Manipulation of online reviews: An analysis of ratings, readability, and sentiments

As consumers become increasingly reliant on online reviews to make purchase decisions, the sales of the product becomes dependent on the word of mouth (WOM) that it generates. As a result, there can be attempts by firms to manipulate online reviews of products to increase their sales. Despite the suspicion on the existence of such manipulation, the amount of such manipulation is unknown, and deciding which reviews to believe in is largely based on the reader's discretion and intuition. Therefore, the success of the manipulation of reviews by firms in generating sales of products is unknown. In this paper, we propose a simple statistical method to detect online reviews manipulation, and assess how consumers respond to products with manipulated reviews. In particular, the writing style of reviewers is examined, and the effectiveness of manipulation through ratings, sentiments, and readability is investigated. Our analysis examines textual information available in online reviews by combining sentiment mining techniques with readability assessments. We discover that around 10.3% of the products are subject to online reviews manipulation. In spite of the deliberate use of sentiments and ratings in manipulated products, consumers are only able to detect manipulation taking place through ratings, but not through sentiments. The findings from this research ensue a note of caution for all consumers that rely on online reviews of books for making purchases, and encourage them to delve deep into the book reviews without being deceived by fraudulent manipulation.     

Text feature selection for sentiment classification of Chinese online reviews

In order to meet the requirement of customised services for online communities, sentiment classification of online reviews has been applied to study the unstructured reviews so as to identify users’ opinions on certain products. The purpose of this article is to select features for sentiment classification of Chinese online reviews with techniques well performed in traditional text classification. First, adjectives, adverbs and verbs are identified as the potential text features containing sentiment information. Then, four statistical feature selection methods, such as document frequency (DF), information gain (IG), chi-squared statistic (CHI) and mutual information (MI), are adopted to select features. After that, the Boolean weighting method is applied to set feature weights and construct a vector space model. Finally, a support vector machine (SVM) classifier is employed to predict the sentiment polarity of online reviews. Comparative experiments are conducted based on hotel online reviews in Chinese. The results indicate that the highest accuracy of the sentiment classification of Chinese online reviews is achieved by taking adjectives, adverbs and verbs together as the feature. Besides that, different feature selection methods make distinct performances on sentiment classification, as DF performs the best, CHI follows and IG ranks the last, whereas MI is not suitable for sentiment classification of Chinese online reviews. This conclusion will be helpful to improve the accuracy of sentiment classification and be useful for further research.     

面向中文网络评论情感分类的集成学习框架

该文针对中文网络评论情感分类任务,提出了一种集成学习框架。首先针对中文网络评论复杂多样的特点,采用词性组合模式、频繁词序列模式和保序子矩阵模式作为输入特征。然后采用基于信息增益的随机子空间算法解决文本特征繁多的问题,同时提高基分类器的分类性能。最后基于产品属性构造基分类器算法综合评论文本中每个属性的情感信息,进而判别评论的句子级情感倾向。实验结果表明了该框架在中文网络评论情感分类任务上的有效性,特别是在Logistic Regression分类算法上准确率达到90.3%。     

Towards effective online review systems in the Chinese context: A cross-cultural empirical study

Online reviews, as one kind of quality indicator of products or service, are becoming increasingly important in influencing purchase decisions of prospective consumers on electronic commerce websites. With the fast growth of the Chinese e-commerce industry, it is thus indispensable to design effective online review systems for e-commerce websites in the Chinese context, by taking into account cultural factors. In this paper, we conduct two empirical studies on online reviews. Firstly, we study how culture differences across countries (i.e., China and the USA) impact the way in which consumers provide online reviews. Secondly, we investigate the impact of online reviews on product sales in the Chinese context, and show that directly copying the ideas of successful online review systems in the USA will deteriorate the effectiveness of the systems in China. Finally, we propose several suggestions for the development of effective online review systems in the Chinese context based on the results of our two empirical studies and the findings in previous studies.     

