A concept-level approach to the analysis of online review helpfulness

Helpfulness of online reviews serves multiple needs of different Web users. Several types of factors can drive reviews' helpfulness. This study focuses on uninvestigated factors by looking at not just the quantitative factors (such as the number of concepts), but also qualitative aspects of reviewers (including review types such as the regular, comparative and suggestive reviews and reviewer helpfulness) and builds a conceptual model for helpfulness prediction. The set of 1500 reviews were randomly collected from TripAdvisor.com across multiple hotels for analysis. A set of four hypotheses were used to test the proposed model. Our results suggest that the number of concepts contained in a review, the average number of concepts per sentence, and the review type contribute to the perceived helpfulness of online reviews. The regular reviews were not statistically significant predictors of helpfulness. As a result, review types and concepts have a varying degree of impact on review helpfulness. The findings of this study can provide new insights to e-commerce retailers in understanding the importance of helpfulness of reviews.     

A study of factors that contribute to online review helpfulness

Helpfulness of online reviews is a multi-faceted concept that can be driven by several types of factors. This study was designed to extend existing research on online review helpfulness by looking at not just the quantitative factors (such as word count), but also qualitative aspects of reviewers (including reviewer experience, reviewer impact, reviewer cumulative helpfulness). This integrated view uncovers some insights that were not available before. Our findings suggest that word count has a threshold in its effects on review helpfulness. Beyond this threshold, its effect diminishes significantly or becomes near non-existent. Reviewer experience and their impact were not statistically significant predictors of helpfulness, but past helpfulness records tended to predict future helpfulness ratings. Review framing was also a strong predictor of helpfulness. As a result, characteristics of reviewers and review messages have a varying degree of impact on review helpfulness. Theoretical and practical implications are discussed.     

The determinants of helpfulness of online reviews

More and more people are gravitating to reading online product reviews prior to making purchasing decisions. Because a number of reviews that vary in usefulness are posted every day, much attention is being paid to measuring their helpfulness. The goal of this paper is to investigate the various determinants of the helpfulness of reviews, and it also intends to examine the moderating effect of product type, that is, the experience or search goods in relation to the helpfulness of online reviews. The study results show that reviewer reputation, the disclosure of reviewer identity, and review depth positively affect the helpfulness of an online review. The moderating effects of product type exist for these determinants on helpfulness. That is, the number of reviews for a product and the disclosure of reviewer identity have a greater influence on the helpfulness for experience goods, while reviewer reputation, review extremity, and review depth are more important for helpfulness in relation to search goods. The interaction effects exist for average review rating and average review depth for a product with review helpfulness on product sales. The results of the study will identify helpful online reviews and assist in designing review sites effectively.     

The effect of content depth and deviation on online review helpfulness: Evidence from double-hurdle model

How does the content of a product review shape its perceived value? We propose two information theory-based constructs derived from probabilistic topic models and show their relationship with review helpfulness. The first construct, content depth, quantifies the breadth-depth tradeoff of a review and has an informational influence on readers’ voting behavior. The second construct, content deviation, indicates the deviance of the review content in comparison with others and exerts a normative influence on readers’ voting behavior. Noting the possibility that a review can get voted but has zero helpfulness score, we use a double-hurdle model to simultaneously estimate the probability of a review being voted and its helpfulness. The analyses on three product categories show that reviews with more depth and less content deviation are rated more helpful. Further, the relationships are moderated by a number of factors, including the deviation of numerical rating, recency of the review, and the reputation of the reviewer. The research contributes to the literature by showing how the content of a review and the interaction of content and numerical ratings jointly create value for consumers.     

