Search Results (299)

The rapid expansion of e-commerce has significantly influenced consumer purchasing behavior, making user reviews a critical source of product-related information. However, the large volume of low-quality and superficial reviews limits the ability to obtain reliable insights. This study aims to classify Turkish e-commerce reviews as either useful or useless, thereby highlighting high-quality content to support more informed consumer decisions. A dataset of 15,170 Turkish product reviews collected from major e-commerce platforms was analyzed using traditional machine learning approaches, including Support Vector Machines and Logistic Regression, and transformer-based models such as BERT and RoBERTa. In addition, a novel Multi-Transformer Fusion Framework (MTFF) was proposed by integrating BERT and RoBERTa representations through concatenation, weighted-sum, and attention-based fusion strategies. Experimental results demonstrated that the concatenation-based fusion model achieved the highest performance with an F1-score of 91.75%, outperforming all individual models. Among standalone models, Turkish BERT achieved the best performance (F1: 89.37%), while the BERT + Logistic Regression hybrid approach yielded an F1-score of 88.47%. The findings indicate that multi-transformer architectures substantially enhance classification performance, particularly for agglutinative languages such as Turkish. To improve the interpretability of the proposed framework, SHAP (SHapley Additive exPlanations) was employed to analyze feature contributions and provide transparent explanations for model predictions, revealing that the model primarily relies on experience-oriented and semantically meaningful linguistic cues. The proposed approach can support e-commerce platforms by automatically prioritizing high-quality and informative reviews, thereby improving user experience and decision-making processes. Full article

We developed a large language model-powered system that classifies complaint categories and adapts response styles for e-commerce reviews. By integrating sentiment clustering, zero-shot classification, and style-conditioned prompt engineering, it enables context-aware, emotionally aligned reply generation for enhancing automated customer interaction and reputation management. Full article

In digital marketplaces, trust in e-commerce platforms has evolved from a protective heuristic into a powerful mechanism of behavioral conditioning. This review interrogates how trust cues such as star ratings, fulfillment badges, and platform reputation shape consumer cognition, systematically displace critical evaluation, and create asymmetries in perceived quality. Drawing on over 47 high-quality studies across experimental, survey, and modeling methodologies, we identify seven interlocking dynamics: (1) cognitive outsourcing via platform trust, (2) reputational arbitrage by low-quality sellers, (3) consumer loyalty despite disappointment, (4) heuristic conditioning through trust signals, (5) trust inflation through ratings saturation, (6) false security masking structural risks, and (7) the shift in consumer trust from brands to platforms. Anchored in dual process theory, this synthesis positions trust not merely as a transactional enabler but as a socio-technical artifact engineered by platforms to guide attention, reduce scrutiny, and manage decision-making at scale. Eventually, platform trust functions as both lubricant and leash: streamlining choice while subtly constraining agency, with profound implications for digital commerce, platform governance, and consumer autonomy. Full article

The growth of e-commerce has led to a rapid increase in user-generated content (UGC), attracting scholars’ attention as a new data source for investigating customer requirements. However, existing requirements analysis methods fail to integrate three critical requirement indicators: stated importance, derived importance, and performance. Using only one or two of these indicators inevitably has its limitations. This paper proposes a novel framework for analyzing and prioritizing customer requirements based on multi-source UGC. First, customer requirements are extracted from online reviews and questions & answers using non-negative matrix factorization. Next, aspect-level sentiment analysis and multi-source data fusion are employed to calculate dual importance and performance. Specifically, we developed an improved importance–performance analysis (IPA) model, named dual importance–performance analysis (Du-IPA), which integrates the three indicators to classify requirement types in a 3D cube with corresponding improvement strategies. Finally, by combining the three indicators, an improved prospect value and PROMETHEE-II are proposed using prospect theory to prioritize CRs for product service improvement. The effectiveness of the proposed method is demonstrated through a case study of fresh food in online retail. Full article

Online review images shape consumer perceptions by offering visual cues of product quality and use. Existing studies focus on aesthetics or object presence but overlook the functional balance among image types. This study introduces the Holistic Image Proportion (HIP)—the ratio of holistic to detailed review images—as a key determinant of visual information completeness. Using deep learning (ResNet-101) to classify over 240,000 images from 4450 clothing products, we find an inverted U-shaped relationship between HIP and sales: a balanced mix (HIP ≈ 0.5) maximizes performance. A follow-up experiment confirms that balanced image composition enhances perceived completeness, which fully mediates its effect on purchase intention. Review sentiment further moderates this relationship, amplifying the effect under positive sentiment. This research extends information completeness theory to visual data, highlighting that completeness emerges from functional image composition rather than quantity or aesthetics, offering new insights for multimodal persuasion and e-commerce design. Full article

