Product Reviews Research Articles

Creativity – help or hindrance? The impact of product review creativity on perceived helpfulness

Although the role of creativity in influencing consumer responses is well explored in research related to marketer-generated content, it has yet to be examined for consumer-generated online reviews. Across two studies (a text analysis of 33,508 online reviews and an experiment), this research examines the influence of creativity in online review narratives on readers' perceptions of review helpfulness. The results from these studies show that: 1. more creative reviews are evaluated as more helpful than less creative reviews, 2. perceived cognitive effort of the review originator mediates this relationship, but perceived sincerity of the review does not, and 3. product type moderates the relationship, such that greater creativity positively impacts perceived helpfulness of reviews for search products but has a negative effect for experience products.

Is the expression of different discrete emotions related to time? Evidence from online Chinese reviews using sentiment analysis and human behavior dynamics.

The online behavior of online users has taken on complex and diverse characteristics, and posting product reviews on e-commerce platforms is no exception. In fact, reviews contain rich and multi-dimensional discrete emotional information, and whether there is a relationship between the expression of these different discrete emotions and the time interval between product purchase and review posting as well as their related characteristics are the issues that this study needs to analyze and solve in depth. Based on the OCC model (named after three proposers) of psychological emotional cognitive evaluation theory as the basis for emotion classification, the study used the massive amounts of Chinese reviews of mobile phones on the Chinese e-commerce platform Jingdong Mall as the research object, applied supervised machine learning methods to classify discrete emotions, and constructed a large corpus containing satisfaction, disappointment, admiration, reproach, love, and hate; then the study delved into the distribution and behavioral dynamics characteristics of consumers' comments containing the different discrete emotions at different "purchase-comment" time intervals. The results showed that the first peak of the distribution curves of the six discrete emotions at different "purchase-comment" time intervals occurs on the first day after purchase and then decreases gradually but at different rates. The three curves for satisfaction, love, and hate emotions also show a second peak on the eleventh day, which is more similar to the bimodal distribution, implying that the corresponding product reviews are more objective. In addition, the distribution of reviews containing the six discrete emotions at different "purchase-comment" time intervals follows a power-law distribution and has the temporal characteristics of human behavioral dynamics, that is, "strong paroxysms and weak memory". However, the reviews containing the admiration and reproach emotions were most intensively written by consumers after the purchase, indicating that the service provided by the seller, logistics, and e-commerce platform stimulates more consumers to give quick responses and detailed reviews. This study is not only of great significance for exploring the internal mechanisms of consumer discrete emotional expression but also provides important decision-making references for potential consumer purchasing decisions, product updates for developers, marketing strategy formulation for marketing teams, and service improvement for sellers, logistics companies, and e-commerce platforms.

A review of hydrogen production and storage materials for efficient integrated hydrogen energy systems

AbstractThe rapidly growing global need for environmentally friendly energy solutions has inspired extensive research and development efforts aimed at harnessing the potential of hydrogen energy. Hydrogen, with its diverse applications and relatively straightforward acquisition, is viewed as a promising energy carrier capable of tackling pressing issues, such as carbon emissions reduction and energy storage. This study conducts a preliminary investigation into effective hydrogen generation and storage systems, encompassing methods like water electrolysis, biomass reforming, and solar‐driven processes. Specifically, the study focuses on assessing the potential of nanostructured catalysts and innovative materials to enhance the productivity and versatility of hydrogen energy systems. Additionally, the utilization of novel materials not only improves hydrogen storage capacity and safety but also opens up possibilities for inventive applications, including on‐demand release and efficient transportation. Furthermore, critical factors such as catalyst design, material engineering, system integration, and technoeconomic viability are examined to identify challenges and chart paths for future advancements. The research emphasizes the importance of fostering interdisciplinary collaborations to advance hydrogen energy technologies and contribute to a sustainable energy future.

