개인화 추천시스템에서 고객 제품 리뷰가 사회적 실재감에 미치는 영향

다국어 초록 (Multilingual Abstract)

Many online stores bring features that can build trust in their customers. More so, the number of products or content services on online stores has been increasing rapidly. Hence, personalization on online stores is considered to be an important technology to companies and customers. Recommender systems that provide favorable products and customer product reviews to users are the most commonly used features in this purpose. There are many studies to that investigated the relationship between social presence as an antecedent of trust and provision of recommender systems or customer product reviews. Many online stores have made efforts to increase perceived social presence of their customers through customer reviews, recommender systems, and analyzing associations among products. Primarily because social presence can increase customer trust or reuse intention for online stores. However, there were few studies that investigated the interactions between recommendation type, product type and provision of customer product reviews on social presence. Therefore, one of the purposes of this study is to identify the effects of personalized recommender systems and compare the role of customer reviews with product types.
This study performed an experiment to see these interactions. Experimental web pages were developed with 2×2 factorial setting based on how to provide social presence to users with customer reviews and two product types such as hedonic and utilitarian. The hedonic type was a ringtone chosen from Nate.com while the utilitarian was a TOEIC study aid book selected from Yes24.com. To conduct the experiment, web based experiments were conducted for the participants who have been shopping on the online stores. Participants were a total of 240 and 30% of the participants had the chance of getting the presents.
We found out that social presence increased for hedonic products when personalized recommendations were given compared to non?personalized recommendations. Although providing customer reviews for two product types did not significantly increase social presence, provision of customer product reviews for hedonic (ringtone) increased perceived social presence. Otherwise, provision of customer product reviews could not increase social presence when the systems recommend utilitarian products (TOEIC study aid books). Therefore, it appears that the effects of increasing perceived social presence with customer reviews have a difference for product types. In short, the role of customer reviews could be different based on which product types were considered by customers when they are making a decision related to purchasing on the online stores.
Additionally, there were no differences for increasing perceived social presence when providing customer reviews. Our participants might have focused on how recommendations had been provided and what products were recommended because our developed systems were providing recommendations after participants rating their preferences. Thus, the effects of customer reviews could appear more clearly if our participants had actual purchase opportunity for the recommendations.
Personalized recommender systems can increase social presence of customers more than nonpersonalized recommender systems by using user preference. Online stores could find out how they can increase perceived social presence and satisfaction of their customers when customers want to find the proper products with recommender systems and customer reviews.
In addition, the role of customer reviews of the personalized recommendations can be different based on types of the recommended products. Even if this study conducted two product types such as hedonic and utilitarian, the results revealed that customer reviews for hedonic increased social presence of customers more than customer reviews for utilitarian. Thus, online stores need to consider the role of providing customer reviews with highly personalized in

참고문헌 (Reference)

• 1 김명규, "텍스트 문서 기반의 감성 인식 시스템" 한국감성과학회 12 (12): 433-442, 2009
• 2 박수환, "전자상거래 개인화 추천을 위한 상품 카테고리 중립적 사용자 프로파일링" 한국경영정보학회 16 (16): 159-176, 2006
• 3 김재경, "유비쿼터스 환경에서 개체간의 자율적 협업에 기반한 추천방법 개발" 한국지능정보시스템학회 15 (15): 31-50, 2009
• 4 이국용, "온라인 상품검색사이트의 이용후기 특성과 친숙성이 충성도에 미치는 영향" 한국전자거래학회 15 (15): 17-37, 2010
• 5 허성혜, "온라인 리뷰 수용에 영향을 미치는 요인 : 온라인 리뷰 품질과 동의성을 중심으로" 한국데이타베이스학회 16 (16): 41-58, 2009
• 6 김근형, "온라인 고객리뷰 분석을 통한 시장세분화에 텍스트마이닝 기술을 적용하기 위한 방법론" 한국콘텐츠학회 9 (9): 272-284, 2009
• 7 Clemons, E., "When online reviews meet hyperdifferentiation：A study of the craft beer industry" 32 (32): 149-171, 2006
• 8 Mudambi, S. M., "What makes a helpful online review? A study of customer reviews on Amazon.com" 34 (34): 185-200, 2010
• 9 Zhang, Z., "Weighing Stars：aggregating online product reviews for intelligent e-commerce applications" 23 (23): 42-49, 2008
• 10 Tam, K. Y., "Understanding the impact of web personalization on user information processing and decision outcome" 30 (30): 865-890, 2006
• 11 Choi, J., "The influence of social presence on evaluating personalized recommender systems" 49-, 2009
• 12 Kumar, N., "The influence of recommendations and consumer reviews on evaluations of websites" 17 (17): 425-429, 2006
• 13 Hassanein, K., "The impact of infusing social presence in the web interfaces ：an investigation across product types" 10 (10): 31-55, 2006
• 14 Al-Natour, S., "The effects of process and outcome similarity on users’ evaluations of decision aids" 39 (39): 175-211, 2008
• 15 Komiak, S. Y. X., "The effects of personalization and familiarity on trust and adoption of recommendation agents" 30 (30): 941-960, 2006
• 16 Huang, P., "Searching for Experience on the Web：An Empirical Examination of Consumer Behavior for Search and Experience Goods" 73 (73): 55-69, 2009
• 17 Wang, W., "Recommendation agents for electronic commerce：effects of explanation facilities on trusting beliefs" 23 (23): 217-246, 2007
• 18 Nunnally, J., "Psychometric Methods" New York Mcgrow Hill Book Co 1967
• 19 Liang, T., "Personalized content recommendation and user satisfaction : theoretical synthesis and empirical findings" 23 (23): 45-70, 2007
• 20 Resnick, P., "GroupLens：an open architecture for collaborative filtering of netnews" 175-186, 1994
• 21 Petrova, P. K., "Fluency of Consumption Imagery and the Backfire Effects of Imagery Appeals" 32 : 442-452, 2005
• 22 Cao, Q., "Exploring determinants of voting for the ‘helpfulness’ on line user reviews：A text mining approach" 50 (50): 511-521, 2011
• 23 Ghose, A., "Estimating the helpfulness and economic impact of product reviews：mining text and reviewer characteristics"
• 24 Xiao, B., "E-commerce product recommendation agents：use, characteristics, and impact" 31 (31): 137-209, 2007
• 25 Geffen, D., "Consumer trust in B2C e-commerce and the importance of social presence：experiments in e-products and e-services" 32 (32): 407-424, 2004
• 26 Phillips, D. M., "Anticipating the Future: The Role of Consumption Visions in Consumer Behavior" 23 : 70-75, 1996
• 27 Chen, P., "All Reviews are not created equal：The disaggregate impact of reviews on sales on Amazon.com" Carnegie Mellon University
• 28 Komiak, S. Y. X., "A two-process view of trust and distrust building in recommendation agents：a process-tracing study" 9 (9): 727-747, 2008