This research investigates how hybrid review systems integrating human-generated reviews and AI-generated summaries shape consumer trust and decision-related confidence. Across three controlled experiments conducted in simulated e-commerce environments, when and how hybrid reviews enhance consumer evaluations were examined. Study 1 demonstrates that hybrid reviews, which combine the emotional authenticity of human input with the analytical objectivity of AI, elicit greater levels of review trust and decision confidence than single-source reviews. Study 2 employs an experimental manipulation of presentation order and demonstrates that decision confidence increases when human reviews are presented before AI summaries, because this sequencing facilitates more effective cognitive integration. Finally, Study 3 shows that AI literacy strengthens the positive effect of perceived diagnosticity on confidence, while information overload mitigates it. By explicitly testing these processes across three experiments, this research clarifies the mechanisms through which hybrid reviews operate, identifying authenticity and objectivity as dual mediators, and sequencing, literacy, and cognitive load as critical contextual moderators. This research advances current theories on human–AI complementarity, information diagnosticity, and dual-process cognition by demonstrating that emotional and analytical cues can jointly foster trust in AI-mediated communications. This integrative evidence contributes to a nuanced understanding of how hybrid intelligence systems shape consumer decision-making within digital marketplaces. Full article

As artificial intelligence (AI) agents become deeply integrated into fitness systems, trustworthy human–AI agent interaction has become pivotal for user engagement in smart home fitness (SHF) e-commerce platforms. Grounded in the Computers Are Social Actors (CASA) framework, this study empirically investigates how, acting as AI fitness coaches, AI agents’ technical and social features shape users’ active engagement in the in-home social e-commerce context. A mixed-method approach was employed, combining computational text mining of 17,582 user reviews from fitness e-commerce platforms with a survey (N = 599) of Chinese consumers. The results show that (1) the technical–social features of AI agents serving as AI fitness coaches include visibility, gamification, interactivity, humanness, and sociability; (2) these five technical–social features of AI agents positively influence user compliance via both cognitive and emotional trust in AI agents; (3) these five technical–social features of AI agents serving as AI fitness coaches positively impact active engagement via both cognitive and emotional trust in AI agents. This study extends the CASA framework to the domain of AI coaching by demonstrating the parallel roles of cognitive and emotional trust in AI agents. For designers and managers in the fitness e-commerce industries, this study offers actionable insights for designing AI agents integrating functional and social features that foster trust and drive behavioral outcomes. Full article

Food systems operate as socio-ecological systems (SES) in which governance, markets, and biophysical constraints interact through feedback. However, how resilience capacities accumulate across sequential shocks, particularly in hyper-arid, import-dependent rentier states, remains under-traced. We analyze Qatar’s food-system SES across three distinct stress tests: the 2017–2021 blockade, the COVID-19 pandemic (multi-node logistics and labor shock), and the post-2022 Russia–Ukraine war (global price and agricultural input-cost shock). Using a qualitative longitudinal case-study design, we combine documentary review with process tracing and a two-layer coding scheme that maps interventions to SES components (actors, governance system, resource systems/units, interactions, outcomes/feedback) and to predominant resilience capacities (absorptive, adaptive, transformative). The results indicate path-dependent capability building: the blockade activated rapid buffering and rerouting alongside early adaptive investments; COVID-19 accelerated adaptive reconfiguration via digitized logistics, e-commerce scaling, and targeted controlled-environment agriculture; and the Russia–Ukraine shock validated an institutionalized portfolio (fiscal buffering, reserves, procurement diversification, and upstream linkages). Across episodes, supply continuity was maintained, but resilience gains also generated water–energy–food tradeoffs, shifting pressures toward energy-intensive cooling/desalination and upstream water demands linked to domestic buffers. We conclude that durable resilience in eco-constrained, import-dependent systems requires explicit governance of these tradeoffs through measurable performance criteria, rather than crisis-driven expansion alone. Full article

E-commerce live streaming is increasingly constrained by the “anchor dilemma” of talent shortages and reputational volatility. Virtual anchors are viewed as a critical nexus for intelligent and sustainable e-commerce transformation, offering scalable and low-carbon potential. Yet, their user experience and perception remain underexplored. Methodologically, this study adopts a mixed empirical design combining literature review, expert interviews, and a structured questionnaire survey (N = 309), followed by reliability testing, paired-sample t-tests, and Importance–Performance Analysis (IPA) to assess user expectations and perceptions. The integrated analysis resulted in a framework of fourteen evaluative attributes, within which spectacle and cross-platformity emerged as distinguishable dimensions observed in participants’ assessments. The results show that expectations (M = 4.41) significantly exceed perceptions (M = 3.74), with all 14 importance–performance gaps reaching significance. Interactivity, professionalism, and technological maturity emerged as priority areas for improvement, while spectacle and novelty were confirmed as key advantages, and credibility and emotional bonding outperformed expectations. Based on these findings, a phased strategy is proposed: short-term optimization of interaction and knowledge support, mid-term development of human–AI collaboration and platform adaptability, and long-term establishment of governance and commercialization ecosystems. The study enriches virtual anchor research and highlights that enhancing core competencies is essential to transform novelty into enduring sales and brand equity, providing a practical pathway for e-commerce’s intelligent and sustainable transformation. Full article