Expressions of pandemic fatigue on digital platforms: a thematic analysis of sentiment and narratives for infodemic insights

BackgroundThe infodemic accompanying the COVID-19 pandemic has led to an overwhelming amount of information, including questions, concerns and misinformation. Pandemic fatigue has been identified as a concern from early in the pandemic. With new and ongoing health emergencies in 2022, it is important to understand how pandemic fatigue is being discussed and expressed by users on digital channels. This study aims to explore and report on key narrative themes associated with expressions of pandemic fatigue by users on digital platforms.MethodsThis paper describes the collection of publicly available data over a 3-month period from multiple online sources using the Meltwater and CrowdTangle platforms to source data from Twitter, Facebook, Instagram, YouTube, TikTok, Pinterest, Product Reviews, Twitch, blogs & forums. A comprehensive search strategy was developed and tested. A total of 1,484,042 social media posts were identified during the time-period that included the defined search terms for pandemic fatigue. These data were initially sorted by highest levels of engagement and from this dataset, analysts reviewed the identified posts to isolate and remove irrelevant content and identify dominant narratives. A thematic analysis was carried out on these narratives to identify themes related to expression of pandemic fatigue. Two researchers reviewed the data and themes.ResultsThe thematic analysis of narratives identified six main themes relating to expression of pandemic fatigue, and one theme of counter narratives against pandemic fatigue. Data volume increased concurrent with the time of the mpox emergency announcement. Emergent themes showed the different ways users expressed pandemic fatigue and how it was interlaced with issues of trust, preventative measure acceptance and uptake, misinformation, and being overwhelmed with multiple or sustained emergencies.ConclusionsThis paper has identified the different ways users express pandemic fatigue on digital channels over a 3-month period. Better understanding the implications of the information environment on user’s perceptions, questions, and concerns regarding pandemic and more broadly emergency fatigue is vital in identifying relevant interventions and, in the longer term, strengthening the global architecture for health emergency preparedness, prevention, readiness and resilience, as evidenced in this paper. There are clear pathways for further research, including incorporating additional languages and reviewing these themes over longer time periods.

Text Attribute Control via Closed-Loop Disentanglement

Abstract Changing an attribute of a text without changing the content usually requires first disentangling the text into irrelevant attributes and content representations. After that, in the inference phase, the representation of one attribute is tuned to a different value, expecting that the corresponding attribute of the text can also be changed accordingly. The usual way of disentanglement is to add some constraints on the latent space of an encoder-decoder architecture, including adversarial-based constraints and mutual-information-based constraints. However, previous semi-supervised processes of attribute change are usually not enough to guarantee the success of attribute change and content preservation. In this paper, we propose a novel approach to achieve a robust control of attributes while enhancing content preservation. In this approach, we use a semi-supervised contrastive learning method to encourage the disentanglement of attributes in latent spaces. Differently from previous works, we re-disentangle the reconstructed sentence and compare the re-disentangled latent space with the original latent space, which makes a closed-loop disentanglement process. This also helps content preservation. In addition, the contrastive learning method is also able to replace the role of minimizing mutual information and adversarial training in the disentanglement process, which alleviates the computation cost. We conducted experiments on three text datasets, including the Yelp Service review dataset, the Amazon Product review dataset, and the GoEmotions dataset. The experimental results show the effectiveness of our model.

Data colonialism on Facebook for personalised advertising: The discrepancy of privacy concerns and the privacy paradox

In the digital era, Facebook serves as one of the leading sources of personal data and monetises that information for advertising purposes. This study aims to investigate users' perceptions regarding personal data collection and privacy concerns on Facebook and to explore the mediation between privacy concerns and the privacy paradox in relation to personalised advertising outcomes. To address this gap, data was collected through a mixed-method approach by conducting an online survey with 155 respondents, followed by five in-depth semi-structured interviews. The study indicated that most respondents were concerned about their personal information being aggregated and monetised by Facebook without users' consent. Most survey respondents never/rarely clicked on Facebook advertising formats such as photo, video, stories, messenger, carousel, slideshow, collection and playable advertising. However, survey and interview data divulged that they sometimes clicked on and engaged in influencer advertising because they regarded this content as informative, attractive and reliable for product reviews. The study discovered that respondents felt ambivalent towards personalised advertising and sometimes traded their privacy to get immediate benefits from the advertising. Their privacy decision-making process was affected by a cost–benefit calculation, which resulted in what is known as the ‘privacy paradox’, where an individual intentionally divulges personal information on social media despite stating privacy concerns.

