• 한국의류학회지
• /
• 제42권5호
• /
• pp.871-883
• /
• 2018

This study tests consumer responses to online product recommendation service offered by a website. A product recommendation service refers to a filtering system that predicts and shows items that consumers would like to purchase based on their searches or pre-purchase information. The survey is conducted on 300 people in an age group between 20 and 40 years in a panel of an online survey firm. Data are analyzed using confirmatory factor analysis and structural equation modeling by AMOS 20.0. The results show that personalization quality does not have a significant effect on trust, but relationship quality and technology quality have a positive effect on trust. Three types of quality of recommendation service also have a positive effect on commitment. Trust and commitment are factors that increase service usage intentions. In addition, this study reveals the moderating effect of light users vs heavy users based on online shopping time. Light users show a negative effect of personalization quality on trust, indicating that they are likely to be uncomfortable to the service using personal information, compared to heavy users. This study also finds that trust vs commitment is an important factor increasing service usage intentions for heavy users vs light users.

• 지역과문화
• /
• 제8권2호
• /
• pp.1-20
• /
• 2021

본 연구는 지역콘텐츠 제작지원 사업에 대한 지각된 서비스 품질이 정책 수혜자의 이용 만족에 미치는 영향과 사업 종료 후 재이용 의도 및 추천 의도에 미치는 영향을 분석하였다. 이를 위해 SERVPERF 모형에 근거한 지역콘텐츠 제작지원 사업에 적합한 품질기준을 구성하였다. 첫째, 지역콘텐츠 제작지원 사업의 서비스 품질평가 기준으로 신뢰성, 대응성, 적정성을 확인하였다. 둘째, 품질평가 기준은 모두 이용 만족에 유의미한 영향을 미쳤다. 셋째, 재이용 의도에 대응성이, 추천 의도에 대응성과 신뢰성이 유의미한 영향을 미쳤다. 넷째, 이용 만족은 재이용 의도와 추천 의도에 유의미한 영향을 미쳤다. 이에 본 연구는 실증 분석으로 지역콘텐츠 제작지원 사업의 품질평가에 대한 이론적 모형을 처음으로 제시하고, 해당 사업의 개선을 위한 정책적 방향성을 제시했다는 의의를 지닌다.

• 유통과학연구
• /
• 제18권5호
• /
• pp.99-109
• /
• 2020

Purpose: This study aims to develop and empirically analyze a research model in order to comprehend the relationship among the service quality of market-wholesaler system, re-transaction intention, and recommendation intention of forwarder. Further, we suggest new six factors reflecting the service quality of market-wholesaler system and highlight market expansion of forwarder as a mechanism in the relationship. Research design, data and methodology: The authors developed the new scales measuring the service quality of market-wholesaler system (i.e. trade price, price fluctuation, payment receipt, settlement period, trade information, and customer service) and conducted a cross-sectional survey for 439 forwarders in a wholesale market. And then we performed a series of path analyses to test hypotheses. The hypotheses are as follows. [H1] The service quality of market-wholesaler system will positively affect forwarders' market expansion, [H2] Forwarders' market expansion will positively affect their re-transaction intention, [H3] Forwarders' market expansion will positively affect their recommendation intention, [H4] Forwarders' re-transaction intention will positively affect their recommendation intention. Results: The results showed that all the six factors for the service quality of market-wholesaler system were positively related to market expansion of forwarders. There was a differential effectiveness in the six factors of the service quality. More specifically, the positive effect of customer service factor was the strongest on market expansion of forwarders. And the respective effects of trade price, price fluctuation, settlement period, trade information factors were followed in order. The positive effect of payment receipt factor was the weakest on market expansion of forwarders. Also, market expansion of forwarders was positively related to their re-transaction intention and recommendation intention. Furthermore, market expansion of forwarders was indirectly related to recommendation intention through re-transaction intention as well. Conclusions: The research findings provide important theoretical and practical implications. This study is the first to attempt to test the perception of forwarders for the service quality of market-wholesaler system by developing and using the new scales. Also, there has been a sharp controversy about the effectiveness of market-wholesaler system. The findings support that market-wholesaler system would be activated by empirically verifying the effectiveness of the service quality on the various outcomes.

