Search

Firms like Uber, Amazon, and TripAdvisor have popularized the rating of people, goods, and services. These entities receive scores (e.g., through online reviews) in a variety of presentation formats: incremental (a raw score per episode; e.g., 5–5–2), cumulative (updated average scores; e.g., 5–5–4), or a combination thereof. This article focuses on prevalent situations in which a score deviates from prior scores and examines how the presentation format of the scores impacts decision makers' (e.g., consumers, managers) evaluations of the entity scored. Across a wide variety of settings, nine experiments document that when a generally well-performing (poorly performing) entity suddenly receives a negative (positive) score, overall performance will be perceived as less negative (positive) when shown in a cumulative format compared with an incremental or combined format. This effect appears to be stronger when the deviating episode is more representative (e.g., due to higher recency or internal attribution). The authors also find evidence for their proposed explanation: a cumulative format distorts individuals' perceptions of the underlying raw score of the deviating episode. These findings imply that presenting scores in alternative formats may affect marketing outcomes (e.g., customer churn, product choice, technology adoption, new product success, and user engagement on peer-to-peer platforms). [ABSTRACT FROM AUTHOR]

As customers increasingly rely on online reviews for making consumption decisions, the dangers arising from misinformation and fakery have become an acute source of concern for consumers, firms, and society at large. Many online review platforms claim a role as guardians of trust in the information exchange process. Yet, little is known about the practices firms can utilize to design platforms that build and safeguard consumer trust. The authors draw on governance and identity disclosure literature streams to propose five practices that mitigate fakery and build trust in the platform: monitoring, exposure, community building, status endowment, and identity disclosure. Five studies (1) establish the individual and joint effects these practices have on platform trust, (2) identify the mediating processes by which the practices build trust, (3) verify the ecological value of the conceptual framework, (4) provide support of the salience of the practices, and (5) show how the practices build trust above and beyond review content–level characteristics. Interaction effects suggest that exposure, community building, and status endowment obviate the need for identity disclosure in building platform trust, a finding that mitigates related privacy concerns. Collectively, the findings can improve platform design, enabling review platforms to fulfill their promise of mitigating fakery and increasing trust. [ABSTRACT FROM AUTHOR]

Prior research has indicated that consumers' decisions are significantly influenced by online reviews. However, existing research has focused mainly on attributes (e.g., average ratings) that are not fully controlled by firms; only limited research has investigated how controllable attributes (e.g., review display formats) affect consumers. Drawing on visual perception research, the authors examine the effectiveness of two prominent graphical display formats used by major e-commerce platforms: one that displays rating distributions in a proportional format (e.g., Amazon) and one that does so in a simple format (e.g., Google). The results indicate that due to the changes in graphs' reference points caused by the shrunken x-axis in simple bar graphs, consumers respond more positively to an item when its rating distribution is displayed in a graphically simple (vs. proportional) format. This effect is moderated by the distribution's peak value (i.e., the share of the most frequent rating) and imbalance score (i.e., the difference between the share of positive and negative ratings). Furthermore, even an item's future ratings are influenced by the graphical format in which its prior ratings are displayed. The contributions to the marketing literature are discussed, and insights that can aid managers in making more informed decisions are provided. [ABSTRACT FROM AUTHOR]

Prior research has shown that user-generated content (UGC) on social media is an important source for consumers seeking information about mental health treatment. The purpose of this study is to explore UGC about prescription drugs for depression as a type of digital 'pharma' folklore. Using qualitative inquiry, the authors analyse 73 user-generated product reviews on YouTube. Findings yield a compelling portrait of people's struggle with mental illness and their treatment journey. Applying folklore as a framework provides a novel strategy for analysing and theorising the cultural worldview of a consumer group. It allows marketers to decipher the knowledge being passed on to others from product review videos and gain a deeper understanding of stories that resonate with consumers seeking treatment with depression drugs. [ABSTRACT FROM AUTHOR]

Regulatory authorities typically require bioequivalence to be demonstrated by comparing pharmacokinetic parameters like area under the plasma concentration-time curve (AUC) and maximum plasma concentration (Cmax). Because in certain cases, AUC and Cmax alone may not be adequate to identify formulation differences in early and/or late segments of the dosing interval, partial AUCs (pAUCs) have been proposed as additional metrics to evaluate bioequivalence. Even though cut-off points for pAUCs are usually decided based on clinical relevance, the identification of the correct cut-off range remains elusive in many other cases and tends to contribute to increased pAUC estimate variabilities. The choice of meaningful cut-off points in pAUC estimates can be especially difficult in the case of long-acting injectable (LAI) products, where long dosing intervals and complex pharmacokinetic (PK) and pharmacodynamic (PD) profiles apply, but most importantly, because there is not always a clear PK/PD relationship established. In this communication, authors discuss the usefulness and challenges associated with the estimation of pAUCs in the development of generic LAI products through the review of six case studies under the lens of regulatory requirements from the two major authorities, namely the FDA and EMA. [ABSTRACT FROM AUTHOR]

Aspect category sentiment analysis aims to automatically identify the sentiment polarities of aspect categories mentioned in text, and is widely used in the data analysis of product reviews and social media. Most existing studies typically limit themselves to utilizing sentence-level local information, thereby failing to fully exploit the potential of document-level and corpus-level global information. To address these limitations, we propose a model that integrates global information for aspect category sentiment analysis, aiming to leverage sentence-level, document-level, and corpus-level information simultaneously. Specifically, based on sentences and their corresponding aspect categories, a graph neural network is initially built to capture document-level information, including sentiment consistency within the same category and sentiment similarity between different categories in a review. We subsequently employ a memory network to retain corpus-level information, where the representations of training instances serve as keys and their associated labels as values. Additionally, a k-nearest neighbor retrieval mechanism is used to find training instances relevant to a given input. Experimental results on four commonly used datasets from SemEval 2015 and 2016 demonstrate the effectiveness of our model. The in-depth experimental analysis reveals that incorporating document-level information substantially improves the accuracies of the two primary 'positive' and 'negative' categories, while the inclusion of corpus-level information is especially advantageous for identifying the less frequently occurring 'neutral' category. [ABSTRACT FROM AUTHOR]