Product recommendation with latent review topics

Online customer reviews complement information from product and service providers. While the latter is directly from the source of the product and/or service, the former is generally from users of these products and/or services. Clearly, these two information sets are generated from different perspectives with possibly different sets of intentions. For a prospective customer, both these perspectives together provide a complementary set of information and support their purchase decisions. Given the different perspective and incentive structure, the information from these two source sets tends to be necessarily biased, clearly with the high probability of negative information omission from that provided by the product/service providers. Moreover, customers oftentimes face information overload during their attempts at deciphering existing online customer reviews. We attempt to alleviate this through mining hidden information in online customer reviews. We use a variant of the Latent Dirichlet Allocation (LDA) model and clustering to generate equivalent options that the customer could then use in their purchase decisions. We illustrate this using online hotel review data.     

Weakness Finder: Find product weakness from Chinese reviews by using aspects based sentiment analysis

Finding the weakness of the products from the customers’ feedback can help manufacturers improve their product quality and competitive strength. In recent years, more and more people express their opinions about products online, and both the feedback of manufacturers’ products or their competitors’ products could be easily collected. However, it’s impossible for manufacturers to read every review to analyze the weakness of their products. Therefore, finding product weakness from online reviews becomes a meaningful work. In this paper, we introduce such an expert system, Weakness Finder, which can help manufacturers find their product weakness from Chinese reviews by using aspects based sentiment analysis. An aspect is an attribute or component of a product, such as price, degerm, moisturizing are the aspects of the body wash products. Weakness Finder extracts the features and groups explicit features by using morpheme based method and Hownet based similarity measure, and identify and group the implicit features with collocation selection method for each aspect. Then utilize sentence based sentiment analysis method to determine the polarity of each aspect in sentences. The weakness of product could be found because the weakness is probably the most unsatisfied aspect in customers’ reviews, or the aspect which is more unsatisfied when compared with their competitor’s product reviews. Weakness Finder has been used to help a body wash manufacturer find their product weakness, and our experimental results demonstrate the good performance of the Weakness Finder.     

中文网络评论的IT产品特征挖掘及情感倾向分析

为探索中文客户评论中的IT产品特征及相关情感倾向的挖掘,帮助IT生产商和服务商提高改进产品和服务质量,提高竞争力。该文将采用情感分析技术,提出基于客户感知价值的产品特征挖掘算法,实现对于评论中IT产品特征及其情感倾向的语义分析、动态提取和综合信息挖掘;并根据用户的关注权重将产品特征和情感倾向进行排列。采用从互联网下载的真实IT产品评论语料中进行实验,初步验证了该方法的有效性。     

Do online reviews reflect a product’s true perceived quality? An investigation of online movie reviews across cultures

This paper investigates when the reported average of online ratings matches the perceived average assessment of the population as a whole, including the average assessments of both raters and non-raters. We apply behavioral theory to capture intentions in rating online movie reviews in two dissimilar countries – China and the United States. We argue that consumers’ rating behaviors are affected by cultural influences and that they are influenced in predictable ways. Based on data collected from IMDB.com and Douban.com, we found significant differences across raters from these two different cultures. Additionally, we examined how cultural elements influence rating behavior for a hybrid culture – Singapore. To study whether online consumer reviews are subjected to under-reporting bias, which is, consumers with extreme opinions are more likely to report their opinions than consumers with moderate reviews causing online reviews to be a biased estimator of a product’s true quality, we compare the consumer reviews posted online with those from an experimental study. Our results shows that under-reporting is more prevalent among US online network, thus online reviews are a better movie perceived quality proxy in China and Singapore than in the US.     

Does sampling influence customers in online retailing of digital music?

Online retailers have taken recourse to many smart marketing strategies to sell digital music. This paper investigates the strategic decisions of online vendors for offering different mechanisms such as sampling and online reviews of digital music to increase their online sales. In this research we seek answers to the following research questions (1) should online retailers offer sampling for experience goods such as music CDs; (2) under what circumstances is offering sampling more important than offering reviews. Our empirical study shows that online markets behave as communication markets, and consumers learn about product quality information both passively (by reading online reviews) and actively but subjectively (by listening to music sampling). Using data from , we empirically show that sampling is a strong product quality signal that reduces product uncertainty and attracts interested shoppers. Products with the sampling option enjoy a higher conversion rate (which leads to better sales) than those without it. Second, the impact of online reviews on conversion rate is lower for experience goods with a sampling option than those without. Third, when the uncertainty of the online reviews is higher, sampling plays a more important role because it mitigates the uncertainty introduced by online reviews. We believe this paper makes an important contribution by comparing and studying the interactions between two commonly adopted online marketing strategies (i.e., sampling versus online reviews) and provides important insights on which strategy is beneficial for vendors in the context of online selling of digital music.     