The accelerated growth of e-commerce and ongoing urban expansion have intensified the challenges associated with last-mile delivery, making it a critical issue in sustainable urban logistics. Therefore, our paper presents a scoping review to systematically delineate the current state of research on smart and sustainable last-mile delivery systems. We explore both innovative technologies—such as artificial intelligence, autonomous vehicles, the Internet of Things, and digital twins—and human-centered dimensions, including urban design, policy development, and collaborative stakeholder engagement. Using the PRISMA-ScR-based methodology, 140 peer-reviewed articles (2015–2025) have been analyzed to highlight key trends, gaps, and prospective directions. The study underlines how the technologies of Industry 4.0 have improved visibility and operational efficiency, but holistic thinking that incorporates environmental, human, and policy factors remains undeveloped. Based on these findings, this article provides a conceptual framework for smart and sustainable last-mile delivery, focusing on the intersection of digital and simulation tools and human-centric governance to achieve optimized efficiency, environmental performance, and equity. This framework helps both academics and decision-makers to advance data-driven, resilient, and integrative city logistic ecosystems. Full article

Folk festivals and other intangible cultural heritage have received widespread attention, and their socio-cultural value can be used to promote tourism, strengthen local identity, and build city brands. However, it remains unclear how these intangible cultural heritage festivals transform their multi-dimensional and multi-configuration material characteristics into economic benefits and image enhancement. This study proposes a practical decision-making framework aimed at understanding how different festival design and governance strategies can work synergistically under different cultural conditions. Based primarily on a literature review and expert questionnaire survey, this study identified six stable materialized practice modules: productization, spatialization, experientialization, digitalization, branding/communication, and co-creation governance. At the same time, this framework also incorporates two other conditional intervention properties: classicism and novelty. The interactions between these modules shape people’s understanding of intangible cultural heritage festivals. Subsequently, this study used a multimodal national dataset that included official statistics, industry reports, e-commerce and social media data, questionnaires, and expert ratings to construct module scores and cultural attributes for 167 festival case studies. Through rough set analysis (RSA), this study simplifies the attributes and extracts clear “if-then” rules, establishing a configurational causal relationship between module configuration and classic/novel conditions to form high economic benefits and enhance local image. The findings of this study reveal a robust core built around spatialization, digitalization, and co-creative governance, with brand promotion/communication yielding benefits depending on the specific context. This further confirms that classicism reinforces the legitimacy and effectiveness of rituals/spaces and governance pathways, while novelty amplifies the impact of digitalization and immersive interaction. In summary, this study constructs an integrated and easy-to-understand process that links indicators, weights, and rules, and provides operational support for screening schemes and resource allocation in festival event combinations and venue brand governance. Full article

With the rapid advancement of virtual human technology and live-streaming e-commerce, virtual anchors have increasingly become key interactive entities in the digital economy. However, emerging issues such as fake reviews, abnormal tipping, and illegal transactions pose significant threats to platform financial security and user privacy. To address these challenges, a multimodal emotion–finance fusion security recognition framework (MSF-Net) is proposed, which integrates visual, audio, textual, and financial transaction signals to achieve cross-modal feature alignment and multi-signal risk modeling. The framework consists of three core modules: the multimodal alignment transformer (MAT), the fake review detection (FRD) module, and the multi-signal fusion decision module (MSFDM), enabling deep integration of semantic consistency modeling and emotion–behavior collaborative recognition. Experimental results demonstrate that MSF-Net achieves superior performance in virtual live-streaming financial security detection, reaching a precision of 0.932, a recall of 0.924, an F1-score of 0.928, an accuracy of 0.931, and an area under curve (AUC) of 0.956, while maintaining a real-time inference speed of 60.7 FPS, indicating outstanding precision and responsiveness. The ablation experiments further verify the necessity of each module, as the removal of any component leads to an F1-score decrease exceeding 4%, confirming the structural validity of the model’s hierarchical fusion design. In addition, a lightweight version of MSF-Net was developed through parameter distillation and quantization pruning techniques, achieving real-time deployment on mobile devices with an average latency of only 19.4 milliseconds while maintaining an F1-score of 0.923 and an AUC of 0.947. The results indicate that MSF-Net exhibits both innovation and practicality in multimodal deep fusion and security risk recognition, offering a scalable solution for intelligent risk control in data-driven artificial intelligence applications across financial and virtual interaction domains. Full article