Sentiment analysis via artificial intelligence, i.e., machine learning and large language models (LLMs), is a pivotal tool that classifies sentiments within texts as positive, negative, or neutral. It enables computers to automatically detect and interpret emotions from textual data, covering a spectrum of feelings without direct human intervention. Sentiment analysis is integral to marketing research, helping to gauge consumer emotions and opinions across various sectors. Its applications span analyzing movie reviews, monitoring social media, evaluating product feedback, assessing employee sentiments, and identifying hate speech. This study explores the application of both traditional machine learning and pre-trained LLMs for automated sentiment analysis of customer product reviews. The motivation behind this work lies in the demand for more nuanced understanding of consumer sentiments that can drive data-informed business decisions. In this research, we applied machine learning-based classifiers, i.e., Random Forest, Naive Bayes, and Support Vector Machine, alongside the GPT-4 model to benchmark their effectiveness for sentiment analysis. Traditional models show better results and efficiency in processing short, concise text, with SVM in classifying sentiment of short length comments. However, GPT-4 showed better results with more detailed texts, capturing subtle sentiments with higher precision, recall, and F1 scores to uniquely identify mixed sentiments not found in the simpler models. Conclusively, this study shows that LLMs outperform traditional models in context-rich sentiment analysis by not only providing accurate sentiment classification but also insightful explanations. These results enable LLMs to provide a superior tool for customer-centric businesses, which helps actionable insights to be derived from any textual data. [ABSTRACT FROM AUTHOR]

Traditional e-commerce recommendation systems often struggle with dynamic user preferences and a vast array of products, leading to suboptimal user experiences. To address this, our study presents a Personalized Adaptive Multi-Product Recommendation System (PAMR) leveraging transfer learning and Bi-GRU (Bidirectional Gated Recurrent Units). Using a large dataset of user reviews from Amazon and Flipkart, we employ transfer learning with pre-trained models (AlexNet, GoogleNet, ResNet-50) to extract high-level attributes from product data, ensuring effective feature representation even with limited data. Bi-GRU captures both spatial and sequential dependencies in user-item interactions. The innovation of this study lies in the innovative feature fusion technique that combines the strengths of multiple transfer learning models, and the integration of an attention mechanism within the Bi-GRU framework to prioritize relevant features. Our approach addresses the classic recommendation systems that often face challenges such as cold start along with data sparsity difficulties, by utilizing robust user and item representations. The model demonstrated an accuracy of up to 96.9%, with precision and an F1-score of 96.2% and 96.97%, respectively, on the Amazon dataset, significantly outperforming the baselines and marking a considerable advancement over traditional configurations. This study highlights the effectiveness of combining transfer learning with Bi-GRU for scalable and adaptive recommendation systems, providing a versatile solution for real-world applications. [ABSTRACT FROM AUTHOR]

In a person's hectic schedule, online buying is a more convenient option. However, this retail center is teeming with fictitious products, fake brands, and low-quality goods. It is the job of shopping sites to employ some kind of sophisticated algorithm to assess the popularity, suggestion, and quality of each accessible product and brand for customers. These shopping sites collect descriptive reviews or comments from users, and sentiment analysis is performed to determine the quality, popularity, and adaptability of these products. A client submits text-based reviews to share his thoughts, opinions, and feelings about the advantage or loss he experienced with that deal. On a shopping or emarket site, thousands of reviews are given by customers, making it impossible to assess each review individually. Such sites can apply a customer satisfaction analysis algorithm to exactly measure the evaluations provided. Using this customer satisfaction research, we may determine whether the product was favorably received by the buyer. However, conventional sentiment analysis methods struggle when a single feeling is combined with many sentiments of distinct elements. For evaluating feelings, an intelligent aspect-based and weighted assessment adaptive CWA-wbiLSTM Model is proposed in this study. The processing featureset in this model is enhanced by combining the BoW and Handcrafted weighted features. The aspect-oriented clustering stage categorizes and strengthens this extended featureset based on aspects and weights. For classifying the reviews, an updated aspect-oriented and weighted LSTM model is implemented in the final stage. The proposed methodology has been tested on Amazon product reviews for cell phones, video games, and electronics. The models being compared include SVM, LSTM, CNN, RNN, biLSTM, Tree + LSTM, and LSTM + FL. The proposed model was 96.15% accurate for Amazon Cell Reviews, 97.98% accurate for Amazon Video Games, and 97.42% accurate for Electronics Products. SVM, CNN, LSTM, RNN, BiLSTM, Tree-LSTM, and LSTM + FL models yield significant increases in an average accuracy of 24.91%, 41.54%, 7.67%, 8.02%, 7.74%, 7.39%, and 2.16%. Current SVM, LSTM, RNN, BiLSTM, Tree-LSTM, and LSTM + FL models have an average accuracy of 72.56%, 88.93%, 88.91%, 89.10%, 89.12%, and 94.16%, respectively. The suggested CWA-wbiLSTM model surpassed all of the prior models, with an accuracy of 97.98%. The proposed model achieved a fantastic accuracy of 96.92% for quality, 96.26% for look, 96.56% for price, 99.31% for color, and 96.90% for size. The findings corroborated each measure's genuine precision rate. With maximum precision and recall rates of 0.981 and 0.98, respectively, the suggested CWA-wbiLSTM model surpassed both baseline machine and deep learning models. [ABSTRACT FROM AUTHOR]

In e‐commerce, online intermediaries give suppliers ready access to consumers and have thereby become an important distribution channel. They can operate as resellers (under a reselling contract) or marketplaces (under an agency contract) for suppliers. This study analyzes the most effective contract arrangement between an online intermediary and two competing suppliers. The suppliers receive online reviews of their products and also be affected by the spillover effect of the intermediary's channel on their own direct sales channels. The intermediary can offer a reselling contract to one supplier and an agency contract to the other or offer the same contract type to both suppliers. We find that when the cross‐channel spillover is positive or only moderately negative, the intermediary offers different contract types if the difference in the review ratings of the two suppliers' products is considerable; otherwise, a reselling contract is offered to both. However, when the cross‐channel spillover is highly negative, the intermediary may offer an agency contract to both suppliers. Our research provides practical guidance for intermediaries on the most effective contract strategies to use when collaborating with suppliers who receive varied online reviews, depending on the cross‐channel spillover scenarios. [ABSTRACT FROM AUTHOR]

