• The Journal of Asian Finance, Economics and Business
• /
• 제10권2호
• /
• pp.83-90
• /
• 2023

This study aims to determine the direct effect of positive online reviews, negative online reviews, the usefulness of online reviews, reviewers' expertise, timeliness of online reviews, the volume of online reviews, and comprehensiveness of online reviews on accommodation booking intentions and also the indirect effect of positive online reviews on the intention of booking accommodations through trust as mediation. Research respondents are users of the accommodation booking application in Yogyakarta. Hypothesis testing was carried out using SEM (Partial Least Square). Data was collected by distributing questionnaires to 135 respondents. The results of this study indicate that the Usefulness of Online Reviews, Volume of Online Reviews, and Comprehensiveness of Online Reviews have a direct positive and significant influence on the accommodation booking Intention of booking application users in Yogyakarta. The variables of Negative Online Reviews and Timeliness of Online Reviews have negative and significant influences on the accommodation booking Intention of booking application users in Yogyakarta. Positive Online Reviews and Reviewer Expertise variables are not significant in this study. At the same time, the Trust variable has a full mediation relationship in an indirect relationship between the Positive Online Reviews variables and the accommodation booking Intention of booking application users in Yogyakarta.

• 패션비즈니스
• /
• 제22권3호
• /
• pp.89-108
• /
• 2018

This paper aims to explore the analysis of the meanings and processes of reading online consumer reviews and to construct a substantive theory that explains the process involved with the phenomenon of reading consumer reviews. In order to explore the phenomenon, this study employs a qualitative methodology. Following the grounded theory perspective, the researcher conducted interviews with 17 participants, who have subsequently shopped online and utilized online consumer reviews for shopping, and decidedly employed in-depth interviews with those participants. Through coding and making constant comparison, several themes emerged: improving confidence, trusting reviews, getting a sense of who reviewers are, seeking balance, processing and handling negative reviews, experiencing vicariously, increasing searchability, getting a sense of who they are in terms of similarity, and seeking benefits and the usage situations from consumer based reviews. Among the emerging themes, improving confidence can be considered a core category, which is influenced by the analysis of trusting reviews and the consumer vicarious experiences with a product. Moreover, this study discusses the relationships among the themes. This study concludes with a discussion of the results, implications, and limitations.

• 복식문화연구
• /
• 제21권3호
• /
• pp.440-451
• /
• 2013

Because of the e-shopping market consumers now have diverse options to choose when placing their orders, and find it easy to obtain the required information through the Internet. Especially, for consumers, product reviews posted on an e-tailer's website have become more important criteria than such information available elsewhere. Hence, this study investigated the influence of the direction and evaluative contents of online reviews on consumer attitudes toward clothing products. Four types of online reviews based on direction (positive/negative) and evaluative content in review information (objective/subjective) were used in the experimental design. Further, stimulus reviews were developed. Credibility, usefulness of reviews, product preference, and purchase intention were the measured dependent variables in each of the four situations of online review presentations. The results indicated that, overall, positive and objective online reviews resulted in a higher level of consumer attitude. The content in these reviews had a relatively stronger influence than the direction on attitudes toward online reviews. Overall, objective reviews generated a higher level of credibility and usefulness of information than subjective reviews. Regarding subjective reviews, negative information was more related to credibility, whereas positive information was more related to usefulness. Further, positive information had a higher influence than negative information on consumer attitudes.

• Fashion, Industry and Education
• /
• 제16권1호
• /
• pp.19-35
• /
• 2018

This study explores the effect of information direction and order of cosmetics power bloggers on consumer responses. A total of 488 undergraduate students participated in experiments with mock-up stimuli of sunscreen product reviews by power bloggers. The study was conducted with four stimuli of product review posts (i.e., positive reviews only, positive-negative reviews in order, negative-positive reviews in order, negative reviews only) of the power bloggers. The results showed a significant difference in consumer responses according to information direction and order of product reviews of the power bloggers. Specifically, negative reviews were considered more objective and more useful than positive reviews were. However, positivity of reviews is crucial in generating more positive attitudes toward products, greater purchase intention, and greater word-of-mouth intention. In regard to information order, the negative-positive reviews generated more positive attitudes toward the product and greater purchase intention than did the positive-negative reviews, emphasizing the importance of ending product reviews with positive information so as to create positive responses. Referring to the findings, power bloggers and marketers using bloggers as a promotional tool would benefit by carefully designing information content in consideration of an appropriate direction and order of information to better fit their purpose.

