Online Review Sentiment Research Articles

Consumer Sentiment and Hotel Aspect Preferences Across Trip Modes and Purposes

Travelers’ perceptions of hotels and their aspects have been the focus of much research and are often studied by analyzing consumers’ online reviews. Yet, little attention has been given to the effect of the trip mode, i.e., whether the person travels alone or with others, on travelers’ preferences as sentiment. Here, we study the influence of the trip mode and purpose using a mixed-methods approach. We conducted a user study to evaluate the perceptions of reviews across trip modes and found that star ratings do not consistently capture the sentiment in text reviews; on average, solo travelers’ text reviews are perceived as more negative than the star ratings they assigned, whether they travel for business or pleasure. We then analyzed over 137,000 reviews from TripAdvisor and Venere and found that a co-occurrence network approach naturally divides the text of reviews into hotel aspects. We used this result to measure the importance of hotel aspects across various traveler modes and purposes and identified significant differences in their preferences. These findings underscore the need for personalized marketing and services, highlighting the role of trip mode in shaping online review sentiment and traveler satisfaction.

Which Receives More Attention, Online Review Sentiment or Online Review Rating? Spillover Effect Analysis from JD.com.

Studies have found that competitive products' online review ratings (ORRs) have a spillover effect on the focal product's sales. However, the spillover effect of online review sentiment (ORS) as an essential component of online review analysis has yet to be studied. In this study, we analyze online review content from JD.com using the latent Dirichlet allocation to identify the product attribute topics that consumers are most concerned about. We then construct a baseline regression model of ORS and ORRs to explore the effects of online competitive product reviews on focal product sales. Moreover, we examine how the interaction between ORS and critical factors of online reviews affect sales. Our results indicate that the ORS of competitive products has a negative effect on focal product sales, and the effect is greater than the ORS and ORRs of focal products, respectively. In addition, the ORS of competitive products inhibits the sale of focal products as evaluations of product attributes become more positive or online review usefulness increases. We also find that the effect of ORRs of competitive products is not significant, which may be because clothing, as an experiential product, requires consumers to gain more information about specific usage scenarios before making a decision. This study provides a more accurate basis for consumer decision-making and offers retailers a novel approach to developing marketing strategies.

Modelling customer requirement for mobile games based on online reviews using BW-CNN and S-Kano models

In the mobile game industry, modelling customer requirements (CRs) is becoming indispensable for developers before improving games. Online reviews, as a common way of customer feedback, can reflect customers’ real experiences and preferences. Several studies have extracted important CRs from online reviews of mobile games, but they have not thoroughly analysed the correlation with customer satisfaction. In this study, we propose a CR modelling framework to extract and classify CRs from online reviews and analyse the influence of each CR on customer satisfaction to assist developers in improving games. First, we extract important CRs based on word embedding from online reviews. Then, we construct a BERT-Word2Vec-Convolutional Neural Network (BW-CNN) model to analyse the sentiment of online reviews by considering semantic features at Chinese character and word granularity. Finally, based on the Shapley Additive Explanation (SHAP) method, the S-Kano model is further proposed to classify the CRs considering customer attention, and the product improvement suggestions are provided for the corresponding mobile game developers based on the CR classification results. To verify the effectiveness and advantages of the proposed method, we crawled the online reviews of four Gacha games from taptap (www.taptap.cn) for case study and compared with the existing methods. The results show that our proposed framework is effective in modelling CRs from online reviews of mobile games. Meanwhile, the results comparison illustrate that customer attention has a significant influence on the CRs classification. In addition, we find that the S-Kano categories of CRs vary across games, and there is strong heterogeneity among games.

TRENDS AND PATTERNS IN ONLINE PURCHASE DECISION RESEARCH A LITERATURE REVIEW AND BIBLIOMETRIC ANALYSIS

Abstract : The internet has transformed the way consumers make purchase decisions, and with the increasing popularity of e-commerce, it has become vital to understand the factors that influence online purchase decisions. The COVID-19 pandemic has further accelerated the shift towards online shopping, making it even more crucial to understand the online purchase decision-making process. This bibliometric review examines 95 articles on the keyword "online purchase decision" from Scopus journals generated by Publish or Perish on April 25th, 2023, to identify the most influential authors and trends in this field from 2002 to 2023. The study reveals a steady rise in publications on this subject over time, with a significant surge in recent times. S. Yoon emerged as the most cited author in this field, followed by H.W. Kim and N. Hu. "Decision Support Systems" is the most productive journal in this field, and M. Ramesh is the most prolific author. The research variables have similarities and differences in focus and approach. For instance, trust, price, reviews, website characteristics, and the influence of social media are the focuses in these studies. The studies differ in their approaches and focuses. For example, (Yoon, 2002) examines the antecedents and consequences of trust, (Kim et al., 2012) compares price and trust, (Hu et al., 2014) analyzes online review sentiment, and (Luo et al., 2012) investigates the impact of website characteristics. There are also differences in the scope of the research, including cross-cultural approaches, the teenage market, gender, and potential customers vs. repeat customers. The findings of this study provide valuable insights into the key trends and patterns in online purchase decision-making research, which can inform marketing strategies and improve online salesKeyword : Keputusan Pembelian Online, Analisis Bibliometrik, Perilaku Konsumen

