Figure - available from: Cognitive Computation
This content is subject to copyright. Terms and conditions apply.
Source publication
With the development of e-commerce, many products are now being sold worldwide, and manufacturers are eager to obtain a better understanding of customer behavior in various regions. To achieve this goal, most previous efforts have focused mainly on questionnaires, which are time-consuming and costly. The tremendous volume of product reviews on e-co...
Citations
... The feedback that customers give voluntarily on these platforms is a key factor in the product development and design phase . In addition, data taken from social media is used for various purposes, such as predicting the company's popularity level based on consumer reactions on social media (Park & Alenezi, 2018 Engineering Management in Production and Services Sebastian (2018) looked at the performance of a product in the market based on ratings and reviews provided by users, and Zhou et al. (2016) compared consumer behaviour when shopping online. ...
The research aims to provide the decision-maker with a framework for determining customer requirements during product development. The proposed framework is based on sentiment analysis and supervised multilabel classification techniques. Therefore, the proposed technique can categorise customer reviews based on the “product design criteria” label and the “sentiment of the review” label. To achieve the research goal, the research presented in this article uses the existing product development framework presented in the literature. The modification is conducted especially in the conceptual stage of product development, in which the voice of the customer or a customer review is obtained from the scraping, and a multilabel classification technique is performed to categorise customer reviews. The proposed framework is tested by using the set data on women’s clothing reviews from an e-commerce site downloaded from www.kaggle.com based on data by Agarap (2018). The result shows that the proposed framework can categorise customer reviews. The research presented in this paper has contributed by proposing a technique based on sentiment analysis and multilabel classification that can be used to categorise customers during product development. The research presented in this paper answers one of the concerns in the categorisation of needs raised by Shabestari et al. (2019), namely, the unclear rules or main attributes of a requirement that make these needs fall into certain categories. Categorising customer requirements allows decision-makers to determine the direction of product development to meet customer needs.
... Understanding the preferences of table grape consumers and studying the factors that affect their preferences provide breeding workers with reference breeding directions (Si-Hong et al., 2020), and enable operators of the table grape industry to accurately predict consumer requirements and formulate appropriate marketing strategies (Qian et al., 2018;Zhou et al., 2015). In order to determine consumer preferences, some studies were conducted on their preferences of table grapes in China; however, preference predictions were not made (Mu et al., 2019;Yao et al., 2020;Zhou et al., 2016). The table grape market in China is facing a situation where grape varieties are highly similar, coupled with the impact of alternative products, resulting in fewer choices for consumers. ...
This paper aims to understand intrinsic attribute preferences of Chinese consumers for table grapes, analyze the influencing factors, and build consumer preference prediction models. In this study, 4324 consumers from various regions of China were investigated. We analyzed consumer preferences and the influencing factors. Finally, binary logistic regression was used to construct prediction models of consumers’ intrinsic attribute preferences for table grapes. The results showed that grapes popular with Chinese consumers had fixed characteristics, including moderate size, spherical or near-spherical shape, purple-red color, strawberry flavor, light aroma, soft flesh and juicy, sweet taste, seedlessness, thin skin powder, and easy to peel. The results of the prediction models showed that age, annual consumption of grapes, and other factors of consumers had significant effects on consumer preferences. The prediction models achieved 80% accuracy in predicting consumer preferences for taste, seedless and peeling degrees. Analyzing the latest attribute preferences of Chinese table grapes, consumers can provide on the one hand, breeding direction reference for breeders, and on the other, marketing suggestions for marketers in the table grape industry. This study comprehensively investigated the intrinsic attribute preferences of Chinese table grape consumers, mastered the latest results of consumer preferences, added the indicators of the intrinsic attributes from the perspective of consumer demand, and conducted relatively more complete prediction research on the preferences of Chinese consumers.
