No.6

As the blockchain technology attracts attention, interest in cryptocurrency that is received as a reward is also increasing. Currently, investments and transactions are continuing with the expectation and increasing value of cryptocurrency. Accordingly, prediction for cryptocurrency price has been attempted through artificial intelligence technology and social sentiment analysis. The purpose of this paper is to develop a deep learning ensemble model for predicting the price fluctuations and one-day lag price of cryptocurrency based on the design science research method. This paper intends to perform predictive modeling on Ethereum among cryptocurrencies to make predictions more efficiently and accurately than existing models. Therefore, it collects data for five years related to Ethereum price and performs pre-processing through customized functions. In the model development stage, four LSTM models, which are efficient for time series data processing, are utilized to build an ensemble model with the optimal combination of hyperparameters found in the experimental process. Then, based on the performance evaluation scale, the superiority of the model is evaluated through comparison with other deep learning models. The results of this paper have a practical contribution that can be used as a model that shows high performance and predictive rate for cryptocurrency price prediction and price fluctuations. Besides, it shows academic contribution in that it improves the quality of research by following scientific design research procedures that solve scientific problems and create and evaluate new and innovative products in the field of information systems

개인의 건강신념이 모바일 헬스케어 앱 이용의도에 미치는 영향

왕진섭(Jin-Seob Wang),송재민(Jaemin Song),양성병(Sung-Byung Yang),윤상혁(Sang-Hyeak Yoon) 한국IT서비스학회 2023 한국IT서비스학회지 Vol.22 No.1

Smart healthcare, combining ICT (Information and Communications Technologies) and medical technologies, has been rapidly emerging. Accordingly, its market has also increased as interest in disease prevention, management, and diagnosis grows due to the COVID-19 pandemic. In particular, using mobile devices to support medical activities, mobile healthcare has been attracting attention as a leading service in the smart healthcare market. However, the intention to use mobile healthcare apps may vary depending on individual beliefs and attitudes. Many studies on the intention to use mobile healthcare apps have used the TAM (Technology Acceptance Model), but there is a lack of studies that have been verified from the perspective of users' health beliefs. This study aims to identify the factors that affect the intention to use mobile healthcare apps based on the HBM (Health Belief Model). Furthermore, it investigates how this influencing mechanism fluctuates based on the user’s mHealth literacy, the ability to find and understand health information through mobile. This study contributes to the empirical examination of the intention to use mobile healthcare apps through the HBM. It also offers insights for app providers and public health officials to increase the use of mobile healthcare apps.