• 한국정보시스템학회지:정보시스템연구
• /
• 제31권2호
• /
• pp.1-18
• /
• 2022

Purpose Customer review is a major factor in choosing certain restaurants. This study investigates the key factors affecting customer's evaluation about restaurants. With the recent intensification of competition among restaurants in the service industry, the analysis results are expected to provide in-depth insights for enhancing customer experiences. Design/methodology/approach We collected information and reviews provided at the restaurants in the Kakao Map platform. The information collected is based on the information of 3,785 restaurants in Daegu registered on Kakao Map. Based on the information collected, seven independent variables, including number of rating registered, number of reviews, presence or absence of safe restaurants, presence or absence of a posting about holding facilities, presence or absence of a posting about business hours, presence or absence of a posting about hashtags, and presence or absence of break times, were used. Dependent variable is restaurant rating. Multiple regression between independent variables and restaurant rating was carried out. Findings The results of the study confirmed that number of rating registered, presence or absence of a posting about business hours, and presence or absence of a posting about hash tags have an positive effects on the restaurant rating. The number of reviews had a negative effect on the restaurant rating. In addition, in order to confirm the role of customer's reviews, we carried out LDA topic modeling. We divided the topics into the positive review and the negative reviews.

• 패션비즈니스
• /
• 제20권4호
• /
• pp.207-225
• /
• 2016

This study aims to analyze the impact of negative review type, brand reputation and opportunity scarcity perception, on preferences of fashion products in social commerce. For the above evaluation, we used the 2 (negative review type: objective/subjective) ${\times}2$ (brand reputation: high/low) ${\times}2$ (opportunity scarcity perception: high/low) model, designed with three mixed elements. We enrolled 260 women in their 20s and 30s, who live in Seoul and have used social commerce; a final total of 207 subjects were considered for analysis. The data were analyzed using the SPSS 18 program and reliability test, t-test and three-way ANOVA were performed. Following observations were made: First, preferences were higher when the subjects read objective negative reviews than subjective negative reviews, and when a fashion product was from a brand of high reputation than a brand of low reputation. Second, the interaction effect between negative review type and brand reputation was greater among the subjects whose opportunity scarcity perception is high, than those having low opportunity scarcity perception. Thus, we conclude that the social commerce should encourage consumers to write more objective reviews, and fashion brands should manage their reputations well. Also, social commerce can use scarcity messages aggressively to increase preferences of global fashion luxury goods, which is actively marketed in social commerce since 2015.

• 디지털산업정보학회논문지
• /
• 제15권1호
• /
• pp.87-97
• /
• 2019

Reaction of people is importantly considered about specific case as a social network service grows. In the previous research on analysis of social network service, they predicted tendency of interesting topic by giving scores to sentences written by user. Based on previous study we proceeded research of sentiment analysis for social network service's sentences, which predict the result as positive or negative for movie reviews. In this study, we used movie review to get high accuracy. We classify the movie review into positive or negative based on the score for learning. Also, we performed embedding and morpheme analysis on movie review. We could predict learning result as positive or negative with a number 0 and 1 by applying the model based on learning result to social network service. Experimental result show accuracy of about 80% in predicting sentence as positive or negative.

• 한국콘텐츠학회논문지
• /
• 제21권4호
• /
• pp.111-122
• /
• 2021

본 연구는 우리나라 대표적인 지역화폐인 동백전과 인천e음 앱 사용자 리뷰를 분석하여 지역화폐 사용자의 긍정/부정 요인을 파악하고, 이를 바탕으로 마케팅 전략을 수립하였다. 앱 사용자 리뷰를 별점을 기준으로 하여 긍정과 부정으로 분류하고 각각 워드클라우드, 토픽모델링, 소셜 네트워크 분석을 수행하였다. 그 결과, 동백전과 인천e음 부정 리뷰에서는 공통적으로 앱 사용과 카드 발급에 대한 불만이 주로 나타났으며, 긍정 리뷰에서는 '캐시백'에 대한 만족감과 함께 '지역경제'와 '소상공인'과 같은 키워드의 출현으로 지역화폐 사용자들은 자신의 소비가 지역경제 활성화에 도움이 된다고 인식하여 지역화폐를 사용하는 데 있어 만족감을 느끼는 것으로 나타났다. 본 연구의 분석결과로 파악된 만족/불만족 요인을 기반으로 개선해야 할 점과 더욱 강화해야 할 점을 파악하고, 이에 적절한 마케팅 전략을 도출하였다. 본 연구에서 활용한 텍스트 마이닝 방법과 연구 결과는 실질적으로 지역화폐 담당 공무원들과 마케터들에게 지역화폐에 대한 유의미한 정보를 제공해 줄 수 있다.