... The comparison at the aspect level identified differences in opinion regarding the aspects battery and keyboard for a given cell phone brand -English customers had a negative sentiment, while Chinese customers had a positive opinion. Zhou et al. [32] analyzed product reviews extracted from Amazon on cameras, smartphones, and tablets. Their method consisted in generating questions and automatically finding the answers to determine the sentiment towards brands and product features. ...
... In line with these works, we provide case study (Section VI) comparing customer opinions from three different countries and languages regarding product features, e.g., the zoom of digital cameras. The main differences in relation to existing work [10], [32], [28] are that we focus a different group of languages and rely on a multilingual product catalog enriched with user opinions to enable such comparisons. ...
... Due to its vast range of applications, it has become one of the most active research domains in natural language processing and text mining in recent years [2]. Opinions influence our behavior and are crucial to almost all human activity [3]. Whenever we need to make a decision, we want to hear other people's perspectives. ...
The computational examination of people's opinions, attitudes, and emotions conveyed in written language is known as sentiment analysis or opinion mining. In recent years, it has become one of the most active study fields in natural language processing and text mining. Sentiment analysis of social media texts, which are predominantly code-mixed for Dravidian languages, is becoming more popular. In a multilingual community, code-mixing is common, and code-mixed writings are produced using native and non-native scripts. The current paper uses machine learning, deep learning, and parallel hybrid deep earning models to identify sentiments in Dravidian code-mixed social media text. The experiments were conducted using a dataset from the Dravidian-CodeMix-FIRE 2021 1 competition, which included YouTube comments in Tamil, Malayalam, and Kannada code-mixed languages.
... In the context of OCRs helpfulness, some research investigated different nationalities' consumer sentiments for diffusing the OCRs. Zhou et al. (2016) found that Chinese consumers concentrate on the commodity's outer features and consumer feelings, whereas American consumers keep attention on the commodity's interior characteristics and detailed information. When transmitting OCRs, Chinese consumers like to express their feelings passively, whereas American consumers tend to express their feelings in a straight way (Zhou et al., 2016). ...
... Zhou et al. (2016) found that Chinese consumers concentrate on the commodity's outer features and consumer feelings, whereas American consumers keep attention on the commodity's interior characteristics and detailed information. When transmitting OCRs, Chinese consumers like to express their feelings passively, whereas American consumers tend to express their feelings in a straight way (Zhou et al., 2016). Consumers trust the OCRs for reviews where the critic's identity and avatar images are integrated (Munzel, 2016). ...
... It is the elementary foundation of OCRs, which is considered as the eWOM and used as a recommendation for future users. Many overwhelming previous research publications had been done based on text reviews such as text mining (Cao et al., 2011;Chatterjee, 2019), consumer behavior (Comscore, 2007;Zhou et al., 2016), review helpfulness (Chua & Banerjee, 2015;Korfiatis et al., 2012), trustworthiness (Tsang & Prendergast, 2009), fake reviews (Hunt, 2015), security issues (Bao et al., 2000), purchasing decisions (Chatterjee, 2019), and so on. In this study, we cope with the review length to understand the usefulness to consumers as well as understand the consumer sentiment for OCRs when expressing different lengths of text reviews. ...
This research endeavors to fill the research cavity in the domain of online consumer reviews (OCRs) through investigating the relationship between review length and rating. Moderated multiple regression (MMR) analysis was used to investigate the moderation interactions of some critical factors, including the product price, brand, product type, and delivery systems. Through data curation, 10,547 sets of cross-sectional product data containing 200,169 sets of reviews refined from the original 12,009 products collected from jd.com in China were used. Research through econometric analysis reveals that review length improves rating. Through interactions, this study found that the price has negative interaction, whereas brand familiarity has a positive interaction with review length and rating. Domestic brands, search goods, and third-party delivery systems have more positive interactions on review length and rating than the overseas brand, experience goods, and platform’s self-delivery systems, respectively. This paper renders insight for managers, especially merchants, who should encourage consumers to write meaningful lengthy reviews to improve their reputation.