Product information asymmetry exists between firms and online customers. Online customer reviews offer product quality information and reduce information asymmetry. Many customers review products online upon receipt and again after use. Initial and additional reviews can be consistent or conflicting, which affects other customers' perceived product value. Firms use different strategies for disclosing product featured quality information, such as exaggerating or depreciating product quality. Different strategies can lead to different customer perceptions, which can affect additional reviews. In addition, even with the same strategy, differences in the degree of information quality can affect the perceived value of customers. [ABSTRACT FROM AUTHOR]

Tenuazonic acid (TeA) is a secondary toxic metabolite produced by fungi species in the genus Alternaria. It is frequently reported as an 'emerging mycotoxin', a group of mycotoxins which have been neither routinely determined nor legislatively regulated. This study aimed at systematically review the occurrence data of TeA in food products from 2011 to date. TeA was detected in numerous foodstuffs from several countries, especially in tomato products, herbs, spices, and cereals. Further studies are needed to understand the occurrence patterns of TeA in different foodstuffs, to orientate preventive measures to avoid Alternaria species and TeA in food production environments. [ABSTRACT FROM AUTHOR]

Online reviews are playing an increasingly important role in consumers' online shopping decisions. Studies on investigating the influencing factors of review helpfulness have been conducted extensively in recent years. Previous studies have proved the positive relationship between review characteristics and review helpfulness, for instance, review depth, review valence, image assistance, and identity disclosure. While, research has neglected the effect of presentation format on review helpfulness, especially its combined effect with other review characteristics. This study adopts online review data to examine how presentation format affects review helpfulness with the influence of word count and the number of responses. Besides, this study also reveals how presentation format moderates the effect of word count and the number of responses on review helpfulness across two product types. The results indicate that video-based reviews are perceived as more helpful than other formats. However, the number of responses is less important for video-based reviews, especially for search products. Moreover, word count of image-based reviews is less influential compared to text-based reviews on review helpfulness. It suggests that consumers may ignore the textual information when images enhance reviews. At last, the theoretical contributions, the practical implications, and limitations are discussed. [ABSTRACT FROM AUTHOR]

E-commerce has experienced rapid growth in recent years and continues to expand dynamically. In this sector, maximizing customer satisfaction and enhancing the shopping experience are recognized as important strategic initiatives. To maximize customer satisfaction, it is essential to accurately determine customer needs and provide appropriate solutions to meet demand. In this context, feedback obtained from customers holds significant importance. However, customer comments often contain spelling errors, complicating the analysis of these comments. This study aims to automatically correct spelling errors in user comments regarding products sold on the e-commerce site Trendyol.com. For this purpose, a system based on transformer architecture has been created. Various spelling error detection and correction models were subsequently developed based on this architecture. Prediction models have been developed using two separate datasets consisting of Trendyol user comments and two additional datasets, including the Turkish Spelling Check Dataset taken from the Hunspell library, and the effects of these four datasets on prediction performance have been examined. The success of the models has been evaluated using the Accuracy metric. The performance of the developed models was also compared with that of the model in the Zemberek library. As a result of the study, it has been observed that the utilization of the Turkish Spelling Check Dataset positively influences prediction performance. The developed system enhanced customer experience by correcting spelling errors in comments. [ABSTRACT FROM AUTHOR]

The food industry is increasingly exploring alternative protein sources, like pulses, to replace animal proteins in human diets. This review investigates the diversity of proteins in pulses and their extraction methods, which significantly affect their functional properties. Understanding these techniques is crucial for optimizing the use of pulse proteins in food products, especially in enhancing the nutritional value of baked goods. However, incorporating pulse proteins can impact processing properties, affecting viscosity, rheological behavior, dough volume, and sensory characteristics. This review provides a comprehensive overview of pulse protein extraction, functional properties, and their effects on baked products. [ABSTRACT FROM AUTHOR]

Sentiment analysis (SA) of several user evaluations on e-commerce platforms can be used to increase customer happiness. This method automatically extracts and identifies subjective data from product evaluations using natural language processing (NLP) and machine learning (ML) methods. These statistics may eventually reveal information on the favourable, neutral, or negative attitudes of the consumer base. Due to its capacity to grasp the complex links between words and phrases in reviews as well as the emotions they imply, deep learning (DL) is very useful for SA tasks. A unique approach termed Weighted Parallel Hybrid Deep Learning-based Sentiment Analysis on E-Commerce Product Reviews (WPHDL-SAEPR) is introduced by the proposed system. Accurately distinguishing between distinct sentiments found in online store reviews is the aim of the WPHDL-SAEPR technique. Additional data pre-processing processes are implemented within the WPHDL-SAEPR architecture to guarantee compatibility. Words are embedded into the paper using the word2vec model, while sentiment is classified using the WPHDL model. The Restricted Boltzmann Machine (RBM) and Singular Value Decomposition (SVD) models are combined in this model. The results of the WPHDL-SAEPR approach's simulation were assessed using a consumer review database, with the results being emphasized at each stage. [ABSTRACT FROM AUTHOR]