Geographical Tweets Based Current Affairs Prediction Using Hybrid Features

Twitter serves as a prominent and freely accessible social networking platform, enabling registered and authorized users to share their opinions and reviews through concise messages known as tweets. This research aims to conduct sentiment analysis on a Twitter dataset, seeking to predict and analyze current affairs based on user behaviour and opinions. The scope of current affairs encompasses a wide range of topics, including product reviews, political discourse, movie ratings, and more, expressed in both real-time streaming and offline data on the Twitter platform. During the feature extraction phase, a hybrid approach is employed, incorporating features such as acronyms, synonyms, and emoticons utilized by users in their tweets. Additionally, a hybrid dictionary is introduced, and instead of relying solely on unigrams and bigrams, a novel algorithm, n-gram, is implemented for more comprehensive analysis. This paper delves into various methodologies, specifically focusing on multiclass classification, and introduces a proposed system along with its noteworthy results. The research contributes to the understanding of user sentiments and behaviours on Twitter, offering valuable insights into diverse current affairs.

An Efficient Aspect-based Sentiment Classification with Hybrid Word Embeddings and CNN Framework

Background: As the e-commerce product reviews and social media posts are increasing enormously, the size of the database for polarity/ sentiment detection is a challenging task, and again, predicting polarities associated with respect to aspect terms end to end in a sentence is a havoc in real-time applications. Human behavior is influenced by the various opinions generated in society. Public opinion influences our decisions most often. Businesses and establishments always need to collect the opinion of the society, which they try to obtain using customer feedback forms and questionnaires or surveys, which help them to be aware of the shortcomings if any, and to use suggestions to improve quality. It works in the same way for customers as well and the opinions of other customers about a particular product can come in handy when deciding to buy a product. Objectives: In this work, an efficient Aspect-based Sentiment Classification technique has been introduced with a hybrid, multiple-word embedding methods and implemented using the CNN framework on large databases. Methods: Most of the traditional models have a limitation on the dependency for one or more similar types of aspect words for sentiment classification problem. However, these conventional models such as TF-ID, Word 2Vec and Glove method consumes much more time for word embedding process and Aspect terms generation and further process of aspect level sentiment classification. Further, these models are facing problems of high true negative rate and misclassification rate on large aspect databases in sentiment classification. In this article, we have introduced an efficient Proposed ensemble word embedding model in the CNN network and defined Hybrid Word2 Vec method, Hybrid Glove word embedding method and Hybrid Random Forest model for sentiment classification. Results: Experiments on a widely used benchmark prove that the proposed word embedding method- based classification technique results in to higher true positive rate with minimal misclassifications and also supports better runtime and accuracy than the traditional word embedding-based aspect level classification approaches. Conclusion: In this article, a hybrid ensemble feature ranking-based classification model is proposed on the large aspect databases. In this work, advanced multiple-word embedding methods are implemented to improve the essential feature extraction problem in the aspect level sentiment process. These multiple-word embedding methods are applied to the sentiment databases in the CNN framework.

Hamiltonian deep neural network fostered sentiment analysis approach on product reviews

Hamiltonian deep neural network fostered sentiment analysis approach on product reviews

Classifying Opinions and Sentiments on Social Networking Sites using Machine Learning Classifiers

Abstract: People now publish evaluations on social media for any product, movie, or location they visit as a result of the Web's rapid development. Customers and product owners can both benefit from the reviews posted on social media in order to assess their offerings. Compared to unstructured data, structured data is simpler to analyze. The reviews are mostly available in an unstructured format. Aspect-Based Sentiment Analysis extracts from the reviews the features of a product and then calculates sentiment for each feature. Sentiment analysis, also known as opinion mining, is a natural language processing (NLP) technique used to determine the sentiment or emotional tone expressed in a piece of text, such as a sentence, paragraph, or document. Machine leaning classifiers are used to classify sentiments. Machine learning classifiers cannot process raw text so raw text needs to be converted into vectorized form. Feature extraction techniques are used to convert raw text to numerical form also called vectorized data. In present research, four feature extraction techniques with five different machine learning classifiers namely, SVM, Logistic Regression, Naïve Bayes, Random Forest, and KNN are applied to classify sentiments associated with tweets. Two online twitter data sets containing tweets about product reviews and tweets about people's thoughts on public policy are selected for experimentation. In the experiments done, it has been found that the SVM classifier using TFIDF and HFE shows better performance as compared to other classifiers. Using the feature sets, 97% accuracy and 98% F1- score is achieved in the aspect category prediction task.