Machine Learning-Based Students’ Sentiment towards E-Learning amid COVID-19 Pandemic

The global COVID-19 pandemic has created exceptional difficulties for college and university students. With educational institutions shutting down and an abrupt shift to online learning, students are facing new challenges. Despite the advantages of e-learning, developing nations like Pakistan are experiencing major hurdles due to a lack of internet access and financial constraints. As a result, students' academic performance is being negatively impacted. Researchers are using sentiment analysis of social media platforms such as Facebook, Twitter, and YouTube to gain insight into student experiences during this challenging time. This study seeks to create a machine learning-based sentiment classification system that can determine the sentiment of higher education students in Pakistan towards e-learning classes during the COVID-19 pandemic. The primary objective of this research is to identify and categorize the challenges and problems that university students face in their e-learning classes. The research employed Multi-Nominal NB and Gaussian NB classifiers to examine the sentiment of online reviews and comments of students. We achieve 99% accuracy using Gaussian NB and 98% accuracy using Multinomial NB classifiers. According to the study's results, the COVID-19 epidemic has had a significant effect on the education system, with the vast majority of feedback being negative (70%). Only a small percentage of feedback was classified as neutral (29.9%) or positive (4%). This finding underscores the urgent need for better solutions for online learning during and after the pandemic. The study underscores the importance of addressing these unique challenges faced by students in developing countries during the pandemic. There is a pressing need for innovative solutions that can ensure that higher education remains effective and accessible in these challenging times. By utilizing sentiment analysis and machine learning, researchers can gain a better understanding of the experiences of students during this crisis. This understanding can inform targeted interventions and solutions that can improve academic performance and well-being for students in developing countries like Pakistan. The findings of this research provide important insights and recommendations for policymakers, educators, and stakeholders in the higher education sector.

Daily online review sentiment and hotel performance

PurposeThis paper aims to investigate the links between daily review sentiment and the hotel performance measures of occupancy rate (OR), average daily rate (ADR) and revenue per available room (RevPAR).Design/methodology/approachThe authors conducted review sentiment analyses in three moments (−1, −7 and −14 days) before arrival time using a data set of budget hotel performance and online reviews. The aim was to identify the effect of review sentiment in the budget hotel market on the three performance metrics.FindingsDaily sentiment positively affects ADR and negatively affects OR and RevPAR, but only up to a certain threshold, after which the trend reverses. Prices increase with the level of sentiment, and high prices lead to low OR and RevPAR only when the sentiment scores are low. When they are high, they are associated with low rates, which lead to high OR and RevPAR.Research limitations/implicationsDaily review sentiment can be viewed as a valuable “barometer” indicating a hotel’s daily operational effectiveness. Daily sentiment can thus allow hotel managers to adjust their dynamic pricing strategies more accurately.Originality/valueThis study identifies daily sentiment as an alternative predictor of hotel performance. In addition to the roles of valence and volume in the decision-making process, the authors found that daily review sentiment can be an “in-the-moment” factor with a high impact, encouraging consumers to complete their transactions. This study suggests that aggregated measures such as the total number of reviews and overall ratings of the hotel should not be the sole consideration in reputation management.

Weak tagging and imbalanced networks for online review sentiment classification

Sentiment classification aims to complete the automatic judgment task of text sentiment tendency. In the sentiment classification task of online reviews, traditional deep learning models require a large number of manually annotated samples of sentiment tendency for supervised training. Faced with massive online review data, the feasibility of manual tagging is worrisome. In addition, the traditional deep learning model ignores the imbalanced distribution of the number of classification samples, which will lead to a decline in classification performance in the practical application of the model. Considering that the online review data contains weak tagging information such as scores and labels, and the distribution is imbalanced, a weak tagging and imbalanced networks for online review sentiment classification is constructed. The experimental results show that the model significantly outperforms the traditional deep learning model in the sentiment classification task of hotel review data.