Social connections are frequently leveraged as supplementary data to enhance recommendation systems (RS). In real-world scenarios, the relationships between multiple users within social networks exhibit considerable complexity and diversity. However, prevalent recommendation approaches based on graphs tend to assume pairwise interactions to uncover user preferences. This approach overlooks the influence of intricate and multifaceted social relationships on user preferences, representing a neglect of the higher-order complexities inherent in user relations. It is also observed that most approaches neglect the fact that similar items have similar shared interests when grouped with users. This shows that there is an active connection among the attributes of items and users. Hence, this study works on the drawback of current Social RS: 1. Relations among the items are not explored well, 2. Social relations and homogeneity are not diverse. To address these issues, this study introduces a novel methodology based on clustering and heterogenous hypergraph, termed as HyperSegRec. HyperSegRec (Hypergraph Segmentation based RS) encompasses three parallel processes: User segmentation through Clustering, Item Similarity through extended graph using the cosine similarity matrix method, and a Weighted aggregation process, these processes models the user-item (where user-user as well as user-item relations is mined through clustering and hypergraph representation) and the item-item relations are mined through an extended graph matrix which is constructed with the cosine similarity method and hypergraph representation. These process in the proposed approach are trained individually over BPR (Bayesian Personalized Ranking) based loss function and evaluated against state-of-the-art models based on Recall and NDCG for top-K (K = 10, 20, 50) Recommendations. The proposed model is evaluated on two datasets: Movieslens 100 k and Amazon beauty products review. Compared to other state-of-the-art models, HyperSegRec achieves a 9.92% improvement on NDGC@50. Hence, HyperSegRec demonstrates superior performance in capturing user preferences and providing precise recommendations compared to existing state-of-the-art models, showcasing its potential to enhance recommendation systems for diverse datasets. [ABSTRACT FROM AUTHOR]

Purpose: We investigate the joint impacts of three trust cues – content, sentiment and helpfulness votes – of online product reviews on the trust of reviews and attitude toward the product/service reviewed. Design/methodology/approach: We performed three studies to test our research model, presenting participants with scenarios involving product reviews and prior users' helpful and unhelpful votes across experimental settings. Findings: A high helpfulness ratio boosts users' trust and influences behaviors in both positive and negative reviews. This effect is more pronounced in attribute-based reviews than emotion-based ones. Unlike the ratio effect, helpfulness magnitude significantly impacts only negative attribute-based reviews. Research limitations/implications: Future research should investigate voting systems in various online contexts, such as Facebook post likes, Twitter microblog thumb-ups and up-votes for article comments on platforms like The New York Times. Practical implications: Our findings have significant implications for voting system-providers implementing information techniques on third-party review platforms, participatory sites emphasizing user-generated content and online retailers prioritizing product awareness and reputation. Originality/value: This study addresses an identified need; that is, the helpfulness votes as an additional trust cue and the joint effects of three trust cues – content, sentiment and helpfulness votes – of online product reviews on the trust of customers in reviews and their consequential attitude toward the product/service reviewed. [ABSTRACT FROM AUTHOR]

Chinese bayberry (Myrica rubra Sieb. et Zucc.) fruit (CBF) is widely growing in eastern Asia, mainly in China. CBFs contain a high number of polyphenols, especially anthocyanins. However, CBF can be easily damaged and cannot be stored for more than 3 days at 20–22 °C and 9–12 days at 0–2 °C. Thus, this review explores the potential options for the long-term preservation and consumption of CBFs, as well as the importance of CBF by-products. The review also addressed the biological activity of CBFs and their food products. The consumption period of CBF can be increased by employing a variety of processing techniques, including rapid freezing, drying, juicing, making wine and vinegar, and adding it to meals. Additionally, oil and bioactive substances can be extracted from the CBF by-products, such as kernel and pomace. The greatest options to extend the duration of CBF ingestion are thought to be CBF powder, juice, and wine. Furthermore, CBFs and their derivatives exhibit many biological activities such as antibacterial, anticancer, antidiabetic, and antioxidant properties. The two biological activities that were most potent were antidiabetic and antioxidant. However, more biological research, particularly clinical-based research, is advised to investigate the potential health benefits of CBF and its products. [ABSTRACT FROM AUTHOR]

Despite their heightening popularity and potential individual and societal implications, little scientific attention has been given to the news coverage of direct-to-consumer genetic testing (DTC-GT) kits (e.g. 23andMe, AncestryDNA, and MyHeritage). To accommodate, this study leverages the Analysis of Topic Model Networks to inductively develop a comprehensive overview of how major U.S. news media framed DTC-GT kits between 2009 and 2022, and then explicates the relationship of such coverage with kit sales between 2012 and 2019. We find four frames through which the news media talk about the kits: Utility, Scientific, Political-Economic, and Commercial. Further, as kit sales increase over time, news media maintain a strong emphasis on the utility of kits in their coverage, but also increasingly promote kits through product reviews. We find that this news coverage is influenced by―but does not influence―kit sales, and discuss implications pertaining to the DTC-GT industry and to biotechnology more broadly. [ABSTRACT FROM AUTHOR]

The emergence of e-commerce is introduced in the golden era. E-commerce product reviews are comments generated by customers of online shopping to estimate the service and product qualities having purchased; these remarks aid users in identifying the facts of the product. The sentiment polarity of e-commerce product analyses is the optimal method to get consumer opinions on a service or product. Hence, sentiment analysis (SA) of product remarks on e-commerce platforms is much more influential. Deep learning (DL) analysis of online consumer feedback can identify user behavior toward a sustainable future. Artificial intelligence (AI) can acquire perceptions from product evaluations to develop efficient products. The main challenge is that numerous ethical products do not satisfy customers' expectations owing to the gap among users' expectations and their perception of sustainable products. This paper focuses on the design of the Fusion of Artificial Intelligence Deep Learning Model for Product Reviews on E-Commerce (FAIDLM-PREC) model. The main intention of FAIDLM-PREC method is to appropriately distinguish the dissimilar types of sentiments that occur in the e-commerce reviews. Initially, data preprocessing is executed to increase the product review quality with Glove based word embedding method. For product reviews classification, the FAIDLM-PREC approach evolves fusion of dual DL methods namely Bidirectional Long Short-Term Memory (Bi-LSTM) and gated recurrent unit (GRU) methods. Eventually, the parameters relevant to the two DL methods are perfectly modified utilizing the Archimedes optimization algorithm (AOA). An extensive experiment of the FAIDLM-PREC technique was conducted utilizing customer review database and outcomes indicated that the FAIDLM-PREC technique highlighted betterment over other recent methods to several measures. [ABSTRACT FROM AUTHOR]