How Do Consumers Evaluate Identical Products at Different Retailers? A Text Mining Approach Using Product Reviews

We examine consumers’ evaluations of identical products at different types of retailers using Amazon’s launch of its fashion division. Drawing on product reviews of brand items carried by Amazon (the general merchandise online retailer) and Macy’s (the traditional fashion-oriented department store), we compare the topics in the reviews as important product attributes of consumer evaluations and examine the effects of these topics on ratings. We show that the product topics mentioned for identical products are generally different between Amazon and Macy’s in terms of their proportions and tone. Also, the different topics discussed at the retailers lead to different ratings, with lower evaluations for Amazon. By demonstrating the significant characteristics of the retailers, we provide managers with insights on developing and maintaining competitive advantages for the general retailer and specialty retailers.

A Comparative Study on Machine Learning Approaches for Sentiment Analysis

Sentiment analysis plays a pivotal role in the operations of online product companies. User reviews are taken into account by others when they search for products, forming the cornerstone for delivering the right product based on user sentiments through sentiment analysis. Sentiment analysis involves the process of collecting, analyzing, and recommending reviews, which are often extensive and contain multiple paragraphs of content. This paper presents a comparative analysis of various machine learning models used to conduct sentiment analysis on customer reviews of Amazon products within the Electronics category. The initial models under scrutiny for our analysis include Logistic Regression, Decision Tree, Naive Bayes Classifier, Random Forest, Support Vector Machines, and BERT Model. The experimental result show that BERT classifier achieves higher accuracy when compare with other machine learning models.

Parasocial Interaction on Indonesia’s Beauty Influencer: Antecedents and Consequences

Consumers' dependence on product reviews provided by other consumers makes consumers make social media influencers (SMI) a source of information and reference before purchasing products. The influence of SMI on their followers has led many beauty marketers to use social media influencers to communicate their products. However, the increasing use of influencer marketing by marketers has led to consumer suspicion of the authenticity of influencers. Therefore, based on the stimulus-organism-response (SOR) framework, this study analyzes consumer behavior through the content attributes and interaction strategies exposed by social media influencers to followers, leading to consumers' internal perceptions of parasocial interaction, hedonic value, and utilitarian, and ultimately influencing consumers' response of purchase intention. This study used PLS-SEM as a data analysis technique using a sample of 363 respondents. The PLS-SEM results revealed that content attributes (i.e., attractiveness and prestige) and interaction strategies (i.e., interactivity) positively influence parasocial interaction and, in turn, lead to purchase intention. The results also show that parasocial interaction has a significant effect on hedonic (perceived transparency and perceived enjoyment) and utilitarian (informativeness and credibility) responses. It was also found that hedonic (perceived enjoyment) and utilitarian (credibility) responses significantly influenced purchase intention. Meanwhile, the hedonic response (perceived transparency) and utilitarian response (informativeness) did not significantly affect followers' purchase intention. This research demonstrates the role of content attributes and interaction strategies in creating parasocial interactions and the effects of parasocial interactions on hedonic, utilitarian and purchase intention responses. Based on the results of this study, marketers and influencers are advised to build intimate relationships with the audience to increase purchase intention. This relationship-building can be created through high interactive power, attractiveness, and prestige content. Then, it is also recommended to pay attention to aspects of audience comfort and maintain the credibility of influencers to shape the buying interest of the audience of social media influencer content.

A Review of the Sediment Production and Transport Processes of Forest Road Erosion