Examining influence factors of Chinese electric vehicle market demand based on online reviews under moderating effect of subsidy policy

Increasing the market penetration of electric vehicles (EVs) is an important step toward reducing carbon emissions in transport and achieving the 'double carbon' target. Currently, the public acceptance of EV is significantly low in China; thus, EVs have not yet become the mainstream of the new vehicles sold. Based on online review sentiment analysis, we investigated the public's perception of EV attributes, and their impact on Chinese EV sales. The public comments on the EV attributes of four brands—Volkswagen, NIO, Xiaopeng, and BYD—were obtained from Autohome and analyzed for sentiment, with the sentiment scores of the corresponding attributes as the independent variables and the monthly sales of EV during 2019–2021 as the dependent variables. A fixed-effects regression model with panel data was constructed to analyze the relationship between the public perceptions of EV attributes and EV sales. Considering the trend of Chinese government subsidies for EV in recent years, a moderating effect model was constructed with government subsidy policy as the moderating variable, sentiment perception as the independent variable, and EV sales as the dependent variable. The results of the study showed that the public perception of the four attributes of EVs—comfort, manipulation, space, and cost performance— significantly impact EV sales. The subsidy policy moderates the relationship between the public's perception of EV attributes and EV sales. The results of this study also reflect that the public is more concerned with the four attributes of comfort, manipulation, space, and cost performance when purchasing EVs. From a practical perspective, by understanding the public perception of EV attributes, government policy makers and EV marketers can appropriately adjust and change their existing policies and marketing strategies for promoting EVs.

Prediction Algorithm of User's Brand Conversion Intention Based on Fuzzy Emotion Calculation.

Branding is a magic weapon for enterprises to participate in international competition, and empowering enterprises through branding has become a national strategy in the new era. Economic and social development has won wide acclaim from the international community, but enterprises generally have the problem of being “big but not strong”, which is not matching with long history and great power influence. The brand bottleneck of Chinese enterprises has been highlighted. Recent brand theory research has been fruitful on the whole, but there are also some weak links, among which “the mechanism of enterprise brand value formation” is a research theme to be strengthened. This paper presents a number of suggestions for the formation of corporate brand value. The empirical analysis was conducted using valid data. The results found that: customer involvement behavior has a significant positive influence on customer citizenship behavior and customer experience value. Customer experience value has a significant positive influence on customer satisfaction and customer commitment. It plays a mediating effect in the relationship between the influence of customer involvement behavior on customer satisfaction and customer commitment, respectively. Customer satisfaction has a significant positive influence on customer commitment, and plays a mediating customer commitment has a significant positive effect on customer citizenship behavior and mediates the effect of customer experience value on customer citizenship behavior. The experimental results show that: the accuracy of crop color recognition by this method is high, and it has the advantages of faster computational efficiency and higher computational accuracy compared with other algorithms, thus verifying the reliability of the algorithm. Based on the fuzzy sentiment of online reviews, this paper improves the continuous use model ECM-ISC and formulates the inference rules of fuzzy affiliation function, and verifies the brand conversion intention and brand conversion type of cell phones by example calculation, which has good accuracy and generality and has important practical significance for brand marketing and early warning management. In addition, the use of brand economics in the study of corporate brand positioning is a development and innovation of brand economics.

Effect of online review sentiment on product sales: The moderating role of review credibility perception

Online reviews are of great importance in supporting the purchasing decision making of online consumers. With the prospering of e-commerce, increasingly fraudulent reviews are appearing on e-commerce platforms to mislead consumers, and an increasing number of consumers are worrying about being deceived in online shopping. However, little research effort has been invested in studying the impact of review credibility perception caused by online review fraud on product sales. To fill this research gap, this paper conducts empirical research on the moderating effect of consumers' credibility perception of online reviews. Here, review credibility perception, or consumers' credibility perception of online reviews, refers to the credibility or trust that consumers have in the reviews they are reading. Building on dual-process theory, this paper develops two hypotheses to understand why and how review credibility perception influence consumer decision-making behaviors. The empirical studies in this paper demonstrate that consumers' credibility perception of online reviews moderate the effect of review sentiment on product sales. This work also found that consumers' credibility perception of online reviews have a heterogeneous effect on product sales. The effect shifts the relationship between review sentiment and product sales from an inverted U-shape relationship to a monotonous positive relationship with the improvement of consumers’ credibility perception of online reviews. This study has great managerial implications for platforms and vendors to improve online review management.