Companies are increasingly introducing conversational reviews—reviews solicited via chatbots—to gain customer feedback. However, little is known about how chatbot-mediated solicitation influences rating valence and review helpfulness compared to conventional online forms. Therefore, we conceptualized these review solicitation media on the continuum of anthropomorphism and investigated how various levels of anthropomorphism affect rating valence and review helpfulness, showing that more anthropomorphic media lead to more positive and less helpful reviews. We found that moderate levels of anthropomorphism lead to increased interaction enjoyment, and high levels increase social presence, thus inflating the rating valence and decreasing review helpfulness. Further, the effect of anthropomorphism remains robust across review solicitors' salience (sellers vs. platforms) and expressed emotionality in conversations. Our study is among the first to investigate chatbots as a new form of technology to solicit online reviews, providing insights to inform various stakeholders of the advantages, drawbacks, and potential ethical concerns of anthropomorphic technology in customer feedback solicitation. [ABSTRACT FROM AUTHOR]

This paper explores the diversity of people's opinions. We have chosen as the research material a corpus of customer reviews on robotic vacuum cleaners in the Russian language. The study aims to identify unique perspectives and recurring patterns in user feedback, examining how individuals express their satisfaction or dissatisfaction with the product. In particular, we estimate the growth rate of the number of different opinions with an increase of the corpus size. By employing both qualitative and quantitative methods, the research explores the linguistic strategies used by users to convey their experiences and emotions. We utilize a large language model (LLM) as a tool for extracting and analyzing opinions from product reviews.

Language is critical to the effectiveness of marketing messages. Achieving a desired outcome requires arranging words to formulate a message (i.e., syntax), but this task is not trivial. The authors study the role of syntax in marketing communications by focusing on syntactic surprise (i.e., how unexpected the syntax of a message is). They introduce a measure that captures syntactic surprise, establishes its internal and external validity, and tests its effectiveness for marketing messages. In a series of studies that include field data and randomized field experiments from contexts such as donations, advertising, and product reviews, the authors show that a message's syntactic surprise is related to its effectiveness. This relationship follows an inverted U-shape, such that medium-syntactic-surprise messages are the most effective. The authors then conduct experiments on Facebook and Instagram to demonstrate how these findings can be used to write effective marketing messages. In collaboration with two independent companies, they show that ads for products, services, or jobs that are written with medium syntactic surprise result in higher click-through rates than ads written with low or high syntactic surprise. [ABSTRACT FROM AUTHOR]

ABSTRACT The rapid growth of online product reviews has made it increasingly challenging for consumers to make informed purchase decisions. However, the abundance of reviews, including fake or augmented and sarcastic reviews, poses a challenge for consumers. To address this challenge, this paper introduces the TOPS (Trusted Opinion analysis of Product reviewS) framework, a novel approach that leverages a hybrid deep learning‐based D2CL (Dual Deep leaning based cleaning) filter to enhance the reliability of online reviews. The proposed methodology employs the D2CL filter to identify and eliminate fake and sarcastic reviews, ensuring that the consolidated sentiment analysis provides users with trustworthy opinions. The framework is equipped with the R‐mGRU, a hybrid deep learning model specifically designed to tackle the nuances of product reviews. This model has demonstrated impressive accuracy rates, achieving 89%, 91%, and 94% for fake, sarcasm, and sentiment analysis tasks, respectively. The TOPS framework makes a significant contribution to improving the overall quality and authenticity of product reviews, empowering consumers with more reliable information for informed decision‐making in online shopping scenarios. [ABSTRACT FROM AUTHOR]

Alkali-activated materials (AAMs) are regarded as a substitute for Portland cement. They have high chloride resistance and a low carbon dioxide footprint. The aim of this review is to provide a multi-scale perspective to understand material–product–microstructure–property relationships in terms of the chloride binding behaviour of AAMs. The physical and chemical chloride stability of different reaction products is summarised from nanostructure, to microstructure to macro properties. An analysis of studies in the literature gives an overview of recent progress in chloride transport in AAMs influenced by different reaction products. Results show that a higher calcium/silicon, aluminium/silicon molar ratios and alkali content increase the formation of amorphous phases, leading to a denser microstructure and lower chloride penetration in AAMs. Higher magnesium oxide and aluminium oxide contents result in increased formation of hydrotalcite. The enhanced physical and chemical absorption of chloride by hydrotalcite leads to higher resistance of chloride penetration in AAMs. Investigation of increasing chloride resistance could potentially focus on increasing gel and hydrotalcite formation. [ABSTRACT FROM AUTHOR]

Based on the perspective of identifying the water army, this paper uses the methods of machine learning and data visualization to analyze the helpfulness of online produce reviews, portray product portraits, and provide real and helpful product reviews. In order to identify and eliminate the water army, the Term Frequency-Inverse Document Frequency Model (TF-IDF) and Latent Semantic Index Model (LSI) are used. After eliminating the water army, three classification methods were selected to perform sentimental analysis, including logistics, SnowNLP, and Convolutional Neural Network for text(TextCNN). The TextCNN has the highest F1 score among the three classification methods. At the same time, the Latent Dirichlet Allocation mode (LDA) is used to extract the topics of various reviews. Finally, targeted countermeasures are proposed to manufacturers, consumers, and regulators. [ABSTRACT FROM AUTHOR]