Forest roads are a common land use feature with a significant impact on sediment yield and the water sediment transport processes within a watershed, seriously disrupting the safety and stability of the watershed. Previous studies have focused on the sediment production processes within the road prism. However, there has been limited attention given to the transport processes of road-eroded sediment at various scales, which is crucial for understanding the off-site effects of road erosion. This paper reviews research conducted on forest road erosion over the past two decades. It summarizes the mechanisms of sediment production from road erosion and provides a detailed analysis of the transport mechanisms of eroded sediments from roads to streams at the watershed scale. The paper also examines the ecological and hydrological effects, research methods, and control measures related to sediment transport caused by forest road erosion. It identifies current research limitations and outlines future research directions. The findings of this review highlight several key points: (1) Most research on forest road erosion tends to be specific and unilateral, often neglecting the broader interaction between roads and the watershed in terms of water–sediment dynamics. (2) Various research methods are employed in the study of forest road erosion, including field monitoring, artificial simulation experiments, and road erosion prediction models. Each method has its advantages and disadvantages, but the integration of emerging technologies like laser scanning and fingerprint recognition remains underutilized, hindering the simultaneous achievement of convenience and accuracy. (3) The transport processes of forest road-eroded sediment, particularly on road–stream slopes, are influenced by numerous factors, including terrain, soil, and vegetation. These processes exhibit significant spatial and temporal variability, and the precise quantification of sediment transport efficiency to the stream remains challenging due to a lack of long-term and stable investigation and monitoring methods. The establishment and operation of runoff plots and sedimentation basins may help offer a solution to this challenge. (4) Both biological and engineering measures have proven effective in reducing and limiting sediment erosion and transport. However, the costs and economic benefits associated with these regulation measures require further investigation. This review provides a comprehensive summary of relevant research on sediment erosion and transport processes on unpaved forest roads, enhancing our understanding of sediment yield in watersheds and offering valuable insights for reducing sediment production and transport to streams.

Sentiment Analysis on a Large Indonesian Product Review Dataset

Background: The publicly available large dataset plays an important role in the development of the natural language processing/computational linguistic research field. However, up to now, there are only a few large Indonesian language datasets accessible for research purposes, including sentiment analysis datasets, where sentiment analysis is considered the most popular task. Objective: The objective of this work is to present sentiment analysis on a large Indonesian product review dataset, employing various features and methods. Two tasks have been implemented: classifying reviews into three classes (positive, negative, neutral), and predicting ratings. Methods: Sentiment analysis was conducted on the FDReview dataset, comprising over 700,000 reviews. The analysis treated sentiment as a classification problem, employing the following methods: Multinomial Naïve Bayes (MNB), Support Vector Machine (SVM), LSTM, and BiLSTM. Result: The experimental results indicate that in the comparison of performance using conventional methods, MNB outperformed SVM in rating prediction, whereas SVM exhibited better performance in the review classification task. Additionally, the results demonstrate that the BiLSTM method outperformed all other methods in both tasks. Furthermore, this study includes experiments conducted on balanced and unbalanced small-sized sample datasets. Conclusion: Analysis of the experimental results revealed that the deep learning-based method performed better only in the large dataset setting. Results from the small balanced dataset indicate that conventional machine learning methods exhibit competitive performance compared to deep learning approaches. Keywords: Indonesian review dataset, Large dataset, Rating prediction, Sentiment analysis

Exploratory Analysis of Different Types of Adjectives for Sentiment Classification

Online websites available on the internet, Amazon, provide a platform for users to share their valuable opinions. Users’ reviews are available in thousands, and extracting useful information from these reviews manually is a crucial task for managers, companies, and users. Reviews are written in natural language, and to extract useful information from the reviews, there is a need for an automatic technique known as sentiment analysis. Researchers have used polarity features like nouns, adjectives, verbs, and adverbs to mine sentiment using one feature or combination. The studies showed that adjectives remain the most prominent feature. However, previous research has not evaluated many types on a comprehensive dataset. This research focuses on the identification of different kinds of adjectives in a given text, the identification of the best type and the classification of different kinds as positive, negative, and neutral. Understanding the nuanced impact of various adjectives in sentiment classification is crucial for developing more accurate and context-aware natural language processing models, essential for applications ranging from sentiment analysis to customer reviews. We comprehensively evaluate different types of adjectives on machine learning algorithms. The experimental results are performed on an annotated dataset of 58,258 office product reviews collected from Amazon. Out of these evaluated adjectives, opinion adjectives have the highest precision of 0.931 on the Naïve Bayes classifier and show the best sentiments.