... In the context of OCRs helpfulness, some research investigated different nationalities' consumer sentiments for diffusing the OCRs. Zhou et al. (2016) found that Chinese consumers concentrate on the commodity's outer features and consumer feelings, whereas American consumers keep attention on the commodity's interior characteristics and detailed information. When transmitting OCRs, Chinese consumers like to express their feelings passively, whereas American consumers tend to express their feelings in a straight way (Zhou et al., 2016). ...
... Zhou et al. (2016) found that Chinese consumers concentrate on the commodity's outer features and consumer feelings, whereas American consumers keep attention on the commodity's interior characteristics and detailed information. When transmitting OCRs, Chinese consumers like to express their feelings passively, whereas American consumers tend to express their feelings in a straight way (Zhou et al., 2016). Consumers trust the OCRs for reviews where the critic's identity and avatar images are integrated (Munzel, 2016). ...
... It is the elementary foundation of OCRs, which is considered as the eWOM and used as a recommendation for future users. Many overwhelming previous research publications had been done based on text reviews such as text mining (Cao et al., 2011;Chatterjee, 2019), consumer behavior (Comscore, 2007;Zhou et al., 2016), review helpfulness (Chua & Banerjee, 2015;Korfiatis et al., 2012), trustworthiness (Tsang & Prendergast, 2009), fake reviews (Hunt, 2015), security issues (Bao et al., 2000), purchasing decisions (Chatterjee, 2019), and so on. In this study, we cope with the review length to understand the usefulness to consumers as well as understand the consumer sentiment for OCRs when expressing different lengths of text reviews. ...
This research endeavors to fill the research cavity in the domain of online consumer reviews (OCRs) through investigating
the relationship between review length and rating. Moderated multiple regression (MMR) analysis was used to investigate
the moderation interactions of some critical factors, including the product price, brand, product type, and delivery systems.
Through data curation, 10,547 sets of cross-sectional product data containing 200,169 sets of reviews refined from the
original 12,009 products collected from jd.com in China were used. Research through econometric analysis reveals that
review length improves rating. Through interactions, this study found that the price has negative interaction, whereas
brand familiarity has a positive interaction with review length and rating. Domestic brands, search goods, and third-party
delivery systems have more positive interactions on review length and rating than the overseas brand, experience goods, and platform’s self-delivery systems, respectively. This paper renders insight for managers, especially merchants, who should encourage consumers to write meaningful lengthy reviews to improve their reputation.
... Unit distribution analysis is used to investigate whether different levels of control variables have a significant effect on the observed variables. By inferring whether there is a significant difference in the overall average of the observed variables at each level of the control variable, analyze whether the control variable has a significant effect on the observed variable (Zhou, et al., 2016). The formula for verification statistics is as follows: ...
This article aims to study e-commerce precision models and consumer behavior models based on clustering algorithms, and at the same time conduct detailed research on the Gaussian mixture distribution algorithm, consumer behavior and model construction, and precision marketing strategies in the clustering algorithm. First, a lot of analysis and demonstration of precision marketing strategies and the construction of consumer behavior models are carried out, and then the clustering algorithm-based electronic some experiments were carried out on the application of commercial precision marketing methods and consumer behavior models. The experimental results show that the precision marketing method using the clustering algorithm is more in line with the development of modern e-commerce. The application of the algorithm in the precision marketing methods of enterprises and consumer behavior models has promoted the vigorous development of enterprises, making the sales volume of enterprises reach 9.8 %.
... Such a negative impact can be attenuated if foreign consumers of a local business can use the domestic language. Similarly, Zhou et al. (2016) examine the differences in eWOM between China and the US based on their beliefs and philosophies. Chinese people follow the Doctrine of the Mean from Confucian philosophy, which is absent from US culture. ...
... Yet, certain values unique to a country can significantly sharpen the eWOM behavior in that nation. For example, Zhou et al. (2016) suggest that Chinese and Americans write eWOM differently because of China's Doctrine of Mean. National heritage reflects the unique history of a country. ...