A significant challenge in e-commerce is the inability of consumers to physically inspect products, forcing them to rely on perceived quality derived from other consumers’ experiences. However, gaps remain in understanding which dimensions of perceived quality are most frequently mentioned and influential for customer satisfaction, particularly in emerging markets like Indonesia. This study investigates these gaps by identifying key perceived quality aspects and examining their impact on satisfaction, with a specific focus on the moderating role of logistics delivery services. Using a large language model (LLM), specifically Google’s Gemma 2, we performed aspect-based sentiment analysis on 5000 smartphone reviews from Indonesian e-commerce. Logistic regression models incorporating interaction variables were employed to evaluate the relationships. The results identify the most frequently mentioned aspects of perceived quality: Logistics delivery services, Functionality, Originality, Responsiveness, and Packaging. While Logistics delivery services was the most mentioned aspect, Packaging had the most significant direct influence on satisfaction. Notably, Logistics delivery services also play a significant moderating role, enhancing the positive effect of other perceived quality aspects on satisfaction. These findings suggest that Logistics delivery services contribute directly to satisfaction and amplify other aspects, resulting in greater customer satisfaction. The study contributes to the literature by demonstrating LLM-driven aspect-based sentiment analysis methods and expanding the concept of perceived quality to include service aspects, thus promoting a more complete consideration of perceived quality in e-commerce. Full article

Effective website usability assessment is crucial for improving user experience, driving customer satisfaction, and ensuring business success, particularly in the competitive e-commerce sector. Traditional methods, such as expert reviews and user testing, are resource-intensive and often fail to fully capture the complex interplay between a site’s aesthetic design and its technical performance. This paper introduces an end-to-end multimodal deep learning framework that automates the usability assessment of fashion e-commerce websites. The framework fuses structured numerical indicators (e.g., load time, mobile compatibility) with high-level visual features extracted from full-page screenshots. The proposed framework employs a comprehensive set of visual backbones—including modern architectures such as ConvNeXt and Vision Transformers (ViT, Swin) alongside established CNNs—and systematically evaluates three fusion strategies: early fusion, late fusion, and a state-of-the-art cross-modal fusion strategy that enables deep, bidirectional interactions between modalities. Extensive experiments demonstrate that the cross-modal fusion approach, particularly when paired with a ConvNeXt backbone, achieves superior performance with a 0.92 accuracy and 0.89 F1-score, outperforming both unimodal and simpler fusion baselines. Model interpretability is provided through SHAP and LIME, confirming that the predictions align with established usability principles and generate actionable insights. Although validated on fashion e-commerce sites, the framework is highly adaptable to other domains—such as e-learning and e-government—via domain-specific data and light fine-tuning. It provides a robust, explainable benchmark for data-driven, multimodal website usability assessment and paves the way for more intelligent, automated user-experience optimization. Full article

Population aging and rising rehabilitation demands highlight the need for advanced assistive devices to improve mobility in individuals with motor impairments. Existing back-support lifting nursing beds often lack adequate human–machine adaptability, safety, and emotional consideration. This study presents a human-centered, data-driven optimization pipeline that integrates behavior–emotion dual recognition, simulation verification, and parameter optimization with user demand mining, biomechanical simulation, and sustainable practices. The design utilizes GreenAI, focusing on low-power algorithms and eco-friendly materials, ensuring energy-efficient AI models and reducing the environmental footprint. A dual-channel data fusion method was developed, combining movement parameters from sit-to-lie transitions with emotional needs extracted from e-commerce reviews using the Term Frequency-Inverse Document Frequency (TF-IDF) and Latent Dirichlet Allocation (LDA) models. The fuzzy Kano model prioritized design objectives, identifying multi-position adjustment, joint protection, armrest optimization, and interaction comfort as key targets. An AnyBody-based human–device model quantified muscle (erector spinae, rectus abdominis, trapezius) and hip joint loads during posture changes. Simulations verified the design’s ability to improve load distribution, reduce joint stress, and enhance comfort. The optimized nursing bed demonstrated improved adaptability across user profiles while maintaining functional reliability. This framework offers a scalable paradigm for intelligent rehabilitation equipment design, with potential extension toward AI-driven adaptive control and clinical validation. This sustainable methodology ensures that the device not only meets rehabilitation goals but also contributes to a more environmentally responsible healthcare solution, aligning with global sustainability efforts. Full article