Website Design Using Laravel Framework for Tart Tuns Shop Sales Pontianak

Websites have become a means of getting information, promotion, sales that are widely used by Indonesians. Tart Tuns has difficulties in managing orders and products because currently sales in this shop are only through Instagram, in making transactions with buyers the admin communicates via Instagram direct messages and via WhatsApp messenger. This transaction process is less efficient because the Tart Tuns cake shop must always update information about product availability on insta story. This research aims to address the existing needs of the store in the form of a website that can be used as a sales medium to be more structured and expand product offerings not limited to the store's Instagram account by utilizing the MVC architecture of the Laravel Framework and collaborative filtering algorithms. System modeling uses UML (Unified Modeling Languange) and the research method used is Design Science Research (DSR). The results of this study are to produce a sales website software at Tart Tuns using the Laravel framework which functions to help shop owners in making it easier to manage orders, products, transactions, and product categories. Suggestions from this research are that other developers can add / develop features such as product reviews, chat on the website, and product liking features.

To Tell, Not to Yell: The Effect of Writer's Intents on Readers’ Perceived Helpfulness of Online Product Reviews

Perceived review helpfulness has been investigated extensively; however, the influence of the review writer's intents behind a review has not been considered. This study investigates the effect of the writer's intents to provide information, express emotions, or call for action on reviews’ helpfulness, thereby bridging the gap between the writer's perspective and the reader's perspective. The writer's intents are operationalized through speech acts, an implicit, yet systematic, approach in which the intents of the writer are captured at the linguistic level, that is, how things are said. A unique parallel approach based on both readers’ perceptions and linguistic theory is used to operationalize the speech acts. This approach allows multiple speech acts per sentence, and it can estimate the influence of each speech act separately. The results show that assertive acts positively impact review helpfulness, whereas expressive and call-for-action acts have a negative effect. This confirms previous literature showing that the major drivers of review helpfulness are often related to information delivery. In this research, the idea that consumers like reviews with accurate information is detected at the level of the writer's intent, which is new in the context of text analysis to assess online review helpfulness.

Exploring Influencing Marketing—Consumer Insights and Creators’ Perspectives

Internet users consistently lean towards influencers who actively participate in their respective areas of interest, providing thorough information and genuine product reviews. These practices have the potential to positively influence the decision-making process related to purchasing. This study delves into the realm of influencer marketing, examining it from the viewpoints of both consumers and specialists. The study is based on the qualitative research, employing both individual in-depth interviews and a focus group. More precisely, the investigation included a series of six individual interviews with influencers from Romania and a focus group comprising eight participants to grasp their perceptions of influencer marketing campaigns. The findings underscore the importance of transparency and authenticity for content creators. Influencers tailor their content based on the platform they use, revealing an existing correlation among them, all while underscoring the significance of transparency and striving to maintain close connections with their followers. This study not only offers a fresh perspective on understanding consumers’ views regarding the activities of content creators in the online domain but also sheds light on influencers’ perspectives concerning collaborations, content creation, and their relationships within their community. Theoretical implications encompass the identification of novel behavioral traits among Romanian consumers regarding influencer marketing actions and contributes to a clearer understanding of the term influencer marketing.

An Overview of the Application of Convolutional Neural Networks inSentiment Analysis

The field of natural language processing, or NLP, uses its understanding of human language to find practical solutions to issues. It mainly includes two parts: the core task and the application. The core task represents the common problem that needs to be solved in various natural language application directions. It includes language models, morphology, grammar analysis, semantic analysis, etc. At the same time, the application section focuses on specific natural language processing tasks such as machine translation, information retrieval, question-answering systems, dialogue systems, etc. Natural language processing has made a significant contribution to the development of human society and the economy and provides strong support for all aspects of research work. Opinion mining, or sentiment analysis, is a subfield of natural language processing that develops systems for identifying and extracting ideas from text. Sentiment analysis is a hot topic since it has many practical applications. Many opinion-expressing texts are available on review sites, forums, blogs, and social media as the amount of publicly available information on the Internet grows. This unstructured information can then be automatically transformed into structured data about products, services, brands, politics, or other topics on which people can express their opinions using sentiment analysis systems. This information can be used for marketing analytics, public relations, product reviews, network sponsor ratings, product feedback, and customer service. With the rapid growth of labeled sample data sets and the notable enhancement in graphics processor (GPU) performance, convolutional neural network research has advanced rapidly and achieved remarkable leads to various computer vision tasks. By reviewing the application of CNN, we see that convolutional operations are naturally suitable for some text processing and, thus, naturally suitable for the background of sentiment analysis.