Conventional radiography (CR) is primarily utilized for fracture diagnosis. Artificial intelligence (AI) for CR is a rapidly growing field aimed at enhancing efficiency and increasing diagnostic accuracy. However, the diagnostic performance of commercially available AI fracture detection solutions (CAAI-FDS) for CR in various anatomical regions, their synergy with human assessment, as well as the influence of industry funding on reported accuracy are unknown. Peer-reviewed diagnostic test accuracy (DTA) studies were identified through a systematic review on Pubmed and Embase. Diagnostic performance measures were extracted especially for different subgroups such as product, type of rater (stand-alone AI, human unaided, human aided), funding, and anatomical region. Pooled measures were obtained with a bivariate random effects model. The impact of rater was evaluated with comparative meta-analysis. Seventeen DTA studies of seven CAAI-FDS analyzing 38,978 x-rays with 8,150 fractures were included. Stand-alone AI studies (n = 15) evaluated five CAAI-FDS; four with good sensitivities (> 90%) and moderate specificities (80–90%) and one with very poor sensitivity (< 60%) and excellent specificity (> 95%). Pooled sensitivities were good to excellent, and specificities were moderate to good in all anatomical regions (n = 7) apart from ribs (n = 4; poor sensitivity / moderate specificity) and spine (n = 4; excellent sensitivity / poor specificity). Funded studies (n = 4) had higher sensitivity (+ 5%) and lower specificity (-4%) than non-funded studies (n = 11). Sensitivity did not differ significantly between stand-alone AI and human AI aided ratings (p = 0.316) but specificity was significantly higher the latter group (p < 0.001). Sensitivity was significant lower in human unaided compared to human AI aided respectively stand-alone AI ratings (both p ≤ 0.001); specificity was higher in human unaided ratings compared to stand-alone AI (p < 0.001) and showed no significant differences AI aided ratings (p = 0.316). The study demonstrates good diagnostic accuracy across most CAAI-FDS and anatomical regions, with the highest performance achieved when used in conjunction with human assessment. Diagnostic accuracy appears lower for spine and rib fractures. The impact of industry funding on reported performance is small. [ABSTRACT FROM AUTHOR]

In this article, an intensive sentiment analysis approach termed Hierarchical attention‐convolutional neural network (HAN‐CNN) has been proposed using product reviews. Firstly, the input product review is subjected to Bidirectional Encoder Representation from Transformers (BERT) tokenization, where the input data of each sentence are partitioned into little bits of words. Thereafter, Aspect Term Extraction (ATE) is carried out and feature extraction is completed utilizing some features. Finally, sentiment analysis is accomplished by the developed HAN‐CNN, which is formed by combining a Hierarchical Attention Network (HAN) and a Convolutional Neural Network (CNN). Moreover, the proposed HAN‐CNN achieved a greater performance with maximum accuracy, recall and F1‐Score of 91.70%, 90.60% and 91.20%, respectively. [ABSTRACT FROM AUTHOR]

BACKGROUND A successfully developed topical botulinum toxin product would overcome the disadvantages of injectable administration commonly seen with all available botulinum products, including the potential pain, bleeding, and bruising associated with injection. OBJECTIVE Identify the disadvantages of injectable administration of botulinum products and advantages of topical administration. Identify previous attempts to develop topical products and review the only topical product candidate (ET-01, abbreviated EMX) under clinical development. MATERIALS AND METHODS A literature review regarding injectable botulinum products and topical botulinum product candidates and an analysis of data regarding EMX development were performed. RESULTS Multiple disadvantages of injectable administration and advantages of topical administration of botulinumtoxin were identified. If available, topical botulinum products are expected to expand the clinical use of botulinum toxin. Several previous attempts to develop topical products were identified. EMX has been studied for lateral canthal line treatment in multiple controlled phase 2 clinical trials including over 700 subjects. EMX has demonstrated statistically significant positive results. CONCLUSION The development of a topical botulinum toxin is appealing because it overcomes multiple disadvantages observed with injectable administration and would likely result in increased clinical use of botulinum toxin. The development of EMX shows promise and merits further study. [ABSTRACT FROM AUTHOR]

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

Summary: The baking industry is greatly interested in incorporating banana peel and mango peel into a broader spectrum of baked goods. This review delves into the innovative utilisation of banana and mango peels in various baked products, specifically focusing on biscuits, cookies, cakes, and cake doughnuts. Incorporating banana peel in baked goods has gained attention for its potential to enhance flavour, texture, and nutritional value. Similarly, mango peel has been explored for its unique properties and potential contributions to baked items. This review emphasises the health benefits of banana and mango peels, including antioxidants, fibre, and other nutritional advantages that help prevent diseases and promote overall well‐being. Limited scientific studies may explore using banana peel and mango peel specifically as functional ingredients in baked products. This review highlights the utilisation of banana peel and mango peel in baked products, focusing on their influence on product quality, sensory characteristics, and potential health benefits. It presents a sustainable and innovative approach to creating healthier, more flavourful baked products. [ABSTRACT FROM AUTHOR]

Summary: Meat is widely regarded as the optimal source of protein for human consumption. However, meat is susceptible to spoilage and microbial contamination throughout processing, transportation and storage stages. Traditional thermal processing techniques are commonly employed in the global meat industry. While these methods ensure microbiological safety, they may also impact the sensory and functional attributes of meat products. To address these limitations, non‐thermal processing technologies are progressively replacing conventional thermal approaches. This review will primarily focus on the potential application of non‐thermal processing technologies as alternatives to traditional methods in meat and meat product processing. Various technologies exhibiting promising performance in this field will be presented. Additionally, the mechanisms, advantages and limitations of these non‐thermal technologies in recent years will be summarised, along with a discussion on the challenges and prospects they present. This study provides an overview of non‐thermal processing technology specifically applied to meat products, emphasising the underlying mechanisms associated with different approaches. By employing diverse technical combinations, it is possible to address existing processing limitations and expand their applicability range. Non‐thermal processing technology demonstrates significant potential for enhancing meat processing practices; however, concerns regarding safety aspects and limited clinical verification about to environmental pollution have impeded its widespread adoption within the food industry. [ABSTRACT FROM AUTHOR]

The significance of product attributes, as reflected in online reviews during the initial purchase, may undergo changes following positive or negative product usage experiences. The purpose of this research was to investigate whether positive or negative product consumption experiences influence shifts in the perceived importance of product attributes and subsequent product choices, based on online quality ratings by attributes. A series of experiments demonstrated that the importance associated with an attribute's expectation in the pre‐purchase process shifted after a negative consumption experience with that attribute. The results showed that a change in both attribute importance and product choice following a negative attribute experience. A less important attribute that malfunctioned during the consumption experience became more crucial in the subsequent purchase (Study 1). However, this shift in importance may not be universal for all attributes; it was observed only when the attribute was moderately important (Study 2). This study contributes valuable empirical findings to the existing literature on online reviews by highlighting dynamic shifts in the importance weight of product attributes. E‐commerce practitioners are encouraged to allow formal purchasers to leave product reviews without time constraints, providing a platform for reflecting users' lifelong consumption experiences. [ABSTRACT FROM AUTHOR]