• 한국IT서비스학회지
• /
• 제21권4호
• /
• pp.75-90
• /
• 2022

From the past to the present, reviews have had much influence on consumers' purchasing decisions. Companies are making various efforts, such as introducing a review incentive system to increase the number of reviews. Recently, as various types of reviews can be left, reviews have begun to be recognized as interesting new content. This way, reviews have become essential in creating loyal customers. Therefore, research and utilization of reviews are being actively conducted. Some studies analyze reviews to discover customers' needs, studies that upgrade recommendation systems using reviews, and studies that analyze consumers' emotions and attitudes through reviews. However, research that predicts the future using reviews is insufficient. This study used a dataset consisting of two reviews written in pairs with differences in usage periods. In this study, the direction of consumer product evaluation is predicted using KoBERT, which shows excellent performance in Text Deep Learning. We used 7,233 reviews collected to demonstrate the excellence of the proposed model. As a result, the proposed model using the review text and the star rating showed excellent performance compared to the baseline that follows the majority voting.

• 한국IT서비스학회지
• /
• 제16권2호
• /
• pp.21-32
• /
• 2017

This study analyzes the variable consumer characteristics that influence the intention to use online product reviews. In online e-commerce, where purchases take place without consumers seeing the products in person, the product reviews left by other consumers who have already purchased the product are believed to be valuable information. However, when different consumers read the same product review, their responses to it may vary. This study analyzes the characteristics of consumers who utilize product reviews for their purchases. Consumer characteristics are categorized into personal information, personality, purchasing tendency, and experience related to product reviews. These factors are examined to see if they have direct or indirect effects on a consumer's intention to use product reviews when making online purchases. We surveyed a total of 240 consumers who had experience using e-commerce and knew about online product reviews. Once the data was collected, path analysis was conducted using the statistics tool AMOS. The study results reveal that consumers who are female, extroverted, and have higher price sensitivity think that product reviews left by others are useful, and that this "perceived usefulness" has a positive effect on the intention to use product reviews for making online purchasing decisions. In addition, consumers who are agreeable to others, have high brand sensitivity, and who have left numerous reviews themselves demonstrated the tendency to trust reviews left by others more. Thus, we conclude that this "perceived reliability" makes it more likely that a consumer will use product reviews when making online purchasing decisions. Future research can be done to develop this study further by analyzing whether providing online product reviews corresponding to the personal characteristics of consumers enhances the effect of product reviews on online purchasing decisions.

• 한국IT서비스학회지
• /
• 제14권4호
• /
• pp.159-169
• /
• 2015

Customer reviews are one of the important sources for purchase decision makings in online stores. Online stores have tried to provide useful reviews in product pages to customers. To assess the usefulness of customer reviews before other users have voted enough on the reviews, diverse aspects of reviews were utilized in prevous studies. Style and semantic information were utilized in many studies. This study aims to test diverse alogrithms and datasets for identifying a proper classification method and threshold to classify useful reviews. In particular, most researches utilized ratio type helpfulness index as Amazon.com used. However, there is another type of usefulness index utilized in TripAdviser.com or Yelp.com, count type helpfulness index. There was no proper threshold to classify useful reviews yet for count type helpfulness index. This study used reivews and their usefulness votes on restaurnats from Yelp.com to devise diverse datasets and applied text mining approaches to classify useful reviews. Random Forest, SVM, and GLMNET showed the greater values of accuracy than other approaches.

• Asia pacific journal of information systems
• /
• 제31권4호
• /
• pp.472-490
• /
• 2021

The detection of products with manipulated reviews has received widespread research attention, given that a truthful, informative, and useful review helps to significantly lower the search effort and cost for potential customers. This study proposes a method to recognize products with manipulated online customer reviews by examining the sequence of each review's sentiment, readability, and rating scores by product on randomness, considering the example of a Russian online retail site. Additionally, this study aims to examine the association between brand awareness and existing manipulation with products' reviews. Therefore, we investigated the difference between well-known and unknown brands' products online reviews with and without manipulated reviews based on the average star rating and the extremely positive sentiment scores. Consequently, machine learning techniques for predicting products are tested with manipulated reviews to determine a more useful one. It was found that about 20% of all product reviews are manipulated. Among the products with manipulated reviews, 44% are products of well-known brands, and 56% from unknown brands, with the highest prediction performance on deep neural network.