• 한국멀티미디어학회논문지
• /
• 제20권2호
• /
• pp.322-329
• /
• 2017

Recently, the importance of personalized healthcare(wellness) services is increasing in the era of the 4th Industrial Revolution. However, the authoring of wellness contents fused with variety of contents and the study of the system which provides the customized recommendation are insufficient. In this paper, we proposes the recommendation service system for the customized convergence wellness contents. The proposed system makes to the wellness contents by the existing cultural/tourism/leisure contents and recommends the customized wellness contents based on a user's profile and the situation information such as location and weather. The proposed systems is expected to contribute to designing the innovative and new service models for the tailored wellness content.

• 지능정보연구
• /
• 제15권1호
• /
• pp.31-50
• /
• 2009

유비쿼터스 컴퓨팅 환경에서는 정적 및 동적인 상황 정보의 양이 무한대로 늘어나게 됨에 따라, 추천서비스에 있어서 정보 과부하 문제와 프라이버시 침해 문제가 중요한 문제로 대두되고 있다. 따라서 본 연구에서는 이러한 문제점을 해결하기 위하여 서버와의 교신 없이 고객 중심의 자체적인 정보처리와 고객들간 직접 커뮤니케이션을 통하여, 효율적이고 안전한 정보 획득이 가능하도록 P2P방식의 협업을 통하여 선호도가 유사한 다른 고객들의 상품에 대한 평가정보가 전달되는 추천서비스를 제안하였다. 제안한 추천방식은 협업필터링의 기본 법칙을 따르고 있지만, 현재 센서 네트워크에 접속해 있는 전체 고객를 대상으로 이웃 고객을 탐색하는 방법대신에 목표 고객 주위의 가까운 이웃을 지역적으로 탐색하는 방법을 채택하여 성능의 저하없이 유비쿼터스 컴퓨팅 환경에서 실시간 추천이 가능하도록 하였다. 또한 유비쿼터스 컴퓨팅 환경에서 적용가능한 프로토타입의 통합 네트워크 시스템의 구현을 통해 실세계 상점에서 유비쿼터스 컴퓨팅 기술의 활용 가능성을 제시하였다. 마지막으로 실제 모바일 회사의 데이터를 이용한 실험을 통하여 그 특징을 제시함으로써 향후 유비쿼터스 서비스 애플리케이션의 범용적인 추천모델을 제공하고자 한다.

• 한국인터넷방송통신학회논문지
• /
• 제21권3호
• /
• pp.99-105
• /
• 2021

넥플릭스,유튜브로 대표되는 OTT 동영상 제작 서비스에 인공지능으로 콘텐츠를 개인별 맞춤식 추천 시스템은 보편화 되었다. 유튜브의 개인별 맞춤 추천서비스 시스템은 두 개의 신경망으로 구성되는데 신경망 하나는 추천 후보생성 모델이고 다른 하나는 순위평가 네트워크로 구성된다. Netflix의 동영상 추천 시스템은 두 개 데이터 분류 시스템으로 구성되어 있으며 콘텐츠 기반 필터링과 협업 필터링으로 나누어진다. 코로나 펜데믹으로 온라인 플랫폼 주도의 콘텐츠 제작이 활성화 되면서 인공지능을 활용한 가상 인플루언서 분야가 부각되고 있다. 가상인플루언서는 GAN(Generative Adversarial Networks) 인공지능으로 제작되는데 성격이 다른 두 시스템이 서로 경쟁하는 방식으로 학습이 반복되는 비교사(Unsupervised) 학습 알고리즘이다. 이 연구는 AI 개인별 추천 기반 플랫폼과 가상인플루언서(메타버스)가 향후 OTT의 핵심콘텐츠로의 발전 가능성도 연구해 보았다.