Effects of membership tier on user content generation behaviors: evidence from online reviews

Online shopping websites typically classify customers into different membership tiers in their customer relationship management systems. This study investigates the effects of membership tiers on user content generation behaviors in the context of an electronic commerce marketplace that has a membership tier program and an online review system. Grounded in theories related to status, our study hypothesizes the effects of membership tiers on user content generation behaviors as well as the helpfulness of the content they generated in the context of online reviews. We collected online data from a world-leading shopping website. The results from our empirical analyses indicate that membership tier has a positive effect on review rating and review delay, whereas it has a negative effect on review depth. Additionally, we tested mediation effects of review rating, depth and delay between membership tiers and review helpfulness, and found that membership tier negatively affected review helpfulness indirectly. Interestingly, reviews posted by high-status customers are perceived as more helpful than those of others when we controlled for review characteristics. This study contributes to research on online product reviews and customer relationship management.     

Consumer perceptions of information helpfulness and determinants of purchase intention in online consumer reviews of services

Online consumer reviews offer an unprecedented amount of information for consumers to evaluate services before purchase. We use the dual process theory to investigate consumer perceptions about information helpfulness (IH) in electronic word-of-mouth (eWOM) contexts. Results highlight that popularity signals, two-sided reviews, and expert sources (but not source trustworthiness) are perceived as helpful by consumers to assess service quality and performance. Although two-sided reviews exercise a significant influence on perceived IH, their influence on purchase intention was indirectly mediated by IH. IH predicts purchase intention and partially mediates the relationship between popularity signals, source homophily, source expertise, and purchase intention.     

Predicting the helpfulness of online reviews using multilayer perceptron neural networks

With the great development of e-commerce, users can create and publish a wealth of product information through electronic communities. It is difficult, however, for manufacturers to discover the best reviews and to determine the true underlying quality of a product due to the sheer volume of reviews available for a single product. The goal of this paper is to develop models for predicting the helpfulness of reviews, providing a tool that finds the most helpful reviews of a given product. This study intends to propose HPNN (a helpfulness prediction model using a neural network), which uses a back-propagation multilayer perceptron neural network (BPN) model to predict the level of review helpfulness using the determinants of product data, the review characteristics, and the textual characteristics of reviews. The prediction accuracy of HPNN was better than that of a linear regression analysis in terms of the mean-squared error. HPNN can suggest better determinants which have a greater effect on the degree of helpfulness. The results of this study will identify helpful online reviews and will effectively assist in the design of review sites.     

Effect of other visible reviews’ votes and personality on review helpfulness evaluation: an event-related potentials study

Electronic Commerce Research - Few studies have investigated the mechanism underlying the connection between votes and review helpfulness. A within-subject experiment with a between-group factor of...     

基于用户回复内容观点支持度的评论有用性计算

针对有监督评论有用性预测方法中的训练数据集难以构造,以及无监督方法缺乏对情感信息支撑的问题,提出基于语义和情感信息构建一种无监督模型,用于对评论有用性进行预测,同时考虑了评论和评论下回复内容对观点的支持度用来计算观点的有用性得分,进而得到评论的有用性。同时,提出结合句法分析和改进潜在狄利克雷分配（LDA）模型的评论摘要方法用于评论有用性预测模型中的观点提取,基于句法分析结果构建must-link和cannot-link两种约束条件指导主题模型学习,在保证召回率的同时提高模型准确率。该方法在实验数据集上能取得70%左右的F1值和90%左右的排序准确率,且实例应用也表明该方法对结果具有较好的解释性。     

Assessing the product review helpfulness: Affective-Cognitive evaluation and the moderating effect of feedback mechanism

Product review length has been demonstrated as one of the key factors that influence the product review helpfulness. However, we have little knowledge of a deeper understanding on how consumers process the product review length to assess product review helpfulness. Anchoring on the human’s affective-cognitive model of decision-making and the prominent coping approaches, this research revisits this key issue from consumers’ affect-oriented and cognition-oriented processing perspective. Our findings show that (1) consumers apply both the affect-oriented and cognition-oriented processing to assess the helpfulness of product review, (2) a significant inverted U-shape relationship between the review length and the review helpfulness, and interestingly, (3) such a relationship is further moderated by whether the product review author has provided a response to consumers’ comments. These findings not only provide further robust evidence to the consideration of both product review length and feedback, but also suggest a refinement of the underlying mechanism on how consumers process product review length to assess product review helpfulness.     