Marketers are concerned about how consumers' readings of product reviews and product rating statistics affect product evaluations. Previous studies generally regard product review content and rating statistics as affecting product evaluations independently. This may not be true. Typically, consumers enter a review section with pre‐existing positive expectations about a product. When they see a negative average rating in this section, their expectancies are disconfirmed. Their surprise stemming from the disconfirmation makes them read the reviews more carefully. Thus, review reading affects product evaluation more strongly when viewing a negative average rating. Conversely, a positive average rating results in expectancy confirmation, which weakens the effect of review reading. An eye‐tracking experiment using a mock‐resort review section supported these predictions. These findings have implications for marketers in behavioral advertising, responding to reviews, and consumers writing reviews. [ABSTRACT FROM AUTHOR]

Many online platforms are now offering free samples to seasoned reviewers, hoping to get feedback. While these reviewers are given free samples to review, they also buy and review products themselves. The regular ratings for the purchased products are the majority. This brings up the question: Does receiving free products make them rate their personal purchases more positively? And if so, why? We explored two possibilities. First, uncertainty reduction mechanism: The idea that trying free samples makes buyers more confident in their purchases, leading to greater satisfaction and higher ratings for the purchased products; Second, reciprocity mechanism: The idea that reviewers might feel obliged to give better ratings as a "thank you" for the free samples or with the expectations of getting more free samples, which could introduce bias. Our research indicates that giving free samples mainly helps in reducing purchase uncertainty, making customers genuinely happier with their subsequent purchases. So, online platforms can benefit from this strategy, as it seems to uplift genuine positive reviews rather than create biased ones. However, it is still essential to monitor for any undue bias to maintain trustworthiness in reviews. Many e-commerce platforms have launched reviewer incentive programs, in which the platforms send free samples to experienced reviewers in exchange for feedback on the samples. This study focuses on an unexplored aspect of this practice: the impact of free samples on reviewers' ratings of subsequently purchased products (i.e., regular ratings). We theorize that free samples can increase regular ratings through two mechanisms: uncertainty reduction and reciprocity. We collect observational data from Amazon.com and find that after beginning to receive free samples from a platform-initiated reviewer incentive program (i.e., Vine), reviewers' regular ratings increased by 2.25% on average (i.e., 0.093 more stars on the five-star scale). Our follow-up analyses at a granular product-category level reveal that the increase in regular ratings occurs primarily for purchased products that share common attributes with free samples that reviewer has received previously. To further delve into the mechanisms at work, we conduct a randomized controlled experiment in which participants are asked to purchase one cookie among three flavor options. Before the cookie purchase, participants either receive nothing in advance (i.e., control group) or receive one of four free samples (i.e., four treatment groups). The results indicate that uncertainty reduction (via the common attribute between the free sample and the options for purchase) significantly mediates the impact of the free sample on the ratings of the purchased cookie. We conclude that platform-initiated free samples can help consumers understand products with similar attributes, thereby reducing prepurchase uncertainty and leading to higher regular ratings. We discuss the managerial and scholarly implications of the findings. History: Ravi Bapna, Senior Editor; Zhenhui (Jack) Jiang, Associate Editor. Funding: This work was supported by the National Natural Science Foundation of China [Grant 72121001]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/isre.2019.0176. [ABSTRACT FROM AUTHOR]

Acquiring a single sentiment score dependent on all the reviews will benefit the buyers and sellers in making the decision more accurately. The raw format of user-generated content lacks a legitimate language structure. It, therefore, acts as an obstacle for applying the Sentiment analysis task, which aims to predict the true emotion of a sentence by providing a score and its nature. This paper concentrates on obtaining a single sentiment score using a hybrid Long Short-Term Memory encoder–decoder model. This research uses the text normalization process to transform the sentences consisting of noise, appearing as incorrect grammar, abbreviations, freestyle, and typographical errors, into their canonical structure. The experimental outcomes confirm that the proposed hybrid model performs well in standardizing the raw E-commerce website review, enriched with hidden information and provided a single sentiment score influenced by all the review scores for the product. [ABSTRACT FROM AUTHOR]

Detecting suggestions in online review requires contextual understanding of review text, which is an important real-world application of natural language processing. Given the disparate text domains found in product reviews, a common strategy involves fine-tuning bidirectional encoder representations from transformers (BERT) models using reviews from various domains. However, there hasn't been an empirical examination of how BERT models behave across different domains in tasks related to detecting suggestion sentences from online reviews. In this study, we explore BERT models for suggestion classification that have been fine-tuned using single-domain and cross-domain Amazon review datasets. Our results indicate that while single-domain models achieved slightly better performance within their respective domains compared to cross-domain models, the latter outperformed single-domain models when evaluated on cross-domain data. This was also observed for single-domain data not used for fine-tuning the single-domain model and on average across all tests. Although fine-tuning single-domain models can lead to minor accuracy improvements, employing multi-domain models that perform well across domains can help in cold start problems and reduce annotation costs. [ABSTRACT FROM AUTHOR]

As the online market grows rapidly, people are relying more on product review when they purchase the product. Hence, many companies and researchers are interested in analyzing product review which essentially a text data. In the current literature, it is common to use only text analysis tools to analyze text dataset. But in our work, we propose a method that utilizes both text analysis method such as topic modeling and statistical network model to build network among individuals and find interesting communities. We introduce a promising framework that incorporates topic modeling technique to define the edges among the individuals and form a network and uses stochastic blockmodels (SBM) to find the communities. The power of our proposed method is demonstrated in real-world application to Amazon product review dataset. [ABSTRACT FROM AUTHOR]