• Asia pacific journal of information systems
• /
• 제17권1호
• /
• pp.123-145
• /
• 2007

There are flourishing studies in the acceptance or usage of information systems literature. Most of them have taken the pro - acceptance view. Undesirably, information technologies often provoke users' reactance or resistance. This paper explores one of the negative reactions -psychological reactance. The present paper studies the effects of the users' perception of threatened freedom and personalization degree on intention to use recommendation services. High personalization can be a major motivation for users to accept recommendation systems. However recommendation services are a two-edged sword, which not only provides users the efficiency of decision making but also poses threats to free choice. When people consider that their freedom is reduced or threatened by others, they experience the motivational state to restore the freedom. This motivational state must be considered in understanding usage of information systems, especially personalized services which are designed for persuasion or compliance. This paper empirically investigates the effect of personalization and the psychological reactance on the intention to use information systems in the personalized recommendation context. Users' perception of personalization increases the usefulness of recommendation service while their perception of threat to freedom reduces the intention to use personalized recommendation service. Findings and implications are discussed.

• 대한가정학회지
• /
• 제47권1호
• /
• pp.13-24
• /
• 2009

The purpose of this study was to examine the effects of banking service quality on consumer satisfaction and positive word-of-mouth. A total of 330 bank consumers were investigated between Sept. 11 and Oct. 11, 2006. After sorting through the data, the responses of 299 consumer ( > 24 yrs old) were used for analysis. SERVPERF, the performance component of the Service Quality scale (SERVQUAL), was used to measure the four dimensions of reliability, responsiveness/empathy, assurance, and tangibles. Responses were partitioned by age and gender. The major findings were as follows; The effect of service quality by SERVPERF on consumer satisfaction and positive word-of-mouth did not differ according to gender. However, positive recommendation in males was directy related to the technical quality evaluate. In females, higher the functional quality evaluate was directly related to higher positive word-of-mouth recommendation. The effect of service quality by SERVPERF on consumer satisfaction was not revealed differently according to age. However, with respect to respondents under the age of 45, tangibles and assurance had a positive relationship with word-of-mouth recommendation. Furthermore, the higher the functional quality evaluate, the higher the level of positive word-of-mouth. Responsiveness/empathy was the most significant factor on positive word-of-mouth recommendation in respondents over the age of 45. In this age group, the higher the technical quality evaluate, the higher the level of positive word-of-mouth recommendation. These results have implications for banking service managers, particularly in improving service quality to increase consumer satisfaction and positive word-of-mouth. Future research is needed to replicate this study using more broad and representative samples in order to test the generalization of these findings.

• 반도체디스플레이기술학회지
• /
• 제18권3호
• /
• pp.93-99
• /
• 2019

It is a wine recommendation application service designed for people who sometimes drink wine but lack information and have no place to recommend. This study is to develop UI display design method of wine recommendation service using chatbot. The research method was a case study on Korean wine market, a case study on artificial intelligence market, SWOT analysis of wine-related chatbots, and a competitor analysis of related industries. In addition, surveys and in-depth interviews examined the level of interest and understanding of chatbots, and what kind of chatbots they had encountered and what requirements and goals they faced. After grasping the needs and requirements of users, we created a service concept sheet according to them and produced an application UI design that users can use most easily. Therefore, this study is meaningful in that it proposes a UI design that can search wine information more sophisticated and convenient than face-to-face communication through artificial intelligence service called chatbot and recommend wines that match the taste.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 춘계학술발표대회
• /
• pp.346-348
• /
• 2022

Keyword-Based Contents Recommendation Web Service(서비스명 'mobodra')는 미디어 종류 및 장르 취향을 유저별로 분석하여 이에 맞는 콘텐츠를 추천하는 웹 서비스이다. 유저들은 회원가입 시 웹에서 제공하는 랜덤한 작품 중에 일부를 선택하며 서버에서 이를 토대로 취향을 분석한다. 해당 분석을 토대로 유저별 선호 콘텐츠를 추천한다. 본 논문에서는 아이템 기반 협업 필터링(Item-Based Collaborative Filtering)을 통해 콘텐츠 추천 알고리즘을 구현한다. 유저의 활동 데이터 혹은 선호도 재조사 시 위 과정을 다시 실행하여 사용자의 취향을 갱신한다.