消费者在线评论有用性影响因素模型研究

消费者在线评论的价值已经得到消费者和在线零售商的公认,对评论有用性的研究已经成为新的研究热点。从减少消费者决策风险出发,在感知诊断性概念基础上定义了评论有用性概念,构建了一个评论有用性影响因素模型。从传播说服理论的维度考察,评论等级、评论长度、好评率和使用互联网经验是影响评论有用性的重要因素。商品类型对评论有用性具有调节作用。     

Opinion helpfulness prediction in the presence of “words of few mouths”

This paper identifies a widely existing phenomenon in social media content, which we call the “words of few mouths” phenomenon. This phenomenon challenges the development of recommender systems based on users’ online opinions by presenting additional sources of uncertainty. In the context of predicting the “helpfulness” of a review document based on users’ online votes on other reviews (where a user’s vote on a review is either HELPFUL or UNHELPFUL), the “words of few mouths” phenomenon corresponds to the case where a large fraction of the reviews are each voted only by very few users. Focusing on the “review helpfulness prediction” problem, we illustrate the challenges associated with the “words of few mouths” phenomenon in the training of a review helpfulness predictor. We advocate probabilistic approaches for recommender system development in the presence of “words of few mouths”. More concretely, we propose a probabilistic metric as the training target for conventional machine learning based predictors. Our empirical study using Support Vector Regression (SVR) augmented with the proposed probability metric demonstrates advantages of incorporating probabilistic methods in the training of the predictors. In addition to this “partially probabilistic” approach, we also develop a logistic regression based probabilistic model and correspondingly a learning algorithm for review helpfulness prediction. We demonstrate experimentally the superior performance of the logistic regression method over SVR, the prior art in review helpfulness prediction.     

基于贝叶斯网络的在线商品评价质量评估

随着电子商务领域的迅速发展,在线商品评价规模日益庞大,评价质量参差不齐,用户难以筛选有用评价信息做出购买决策,因此如何有效识别高质量评价信息成为重要议题。以在线商品评价的有用性投票为基础定义评价质量,使用贝叶斯网络表示在线商品评价的相似性及不确定性,通过对在线商品评价信息进行多维度特征统计,构建在线商品评价质量评估模型,使用概率推理机制对在线商品评价质量进行分类预测,并给出评价质量分类置信度。在真实数据集上验证模型有效性及高效性。     

An information gain-based approach for recommending useful product reviews

Recently, many e-commerce Web sites, such as Amazon.com, provide platforms for users to review products and share their opinions, in order to help consumers make their best purchase decisions. However, the quality and the level of helpfulness of different product reviews are not disclosed to consumers unless they carefully analyze an immense number of lengthy reviews. Considering the large amount of available online product reviews, this is an impossible task for any consumer. Therefore, it is of vital importance to develop recommender systems that can evaluate online product reviews effectively to recommend the most useful ones to consumers. This paper proposes an information gain-based model to predict the helpfulness of online product reviews, with the aim of suggesting the most suitable products and vendors to consumers. Reviews are analyzed and ranked by our scoring model and reviews that help consumers better than others will be found. In addition, we also compare our model with several machine learning algorithms. Our experimental results show that our approach is effective in ranking and classifying online product reviews.     

When Consistency Matters: The Effect of Valence Consistency on Review Helpfulness

When evaluating the helpfulness of online reviews, review valence is a particularly relevant factor. This research argues that the influence of review valence is highly dependent on its consistency with the valence of other available reviews. Using both field and experimental data, this paper show that consistent reviews are perceived as more helpful than inconsistent reviews, independent of them being positive or negative. Experiments show that this valence consistency effect is driven by causal attributions, such that consistent reviews are more likely to be attributed to the actual product experience, while inconsistent reviews are more likely to be attributed to some reviewer idiosyncrasy. Supporting the attribution theory framework, reviewer expertise moderates the effect of consumers' causal attributions on review helpfulness.     