• 한국의류산업학회지
• /
• 제18권1호
• /
• pp.15-26
• /
• 2016

Many brands are attempting to use consumers as a part of their marketing strategies, due to the fashion industry's sensitive response to consumers' reaction. In addition, due to the popularity of e-WOM(electronic Word-Of-Mouth), fashion brands are highly sensitive to their supporters' online reviews. Amid this background, the main objectives of this study are as follows: 1) to analyze the effect of online reviews' attributes and valences on forming an impression about a fashion brand; 2) to examine the online re-WOM(word-of-mouth) effect of online reviews by fashion brand supporters on brand attitude; and 3) to measure the moderating effect of fashion involvement in online re-WOM intention. In order to verify the research model and to test the proposed hypotheses, a 2 (utilitarian vs. hedonic review attributes) by 2 (positive vs. negative review valences) model is constructed and gathers 215 respondents. The results demonstrate that consumers form the highest reliable impression based on utilitarian and negative online reviews. However, there is no relationship between the types of online reviews and the formation of a favorable impression. Findings also reveal that the impression formed by online reviews has a positive effect on re-WOM intention, contributing to brand attitude. In addition, the hypothesis about the moderating effect produced by fashion involvement on re-WOM is supported. In conclusion, these results suggest that online reviews by fashion brand supporters have a powerful effect on forming a consumer's impression towards a fashion brand, affecting re-WOM intention and brand attitude.

• Journal of Multimedia Information System
• /
• 제9권1호
• /
• pp.61-68
• /
• 2022

The importance of online reviews is prevalent as more people access goods or places online and make decisions to visit or purchase. However, such reviews are generally provided by short sentences or mere star ratings; failing to provide a general overview of customer preferences and decision factors. This study explored and broke down restaurant reviews found on Google Maps. After collecting and analyzing 5,427 reviews, we vectorized the importance of words using the TF-IDF. We used a random forest machine learning algorithm to calculate the coefficient of positivity and negativity of words used in reviews. As the result, we were able to build a dictionary of words for positive and negative sentiment using each word's coefficient. We classified words into four major evaluation categories and derived insights into sentiment in each criterion. We believe the dictionary of review words and analyzing the major evaluation categories can help prospective restaurant visitors to read between the lines on restaurant reviews found on the Web.

• 지능정보연구
• /
• 제22권1호
• /
• pp.63-82
• /
• 2016

최근 정보기술의 발달로 인해 소비자들은 온라인상에서 많은 정보를 쉽고 빠르게 획득할 수 있다. 소비자가 제품 구매시에는 소비자들이나 전문가들이 작성한 제품 후기 정보를 주로 탐색한다. 기존의 연구들이 소비자들이 창출한 제품 후기 중심으로 주로 진행되어 왔기 때문에, 전문가 제품 후기의 영향력에 대해서는 상대적으로 소수의 연구들만 존재하고 있다. 본 연구는 전문가가 생성하는 제품 후기에 초점을 맞추어, 방대한 실제 비정형데이터인 전문가의 후기를 어떻게 언어학적인 차원과 심리학적인 차원으로 나눌 수 있는지의 방법론을 제안하며, 실제 전문가 제품 후기를 사용하여 의미 있는 다섯 가지 차원의 새로운 변수들을 도출하였다. 그 결과 소비자들이 전문가 후기에서 반응하고 있는 언어적 특성은 제품에 대한 깊이 있는 정보의 양이나 충분한 설명을 나타내는 변수인 Review Depth, 그리고 전문가가 기술하는 방식이 제품에 대한 확신이 없는 듯한 말투를 나타내는 변수인 Lack of Assurance는 소비자의 전반적인 제품평가에 유의한 상관관계가 있는 것으로 밝혀졌다. 또한, 제품에 대한 칭찬이나 긍정적인 면을 서술하는 방식인 Positive Polarity가 소비자의 제품 평가에 영향을 미치지 않았지만, 전문가가 하는 제품에 대한 비관적인 평가인 Negative Polarity는 소비자들의 평가와 유의한 음의 상관관계가 있었다는 점이다. 전문가가 스토리텔링 관점에서 자주 사용하는 Social Orientation 특성은 유의한 관계를 미치지 못함이 밝혀졌다. 본 연구는 새로운 방법론을 제안하고 이를 실제로 활용한 결과를 보여준다는 차원에서 이론적이고 실무적인 공헌을 가진다.

