• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.10
• /
• pp.1010-1014
• /
• 2010

Article-related product recommender system is an emerging e-commerce service which recommends items based on association in contexts between items and articles. Current services recommend based on the similarity between tags of articles and items, which is deficient not only due to the high cost in manual tagging but also low accuracies in recommendation. As a component of novel article-related item recommender system, we propose a new method for tagging item images based on pre-defined categories. We suggest a hypernetwork-based algorithm for learning association between images, which is represented by visual words, and categories of products. Learned hypernetwork are used to assign multiple tags to unlabeled item images. We show the ability of our method with a product set of real-world online shopping-mall including 1,251 product images with 10 categories. Experimental results not only show that the proposed method has competitive tagging performance compared with other classifiers but also present that the proposed multi-tagging method based on hypernetworks improves the accuracy of tagging.

• The Journal of the Korea Contents Association
• /
• v.22 no.7
• /
• pp.164-175
• /
• 2022

The OTT (Over The Top) platform promotes itself as a distinctive competitive advantage in that it allows users to stay on the platform longer and visit more often through a Personalized Recommendation Service. In this study, the characteristics of the Personalized Recommendation Service are divided into three categories: recommendation accuracy, recommendation diversity, and recommendation novelty. Then proposed a research model which affects the usefulness of users to recognize recommendation services by each characteristics and leads to satisfaction of expectations. The result of conducting an online survey of 300 people in their 20s and 30s who subscribe Netflix shows that the perceived usefulness increased when the accuracy, variety, and novelty of Netflix's Recommendation Service were high. It was also confirmed that high perceived usefulness leads to satisfaction of expectations before and after Netflix use. The derived research results can confirm the importance of evaluating the personalized recommendation service in terms of user experience and provide implications for ways to improve the quality of recommendation services.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.372-375
• /
• 2018

본 논문은 온라인 음원 서비스 이용자들이 겪는 선택의 어려움을 최소화하고, 낭비되는 시간을 줄이기 위한 음악 추천 기법을 제안하고자 한다. 제안하는 기법은 개인정보의 이용 없이 아이템을 추천할 수 있는 아이템 기반 협업필터링 알고리즘을 사용한다. 더 정확한 추천을 위해 음원의 메타데이터를 이용한다. 실험을 통해 제안하는 기법이 메타데이터를 이용하지 않을 때보다 추천 성능이 향상되는 것을 확인하였다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.109-112
• /
• 2007

온라인 음악 시장이 점차 커지고 있다. 이에 따라 사용자를 위한 다양한 서비스가 요구되고 있다. 하지만 현재 적용되는 서비스는 통계적인 수치에 기반하는 순위권 나열 혹은 테마나 장르별 음악 소개에 그치고 있다. 따라서 본 논문에서는 사용자의 성향에 가까운 음악을 분석하고 이를 추천하는 방법을 제시한다. 음악 추천 시스템을 위해 우선 사용자의 성향을 분석하기 위하여 사용자가 청취했던 음악의 음파를 분석하여 특성을 추출하여 벡터로 나타낸다. 하지만 추출된 성향과 다른 음악의 성향을 비교해야 하는데 음악의 양이 방대하기 때문에 시간이 오래 걸릴 수 있다. 따라서 이 문제를 해결하기 위해 공간 분할을 통해 검색의 범위를 축소시키고, 음악을 빠르게 추천한다. 실험 결과, 사람의 주관적인 해석이 아닌 음파의 해석을 통해 보다 객관적이고 자동화된 추천 방법을 구현할 수 있었다. 그리고 같은 성질의 음악이 추천되어짐을 확인할 수 있었다.

• Smart Media Journal
• /
• v.13 no.1
• /
• pp.67-75
• /
• 2024

The purpose of this research is to investigate consumer responses to production recommendations by AI. In order to test hypotheses of this study, we conducted experimental study that was a 2(Brand reputation: high vs. low) X 2(Social comparison: high vs. low). The results of this study showed the interaction effects of brand reputation and social comparison on brand attitude. Based on the results, we provide theoretical implications to extent the existing research regarding product recommendations. Moreover, the results of this study provide some practical implications and a new aspect about AI recommendations.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.10a
• /
• pp.587-590
• /
• 2016