Helpfulness of user-generated reviews as a function of review sentiment,product type and information quality

Helpfulness of user-generated reviews has not been studied adequately in terms of the interplay between review sentiment (favorable, unfavorable and mixed) and product type (search and experience). Moreover, the ways in which information quality relates to review helpfulness remain largely unknown. Hence, this paper seeks to answer the following two research questions: (1) How does the helpfulness of user-generated reviews vary as a function of review sentiment and product type? (2) How does information quality relate to the helpfulness of user-generated reviews across review sentiment and product type? Data included 2190 reviews drawn from Amazon for three search products—digital cameras, cell phones, and laser printers—as well as three experience products—books, skin care, and music albums. Review sentiment was ascertained based on star ratings. Investigation of the research questions relied on the statistical procedures of analysis of variance, and multiple regression. Review helpfulness was found to vary across review sentiment independent of product type. Besides, the relationship between information quality and review helpfulness was found to vary as a function of review sentiment as well as product type. The paper concludes with a number of implications for research and practice.     

Can online user reviews be more helpful? Evaluating and improving ranking approaches

Given the sharply increasing number of online reviews, the selection of strategies by review-hosting firms to help users access more helpful reviews is an intriguing but insufficiently studied issue. We first propose a model to help us understand how reviews receive helpful votes (HV) and non-helpful votes. According to this model, the performances of different ranking approaches are compared using several simulated datasets with empirical features. In addition to three well-known ranking approaches, we develop a novel approach based on Bayesian statistics that is easy to implement in existing websites and can be combined with other content recommendation techniques to determine the prior belief in online reviews. More importantly, we suggest two simple ways to enhance existing ranking approaches. The numerical evidence demonstrates the advantages of two enhanced approaches, as indicated by higher helpful ratios and a reduced Matthew effect. These findings have important practical implications for consumers, online retailers, and review-hosting firms.     

融合情感极性与信任函数的虚假评论检测方法

在线评论是用户判断商品质量的一个依据。虚假评论严重影响了消费者的购买行为,现有的虚假评论检测方法从文本出发,忽略了评分的虚假性,评分通常是不精确和不确定的,对虚假评论检测效果不佳。提出融合情感极性与信任函数的虚假评论检测方法(EP-BFRD),利用信任函数处理给定评论者评分中的不确定性和不准确性,考虑与其他评分者提供的评分的相似性,以检测误导性,并判断评论文本情感极性与评分一致性。综合考虑信任函数处理的结果以及评分与文本情感一致性的结果来判断评论的虚假性。在一个真实的数据库上进行实验,实验表明该方法可有效解决虚假评论检测问题。     

Towards Easy Comparison of Local Businesses Using Online Reviews

With the rapid development of e‐commerce, there is an increasing number of online review websites, such as Yelp, to help customers make better purchase decisions. Viewing online reviews, including the rating score and text comments by other customers, and conducting a comparison between different businesses are the key to making an optimal decision. However, due to the massive amount of online reviews, the potential difference of user rating standards, and the significant variance of review time, length, details and quality, it is difficult for customers to achieve a quick and comprehensive comparison. In this paper, we present E‐Comp, a carefully‐designed visual analytics system based on online reviews, to help customers compare local businesses at different levels of details. More specifically, intuitive glyphs overlaid on maps are designed for quick candidate selection. Grouped Sankey diagram visualizing the rating difference by common customers is chosen for more reliable comparison of two businesses. Augmented word cloud showing adjective‐noun word pairs, combined with a temporal view, is proposed to facilitate in‐depth comparison of businesses in terms of different time periods, rating scores and features. The effectiveness and usability of E‐Comp are demonstrated through a case study and in‐depth user interviews.