• 지능정보연구
• /
• 제28권4호
• /
• pp.347-364
• /
• 2022

고객의 구매 의사결정에 영향을 주는 온라인 리뷰의 부적절한 조작을 통해 이익을 얻고자 하는 기업 또는 온라인 판매자들 때문에, 리뷰의 신뢰성은 온라인 거래에서 매우 중요한 이슈가 되었다. 온라인 쇼핑몰 등에서 온라인 리뷰에 대한 소비자들의 의존도가 높아짐에 따라 많은 연구들이 조작된 리뷰를 탐지하는 방법에 개발하고자 하였다. 기존의 연구들은 온라인 리뷰를 기반으로 정상 리뷰와 조작된 리뷰를 대상으로 기계학습으로 이용함으로써 조작된 리뷰를 탐지하는 모형을 제시하였다. 기계학습은 데이터를 이용하여 이진분류 문제에서 탁월한 성능을 보여왔으나, 학습에 충분한 데이터를 확보할 수 있는 환경에서만 이러한 성능을 기대할 수 있었다. 조작된 리뷰는 학습용으로 사용할 수 있는 데이터가 충분하지 못하며, 이는 기계학습이 충분한 학습을 할 수 없다는 치명적 약점으로 내포하게 된다. 본 연구에서는 기계학습이 불균형 데이터 셋으로 인한 학습의 저하를 방지할 수 있는 방안으로 부족한 조작된 리뷰를 인공지능을 이용하여 생성하고 이를 기반으로 균형된 데이터 셋에서 기계학습을 학습하여 조작된 리뷰를 탐지하는 방안을 제시하였다. 파인 튜닝된 GPT-3는 초거대 인공지능으로 온라인 플랫폼의 리뷰를 생성하여 데이터 불균형 문제를 해결하는 오버샘플링 접근방법으로 사용되었다. GPT-3로 생성한 온라인 리뷰는 기존 리뷰를 기반으로 인공지능이 작성한 리뷰로써, 본 연구에서 사용된 로짓, 의사결정나무, 인공신경망의 성능을 개선시키는 것을 SMOTE와 단순 오버샘플링과 비교하여 실증분석을 통해서 확인하였다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제4권3호
• /
• pp.121-128
• /
• 2015

소수 의견을 포함하는 온라인 상품평은 긍정 또는 부정 일변도인 상품평에서는 찾기 어려운 유익한 정보를 내포하기도 한다. 본 논문에서는 주어진 상품평 집합 속에서 소수상품평을 검색하는 방법을 제안한다. 제안방법은 개별 상품평을 먼저 긍정/부정 상품평으로 자동분류한 뒤, 주어진 상품평 집합의 긍정/부정 상품평의 비대칭도를 계산하여 소수상품평을 검색한다. 소수상품평 검색에서는 긍정/부정 자동분류 성능이 소수상품평 검색성능에 영향을 주는데, 본 논문에서는 도메인에 특화된 감성사전과 그렇지 않은 일반적인 감성사전을 가지고 상품평을 긍정/부정으로 감성분류한 뒤 비대칭도를 계산하여 소수상품평 검색성능을 비교한다. 스마트폰과 영화를 다룬 온라인 영문 상품평에 대하여 도메인에 특화된 감성사전을 가지고 소수상품평 검색성능을 평가한 결과, F1점수는 각각 24.6%와 15.9%였고, 정확도는 각각 56.8%와 46.8%였다. 이는 스마트폰과 영화의 개별 상품평 긍정/부정 분류 정확도가 각각 85.3%와 78.8%일 때의 성능이다. 본 논문에서는 또 긍정/부정 자동분류 성능이 주어졌을 때의 이론적인 소수상품평 검색성능에 대해서도 논의한다.