Online review manipulation by asymmetrical firms: Is a firm’s manipulation of online reviews always detrimental to its competitor?

Online reviews have a significant influence on consumers, and consequently firms are motivated to manipulate online reviews to promote their own products. This paper develops an analytical model to systematically explore the impact of online review manipulation on asymmetrical firms who sell substitutable search products in a competing market. Results show that a firm’s manipulation of online reviews is not necessary to hurt its competitor’s profit. In addition, if firms are free to choose whether to manipulate online reviews, both firms will always choose to manipulate online reviews. Moreover, there exists a prisoner’s dilemma in which online reviews are overmanipulated.     

“When Is Short,Sweet?” Selection Uncertainty and Online Review Presentations

Online customer reviews are an important part of e-commerce product selection. When used effectively, online reviews may reduce the uncertainty inherent in making product selection decisions online, but how best to deal with thousands of online customer reviews? Past research considers online review summarization, where reviews are reduced to numeric ratings, key phrases, keywords or product characteristics. However, in their original form, online reviews contain the carefully crafted narratives of past customers, elements of which may not be amenable to summarization. In this research, we present findings of a laboratory experiment which examines the impact of review summarization when evaluating different types of products online. Key findings include evidence that perceptions of product selection uncertainty depend on online review presentation format and the category of the product under consideration. Additionally, the study provides evidence that the e-commerce retailers may benefit from varying online review presentations across specific types of products.     

中文网络评论中提取产品特征的研究

大量的网络评论已经成为挖掘用户意见、改进产品质量的重要信息来源,而特征抽取作为后续分析的基础,直接影响到最终意见挖掘结果的准确性. 本文提出了一种PMI-Bootstrapping算法,并结合了语言规则实现中文网络评论的产品特征抽取. 首先利用语言规则产生候选特征集,计算每个候选特征与初始给定种子集的加权平均互信息,将满足阈值的候选特征添加到种子集中,如此循环迭代,直到种子集合收敛,输出排队后的种子集合作为抽取结果. 实验证明,该算法取得良好的准确率和召回率.     

基于书写行为与语义特征的评论有用性评估

针对电子商务网站充斥着大量有用性较低的评论,提出一种基于用户书写行为与语义特征的中文评论有用性评估方法。方法通过在Web客户端实时监听评论文本框值的变化,识别出句尾插入、非句尾插入、句尾删除、非句尾删除等书写行为,在服务器端根据书写行为、评论的语义特征建立的线性评估模型计算用户评论的有用性。实验结果表明该方法能够较为准确地识别有用性较高的评论。     

A trustworthy online recommendation system based on social connections in a privacy-preserving manner

Certain consumer websites provide reviews from previous buyers to help new customers make purchasing decisions. However, fake reviews can have an adverse impact on user trust. Most previous suggestions for addressing this problem are still subject to various security concerns in terms of privacy, reliability, and authenticity. To ensure the security of online review systems, this paper proposes the development of a secure online-evaluation method based on social connections to establish evaluation authenticity and provide protection against evaluation forgery while preserving the reviewer’s identity. The proposed method enables users to recognize evaluations from their friends to identify reviews from more trustworthy sources, and authenticates online reviews to prevent possible forgery. In addition, it preserves the privacy of friendship relationships from application server and other users and identifier relations between the personal identifier and online identifier. The proposed approach can be applied to Internet auctions and online games, and is shown to be secure and efficient, with sufficient matching probability to be practical.     