This study examines how unexpected exogenous events, labelled as suprises , affect the utility of experience goods reported in online rating systems. Using over 300,000 reviews of accommodation services listed on Booking.com, the research investigates whether online ratings capture the impact of surprises related to meteorological conditions and whether they create additional biases in service evaluation. The study finds that sudden changes in weather conditions have a significant impact on experienced utility, with the effect varying based on the direction of the surprise. Additionally, in line with the hedonic adaption theory, we find that the duration of consumption moderates the surprise effect, reducing its impact on reported utility. • Online reviews signal product/service quality. • External factors may influence consumers' quality assessments. • We perform an empirical analysis of unexpected externalities on experience goods evaluations. • Positive/negative surprises have asymmetric impacts on reported utility. • Surprise effects on utility are moderated by the consumption span. [ABSTRACT FROM AUTHOR]

Food waste production reaches actually about 1.3 billion tonnes per year, corresponding to the emission of 3.3 billion tonnes equivalent of CO2, thus calling for improved recycling. Here we review food waste conversion into energy and products such as biohydrogen, biogas, biofuel, biodiesel, biochar, bioplastics, fertilizers, animal feed, organic acids, enzymes, and proteins. Food waste can be treated by incineration, pyrolysis, composting, anaerobic digestion, hydrothermal carbonization, and landfilling. Properties of food waste influence the efficiency of conversion. [ABSTRACT FROM AUTHOR]

This article presents validation and conformity testing of the Sentinel-3 Ocean Land Colour Instrument (OLCI) green instantaneous fraction of absorbed photosynthetically active radiation (FAPAR) and OLCI terrestrial chlorophyll index (OTCI) canopy chlorophyll content (CCC) products with fiducial reference measurements (FRM) collected in 2018 and 2021 over two sites (Las Tiesas—Barrax, Spain, and Wytham Woods, UK) in the context of the European Space Agency (ESA) Fiducial Reference Measurement for Vegetation (FRM4Veg) initiative. Following metrological principles, an end-to-end uncertainty evaluation framework developed in the project is used to account for the uncertainty of reference data based on a two-stage validation approach. The process involves quantifying uncertainties at the elementary sampling unit (ESU) level and incorporating these uncertainties in the upscaling procedures using orthogonal distance regression (ODR) between FRM and vegetation indices derived from Sentinel-2 data. Uncertainties in the Sentinel-2 data are also accounted for. FRM-based high spatial resolution reference maps and their uncertainties were aggregated to OLCI's native spatial resolution using its apparent point spread function (PSF). The Sentinel-3 mission requirements, which give an uncertainty of 5% (goal) and 10% (threshold), were considered for conformity testing. GIFAPAR validation results revealed correlations > 0.95, RMSD ~0.1, and a slight negative bias (~−0.06) for both sites. This bias could be partly explained by the differences in the FAPAR definitions between the satellite product and the FRM-based reference. For the OTCI-based CCC, leave-one-out cross-validation demonstrated correlations > 0.8 and RMSDcv ~0.28 g·m−2. Despite the encouraging validation results, conclusive conformity with the strict mission requirements was low, with most cases providing inconclusive results (driven by large uncertainties in the satellite products as well as by the uncertainties in the upscaling approach). It is recommended that mission requirements for bio-geophysical products are reviewed, at least at the threshold level. It is also suggested that the large uncertainties associated with the two-stage validation approach may be avoided by directly comparing with spatially representative FRM. [ABSTRACT FROM AUTHOR]

The abundance of publicly available data on the internet within the e-marketing domain is consistently expanding. A significant portion of this data revolve around consumers' perceptions and opinions regarding the goods or services of organizations, making it valuable for market intelligence collectors in marketing, customer relationship management, and customer retention. Sentiment analysis serves as a tool for examining customer sentiment, marketing initiatives, and product appraisals. This valuable information can inform decisions related to future product and service development, marketing campaigns, and customer service enhancements. In social media, predicting ratings is commonly employed to anticipate product ratings based on user reviews. Our study provides an extensive benchmark comparison of different deep learning models, including convolutional neural networks (CNN), recurrent neural networks (RNN), and bi-directional long short-term memory (Bi-LSTM). These models are evaluated using various word embedding techniques, such as bi-directional encoder representations from transformers (BERT) and its derivatives, FastText, and Word2Vec. The evaluation encompasses two setups: 5-class versus 3-class. This paper focuses on sentiment analysis using neural network-based models for consumer sentiment prediction by evaluating and contrasting their performance indicators on a dataset of reviews of different products from customers of an online women's clothes retailer. [ABSTRACT FROM AUTHOR]

This research aims to investigate the impact of TikTok marketing content on impulsive purchasing moderated by product reviews, focusing on cosmetic products with the highest sales in Indonesia, specifically Luxcrime as the brand under study. The research method employed is quantitative research involving 100 respondent samples. Data management was conducted using Smart-PLS 4.0. The findings reveal that, without considering the contribution of the moderating variable, TikTok marketing content significantly influences impulsive purchasing. However, when considering the moderating effect of product reviews, TikTok marketing content does not significantly affect impulsive purchasing. The study also found that TikTok marketing content and product reviews collectively explain 40.9% of the variance in impulsive purchasing behavior. In conclusion, impulsive purchasing of Luxcrime cosmetic products can be significantly influenced by TikTok marketing content. Furthermore, product reviews were found to not significantly influence TikTok marketing content and impulsive purchasing when serving as a moderating variable, but they may have a significant impact on TikTok marketing content when acting as an independent variable Keywords: TikTok Marketing; Impulsive Purchasing; Cosmetic Products; Luxcrime; Product Review

The "Natural Product Reports" journal published by The Royal Society of Chemistry features critical reviews in natural product research. The latest issue includes articles on organismic interactions in the discovery of bioactive natural products and the isolation, biological activity, and synthesis of isoquinoline alkaloids. The journal also covers topics such as the assembly of monoterpene indole alkaloids oligomers, tools for identifying bioactive natural products, and the metabolic engineering of plant specialized benzylisoquinoline alkaloids. [Extracted from the article]