• Asia Marketing Journal
• /
• 제21권2호
• /
• pp.1-19
• /
• 2019

Prior studies on the effects of online consumer reviews have mainly focused on review valence, but little research has investigated how two-sided (both positive and negative) and one-sided (only positive) reviews influence consumers' response to online review. In addition, little attention has been paid to how sponsorship presence (firm-sponsored reviews vs. consumer-voluntary reviews) influences individuals' attitude toward online review. Unlike consumer-voluntary reviews without any monetary incentive, firm-sponsored reviews include messages about brands providing monetary compensation. This study examines whether review valence (two-sidedness vs. one-sidedness) influences attitude toward online review via its influence on review credibility. Further, this study examines whether sponsorship presence affects when review valence influences attitude toward review. Thus, this research investigates the effect of review valence on attitude toward review and the moderating role of sponsorship presence in the relationship between review valence and attitude toward review. The first experiment reveals that attitude toward review is more favorable when the review is two-sided (vs. one-sided). The second study demonstrates that differences between the two-sided and the one-sided review occur only for firm-sponsored reviews, not for consumer-voluntary reviews. The theoretical and practical implications are also discussed.

• 한국정보시스템학회지:정보시스템연구
• /
• 제31권4호
• /
• pp.161-187
• /
• 2022

Purpose This study analyzed the topic of game review contents and how the characteristics of game reviews affect the reviews helpfulness. In addition, this study explore the content of game reviews according to the game's sales strategy such as early access strategy and releasing without early access. Design/methodology/approach We collected a list of 3,572 action genre games released in 2020. 58,336 online reviews were collected by random sampling 50 reviews in each games, and topic modeling was performed on those reviews. We dynamized the results of topic modeling and analyzed the effect on review helpfulness with multiple regression analysis. Findings The results of analysis indicate that the longer the review is or the shorter the time it is written, the more helpful the review is. In addition the topic with positive and negative review has a significant effect on the review helpfulness. As a result of exploratory analysis, games from early access had relatively fewer reviews of story-related topics than games that were released without early access. These findings can present direct guidelines for collecting specific opinions from customers in the game industry when releasing games.

• International Journal of Computer Science & Network Security
• /
• 제23권11호
• /
• pp.83-92
• /
• 2023

The majority of product users rely on the reviews that are posted on the appropriate website. Both users and the product's manufacturer could benefit from these reviews. Daily, thousands of reviews are submitted; how is it possible to read them all? Sentiment analysis has become a critical field of research as posting reviews become more and more common. Machine learning techniques that are supervised, unsupervised, and semi-supervised have worked very hard to harvest this data. The complicated and technological area of feature engineering falls within machine learning. Using deep learning, this tedious process may be completed automatically. Numerous studies have been conducted on deep learning models like LSTM, CNN, RNN, and GRU. Each model has employed a certain type of data, such as CNN for pictures and LSTM for language translation, etc. According to experimental results utilizing a publicly accessible dataset with reviews for all of the models, both positive and negative, and CNN, the best model for the dataset was identified in comparison to the other models, with an accuracy rate of 81%.