The restaurant business value proposition: Intellectual text analysis of online customer reviews

The purpose of the paper is to study value proposition factors in online customer reviews that characterize consumer inquiries in the catering industry by using text-mining techniques. Research design focuses on the analysis of chain restaurants, which focuses on the usefulness of the service indicated in a quantitative system of scores, and pleasure factors for customers. Empirical data from 201 online reviews were collected from TripAdvisor service, which contains information about consumer preferences from St. Petersburg, Russia. In addition to text mining, the study used a correlation analysis of rating scores for chain restaurant customers, comparing the online reviews of satisfied and dissatisfied customers. The results of the study identified certain categories of factors of high importance for customers, which were revealed in both positive and negative online reviews. The results of the analysis help the restaurant management system to manage the value proposition properly taking into account various consumer factors of the preferences indicated in reviews. The research implications allow managers to purposefully select those factors that will achieve the desired effect in the relationship marketing. This study gives an opportunity to management specialists of the company explore and account for the emotional sentiment of online reviews.

Using Machine Learning to Predict the Sentiment of Online Reviews: A New Framework for Comparative Analysis

Online reviews are becoming increasingly important for decision-making. Consumers often refer to online reviews for opinions before making a purchase. Marketers also acknowledge the importance of online reviews and use them to improve product success. However, the massive amount of online review data, as well as its unstructured nature, is a challenge for anyone wanting to derive a conclusion quickly. In this paper, we propose a novel framework for gauging the ratings of online reviews using machine learning techniques. This framework uses a combination of text pre-processing and feature extraction methods. Here, we investigate four different aspects of the new framework. First, we assess the performance of single and ensemble classifiers in predicting sentiment—positive or negative—initially on a specific dataset (Yelp), but subsequently also on two other datasets (Amazon's product reviews and a movie review dataset). Second, using the best identified classifiers, we improve the accuracy with which neutral polarity can be predicted, an ability largely overlooked in the literature. Third, we further improve the performance of these classifiers by testing different pre-processing and feature extraction methods. Finally, we measure how well our deep learning approach performs on the same task compared to the best previously identified classifiers. Our extensive testing shows that the linear-kernel support vector machine, logistic regression and multilayer perceptron are the three best single classifiers in terms of accuracy, precision, recall, and F-measure. Their performance could be further improved if they were used as base classifiers for ensemble models. We also observe that several text pre-processing techniques—negation word identification, word elongation correction, and part of speech lemmatisation (combined with Terms Frequency and N-gram words)—can increase accuracy. In addition, we demonstrate that the general sentiment of lexicons such as SentiWordNet 3.0 and SenticNet 4 can be used to generate features with good results, although deep learning models can perform equally well. Experiments with different datasets confirm that our framework provides consistent outcomes. In particular, we have focused on improving the accuracy of neutral sentiment, and we conclude by showing how this can be achieved without sacrificing the accuracy of positive or negative ratings.

Influence of MOOC learners discussion forum social interactions on online reviews of MOOC

Although some studies have explored massive open online courses (MOOCs) discussion forums and MOOC online reviews separately, studies of both aspects are insufficient. Based on the theory of self-determination, this paper proposes research hypotheses that MOOC learning progress has a direct impact on MOOC online reviews and an indirect influence on MOOC online reviews through social interactions in discussion forums, as well. Coursera the largest MOOC platform, is selected as the empirical research object, and data from learners who participated in the MOOC discussion forum and provided MOOC online reviews from August 2016 to December 2019 are obtained from the most popular course, “Machine Learning”. After processing, data from 4376 learners are obtained. Then, according to research hypotheses, multi regression models are constructed accordingly. The results show that the length of MOOC online review text is affected by the MOOC learning progress, the number of discussion forum posts, the number of follow, the online review sentiment and MOOC rating. This study highlights the main factors that affect MOOC online reviews. As a result, some suggestions are put forward for the construction of MOOC.