基于商品特征关联度的购物客户评论可信排序方法

电子商务网站中,海量无序的用户评论可能导致消费者客户“迷失”其中,无法识别评论的可信和真假。针对这个问题,提出了一种根据用户评论的可信度对其重新排序的方法。首先,针对网站商品广告信息,关注在线用户评论内容是否和商品功能属性密切相关,设计了基于HTML脚本格式的购物网站中商品关键特征提取算法,给出了基于自然语言处理的用户评论特征词提取方法;然后,利用词语相似度来分析商品特征和用户评论内容之间的关联度,提出了购物客户评论的可信度计算方法;最后,通过实例分析,实现了大量购物客户评论的可信排序,使得用户无须浏览全部或者大部分之后就能判断哪些评价可以信任或者具有实际的参考价值,降低了信息搜索成本,提高了决策效率。     

Scores vs. stars: A regression discontinuity study of online consumer reviews

With the rising popularity of consumer reviews, the design of the review system becomes increasingly crucial for e-commerce platforms and online retailers in their business decision-makings. Though the relationship between consumer reviews and sales has been extensively studied, only few studies have been conducted on the effects of different review designs. In this paper, we collect detailed review data from Meituan.com, a popular Chinese shopping website, to examine the effects of numerical presentation of consumer reviews (detailed to one decimal place) and graphical presentation of consumer reviews (in half-stars) on sales. By using a regression discontinuity design, we find that while consumer review scores may affect sales positively, the star presentation can create negative, rather than positive, jumps at cutoffs. Consumers restrict their attention to a star category; therefore, the “best” sellers in a lower star category are better off than the “worst” sellers in a higher star category. The incentive for review manipulation is strongly reduced, which in the long run will create trust and confidence for the review system as well as the sellers. For those sellers that are just below the cutoffs, simply crossing over the cutoffs would not lead to higher sales. Instead, they will have to substantially improve their service quality to attract consumers.     

Predicting consumer sentiments using online sequential extreme learning machine and intuitionistic fuzzy sets

Predicting consumer sentiments revealed in online reviews is crucial to suppliers and potential consumers. We combine online sequential extreme learning machines (OS-ELMs) and intuitionistic fuzzy sets to predict consumer sentiments and propose a generalized ensemble learning scheme. The outputs of OS-ELMs are equivalently transformed into an intuitionistic fuzzy matrix. Then, predictions are made by fusing the degree of membership and non-membership concurrently. Moreover, we implement ELM, OS-ELM, and the proposed fusion scheme for Chinese reviews sentiment prediction. The experimental results have clearly shown the effectiveness of the proposed scheme and the strategy of weighting and order inducing.     

基于在线评论的情感分析方法及应用

互联网以及电子商务的快速发展,使得网络成为人们交流和沟通的公共平台.消费者在网络平台生成的大量在线评论信息产生广泛影响,并引起专家学者的积极关注,基于在线评论进行的情感分析相关研究也不断发展.鉴于此,重点关注基于在线评论的情感分析方法及其应用,在对上述内容概述的基础上分析和思考现有研究存在的问题,并指出未来可能的研究方向和内容.     

Toward an Understanding of Consumer Attitudes on Online Review Usage

ABSTRACT

With the rapid e-commerce growth and changes in consumers’ behaviors, many businesses are forced to adapt their business model to match their target customers’ needs. To provide consumers with more product details and increase their confidence in making online purchases, online businesses offer an online review as an alternative to physically interacting with a product. Although consumers have become familiar with the use of online product reviews, many aspects of user behavior toward the usage of online reviews are still not well understood. This study explores the factors underlying the acceptance of consumers’ online review usage when considering purchasing an item. The study results provide insight into the factors that affect customers’ use of online reviews prior to a purchase. This study furthers the body of knowledge that deals with online reviews and system usage, providing results that allow e-commerce businesses to adapt their business model to better fit consumers’ expectations.     

基于语义约束LDA的商品特征和情感词提取

随着网络购物的发展,Web上产生了大量的商品评论文本数据,其中蕴含着丰富的评价知识。如何从这些海量评论文本中有效提取商品特征和情感词,进而获取特征级别的情感倾向,是进行商品评论细粒度情感分析的关键。本文根据中文商品评论文本的特点,从句法分析、词义理解和语境相关等多角度获取词语间的语义关系,然后将其作为约束知识嵌入到主题模型,提出语义关系约束的主题模型SRC-LDA（semantic relation constrained LDA）,用来实现语义指导下LDA的细粒度主题词提取。由于SRC-LDA改善了标准LDA对于主题词的语义理解和识别能力,从而提高了相同主题下主题词分配的关联度和不同主题下主题词分配的区分度,可以更多地发现细粒度特征词、情感词及其之间的语义关联性。通过实验表明,SRC-LDA对于细粒度特征和情感词的发现和提取具有较好的效果。