최근 통신 기술의 발달로 온라인을 통한 대규모 콘텐츠의 유통이 가능해졌으나, 사용자들은 수많은 콘텐츠 사이에서 원하는 정보를 찾는 시간이 단축되는 것을 원했다. 이로 인해 다양한 분야에서 개인화된 콘텐츠를 추천해주는 추천 시스템(recommendation system)에 대한 요구가 점차 높아졌다. 그럼에도 불구하고 시니어를 위한 추천 시스템에 대한 연구는 매우 부족하다. 또한, 시니어 세대의 변화에 따라 시니어 관련 콘텐츠 연구도 다양하게 진행되고 있으나, 스마트 기기 및 서비스가 젊은 층에 친화적으로 개발됨으로써 시니어 층의 접근성을 감소시키고 있다. 이에 본 연구에서는 다양한 신체적 변화를 겪는 시니어 세대 위해 추천 시스템에서 인지반응 데이터를 이용하여 콘텐츠를 시청하기 적합한 환경을 제공함과 동시에 활동 이력을 중심으로 개인화 추천 시스템을 설계하여 시니어 사용자들의 개념 변화(concept drift) 문제로 사용자가 원하지 않는 콘텐츠를 추천받을 가능성을 줄일 수 있도록 한다.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.5
• /
• pp.37-45
• /
• 2022

As of 2020, about 500 hours of videos are uploaded to YouTube, a representative online video platform, per minute. As the number of users acquiring information through various uploaded videos is increasing, online video platforms are making efforts to provide better recommendation services. The currently used recommendation service recommends videos to users based on the user's viewing history, which is not a good way to recommend videos that deal with specific purposes and interests, such as educational videos. The recent recommendation system utilizes not only the user's viewing history but also the content features of the item. In this paper, we extract the content features of educational video for educational video recommendation based on YouTube, design a recommendation system using it, and implement it as a web application. By examining the satisfaction of users, recommendataion performance and convenience performance are shown as 85.36% and 87.80%.

• The Journal of the Korea Contents Association
• /
• v.18 no.12
• /
• pp.660-675
• /
• 2018

This study is designed to investigate how the perceived characteristics of the online video services affect the 'flow' as positive experience and satisfaction of users. For the study, we conducted a questionnaire survey on 289 people using YouTube, and then analyzed the relationships among variables using hierarchical regression analysis. As a result, it was confirmed that interactivity, newness of recommendation service, diversity of content, and entertainingness of contents all affect the lower level of flow experience. On the other hand, the accuracy of the recommendation service did not affect the flow experience, but positively affects the level of satisfaction. Finally, it is also confirmed that flow has a direct effect on user satisfaction, and mediates relationship between the characteristics of YouTube and satisfaction. The results of this study are helpful to understand user's perception and experience of online video platform service and suggest the discussion points to be considered by the industry to satisfy users.

• Journal of The Korean Association of Information Education
• /
• v.26 no.6
• /
• pp.517-532
• /
• 2022

As a result of the this exploratory study the main results show that the factors affecting the satisfaction of non-face-to-face class learning of graduate students in Education were social presence, perceived easy to use, and information service quality. Secondly, the factors affecting the learning persistence were perceived usefulness and social presence, but perceived usefulness's influence was relatively small. Next, it was found that the subjective satisfaction of graduate students of education did not significantly affect their intention to continue learning persistence. Lastly, learning satisfaction and learning persistence had a positive effect on the intention to recommend to others. Through the this empirical analysitics study it demonstrate the factors such as attitudes, feelings, and friendliness toward non-face-to-face online classes perceived by graduate students in Education.

• Information Systems Review
• /
• v.25 no.4
• /
• pp.67-87
• /
• 2023

Many online shopping sites now offer personalized recommendation systems to improve consumers' shopping experiences by lowering costs (time, cost, etc.), catering to consumers' tastes, and stimulating consumers' potential shopping needs. So far, domestic and foreign research on the personalized recommendation system has mainly focused on the field of computer science, which is advantageous for obtaining accurate personalized recommendation results for users but difficult to continuously track the users' psychological states or behavioral intentions. This study attempted to investigate the effect of the characteristics of the personalized recommendation system in the online shopping environment on consumer perception and purchase intention for consumers using the Stimulus-Organism-Response (S-O-R) model. The analysis results adopted all hypotheses on the effect of the quality of the personalized recommendation system and information quality on trust and perceived value. Through the empirical results of this study, the factors influencing consumers' use of personalized recommendation system can be identified. In order to increase more purchase, online shopping companies need to understand consumers' tastes and improve the quality of the personalized system by improving the recommendation algorithm thus to provide more information about products.