An analysis of Airbnb online reviews: user experience in 16 U.S. cities

分析Airbnb网络评论：论美国16城市的用户体验研究目的本论文旨在深入了解Airbnb用户体验的复杂属性, 通过分析网络评论以及网络评分之间的关系来找寻规律和判断用户情感偏差。研究设计/方法/途径本论文采用大数据方式, 样本为美国16个城市的Airbnb用户网络评论; 本论文还对网络评分做出相关性分析。研究结果研究结果表明, Airbnb用户网络评论的关键主题为“清洁性”、“地理位置”、“居住体验”、“家”、“地点”、“招待主人”、“邻里”、以及“推荐”和用户总体来说对Airbnb有着积极体验。“清洁度”分数对“整体评论”有显著影响。研究理论限制/启示本论文属于开拓性研究作品; 未来项目可以采用混合方法来衡量用户体验与主题的关系。本论文的研究背景是美国, 未来研究可以做不同国家和区域的评论模式对比分析。此外, 未来研究还应该考虑Airbnb用户的统计数据指标、个性、以及住宿设施偏好。研究实际启示对于Airbnb住宿提供者来说, 保持设施的清洁和可使用性是非常重要的。对于Airbnb住宿主人和酒店经营者来说, 加强服务属性来提高网络好评是非常重要的。每个城市都应该基于“清洁度”和“整体评论”来开发不同的战略。研究原创性/价值本论文补充了现有Airbnb用户体验的文献, 通过Leximancer 4.0 软件来分析美国16个城市的网络评论

Analysis of book sales prediction at Amazon marketplace in India: a machine learning approach

Prediction of customer demand is an important part of Supply Chain Management, as it helps to avoid over or under production and reduces delivery time. In the context of e-commerce, accurate prediction of customer demand, typically captured by sales volume, requires careful analysis of multiple factors, namely, type of product, country of purchase, price, discount rate, free delivery option, online review sentiment etc., and their interactions. For e-tailers such as, Amazon, this kind of prediction capability is also extremely important in order to manage the supply chain efficiently as well as ensure customer satisfaction. This study investigates the efficacy of various modeling techniques, namely, regression analysis, decision-tree analysis and artificial neural network, for predicting the sales of books at amazon.in, using various relevant factors and their interactions as predictor variables. Sentiment analysis is carried out to measure the polarity of online reviews, which are included as predictors in these models. The importance of each independent predictor variable, such as discount rate, review sentiment etc., is analyzed based on the outcome of each model to determine top significant predictors which can be controlled by the marketer to influence sales. In terms of accuracy of prediction, the artificial neural network model is found to perform better than the decision-tree based model. In addition, the regression analysis, with and without sentiment and interaction factors, generates comparable results. The comparative analysis of these models reveals several significant findings. Firstly, all three models confirm that review volume is the most important and significant predictor of sales of books at amazon.in. Secondly, discount rate, discount amount and average ratings have minimal or insignificant effect on sales prediction. Thirdly, both negative sentiment and positive sentiment of the reviews are individually significant predictors as per regression and decision-tree model, but they are not significant at all as per neural network model. This observation from the neural network model is contrary to the extant research which claims that both negative and positive sentiment are significant with the former having more influence in predicting sales. Finally, the interaction effects of review volume with negative and positive sentiment are also found to be significant predictors as per all three models. Hence, overall, out of various factors used for sales prediction of books, review volume, negative sentiment, positive sentiment and their interactions are found to be the most significant ones across all models. The results of this study can be utilized by online sellers to accurately predict the sales volume by adjusting these significant factors, thereby managing the supply chain effectively.

Sentiment analysis for online reviews using conditional random fields and support vector machines

Sentiment analysis of online reviews is an important way of mining useful information from the Internet. Despite several advantages, the accuracy of sentiment analysis based on a domain dictionary relies on the comprehensiveness and accuracy of the dictionary. Instead of creating a domain dictionary, we propose an approach for online review sentiment classification, which uses a conditional random field algorithm to extract the emotional characteristics from fragments of the review. The characteristic (feature) words are then weighted asymmetrically before a support vector machine classifier is used to obtain the sentiment orientation of the review. In our experiments, the average accuracy reached 90%, showing that using sentiment feature fragments instead of whole reviews and weighting the characteristic words asymmetrically can improve the sentiment classification accuracy.

Exploring performance of clustering methods on document sentiment analysis

Clustering is a powerful unsupervised tool for sentiment analysis from text. However, the clustering results may be affected by any step of the clustering process, such as data pre-processing strategy, term weighting method in Vector Space Model and clustering algorithm. This paper presents the results of an experimental study of some common clustering techniques with respect to the task of sentiment analysis. Different from previous studies, in particular, we investigate the combination effects of these factors with a series of comprehensive experimental studies. The experimental results indicate that, first, the K-means-type clustering algorithms show clear advantages on balanced review datasets, while performing rather poorly on unbalanced datasets by considering clustering accuracy. Second, the comparatively newly designed weighting models are better than the traditional weighting models for sentiment clustering on both balanced and unbalanced datasets. Furthermore, adjective and adverb words extraction strategy can offer obvious improvements on clustering performance, while strategies of adopting stemming and stopword removal will bring negative influences on sentiment clustering. The experimental results would be valuable for both the study and usage of clustering methods in online review sentiment analysis.