Purpose
This paper aims to summarize peer-reviewed journal articles on national cultures and electronic word of mouth (eWOM) behavior, identify the main findings and patterns among those studies and discuss research gaps that need to be addressed in the future.
Design/methodology/approach
A systematic review process was utilized to analyze peer-reviewed journal articles on both eWOM and national cultures. The main research questions were defined, then proceeded by the identification of exclusive and inclusive criteria to search for relevant articles, which were further filtered based on abstracts and full texts, and then scrutinized for main findings and major variables such as countries, cultural variables and data collection methods.
Findings
An analysis of 52 papers shows that national cultures, primarily Hofstede's dimensions, influence the willingness of individuals to share eWOM, how they write eWOM and the extent to which they use eWOM to make decisions. Although the reviewed studies have provided insightful implications for marketing theory and practice, the present paper has identified a number of important questions that warrant future research attention.
Originality/value
eWOM is continually being employed as a popular source of information for consumers throughout different countries to make their purchase decisions. However, eWOM behavior differs from country to country due to national cultures, and managers' eWOM strategies that work in one country may not be applicable in another. Therefore, there has been an increasing interest in this topic. Nevertheless, it remains unclear which subjects have been addressed and what areas are yet to be investigated. This paper presents a comprehensive review of how national cultures affect eWOM behavior by drawing upon prior research and provides directions for future research contributions.
... By conducting review content mining, research has confirmed the usefulness of perceived academic review [39]. Additionally, based on review information, research across China and the United States has pointed out differences in consumer behavior among different countries [40]. ...
... Furthermore, cue diagnosticity theory holds that information cues are not independent in actual judgments, but jointly influence the decision-making process [18]. For instance, Khare et al. [40] found interactions among various cues, including the need for uniqueness, WOM volume, and WOM valence interaction. Wen et al. [26] indicated that online reviews could moderate brand familiarity with booking intentions. ...
Online reviews are increasingly being used and researched by people worldwide. Compared with previous studies on traditional products or services, research focused on online health communities (OHCs) is still insufficient. Thus, based on cue diagnosticity theory, this research concentrates on combining two mainstream studies by incorporating the patient-generated review with the unique characteristics of online medical services–the doctor-patient relationship–to study the information processing issues in choosing consultations. We clawed the dataset, including 2865 doctors related to 152,864 patient-generated reviews and information, from the GoodDoctor website. We then employed a negative binomial regression to test our hypotheses. Interestingly, we found that the effects of review length and review volume on doctors’ consultations can be negatively moderated by the doctor-patient relationship. Our findings can serve patients, doctors, platform managers, and others to optimize the application of patients’ information processing when choosing consultations.
... In addition, a recent study by Phang & Goh (2019) revealed that when consumers develop a sense of attachment and a degree of consistency with a particular brand like to a person, they are likely to have positive behaviour towards the use of such products bearing such a brand. Positive sentiment brings a better attitude towards online shopping behaviour because Chinese customers' overall sentiment in relation to both overall satisfaction and product-level satisfaction was high (Zhou, Xia & Zhang, 2016). Gillitzer and Prasad (2016) found that consumer sentiment influences actual consumer behaviour and has predictive power for future movements in consumption. ...
The internet has become part and parcel of consumers' daily activities. One can simply buy anything online only with a click of a button and have products delivered right at their doorstep. This paper explores the antecedents of Chinese consumer behaviours towards online shopping. The quantitative research method was adopted for the study using data obtained from 208 Suzhou residents and analysed using Structural Equation Model (SEM) with Amos 23. The paper concludes that attitude is the sole predictor of online shopping behaviour. The results of the study further show that online shopping experience and social influence relate positively and significantly with the sentiment but price motion was not significant. Also, convenience has a positive significant relationship with attitude whiles perceived usefulness, but the ease of use was not significant. Finally, the results reveal that sentiment is not directly related to online shopping behaviour but indirectly related through attitude. Based on these findings, we provide important managerial implications and offer guidance to e-commerce platforms and online retailers.
