Electronic Commerce Research and Applications最新文献

{"title":"Detecting fake review intentions in the review context: A multimodal deep learning approach","authors":"Jingrui Hou ,&nbsp;Zhihang Tan ,&nbsp;Shitou Zhang ,&nbsp;Qibiao Hu ,&nbsp;Ping Wang","doi":"10.1016/j.elerap.2025.101485","DOIUrl":"10.1016/j.elerap.2025.101485","url":null,"abstract":"<div><div>The proliferation of fake reviews on the internet has had significant repercussions for both consumers and businesses. However, existing research predominantly employs a binary classification approach to ascertain review authenticity, often neglecting the rich multimodal context information and nuanced intentions embedded within them. To bridge this gap, our study introduces a novel task, Fake Review Intention Detection in Review Context (FRIDRC), which aims to detect fake review intentions by leveraging both textual and visual information, and constructs a dataset comprising both manually and AI-generated fake reviews. Additionally, we develop a predictive framework encompassing modules for multimodal representation and modality fusion. These modules, while independent, are synergistic and effectively tackle the challenge of discerning fake review intentions. Our framework demonstrates outstanding performance, achieving an average F1 score exceeding 0.97 and a Macro F1 score surpassing 0.96 in this task and outperforming advanced pre-trained models. This research not only presents an effective methodology for accurately identifying and addressing fake review intentions but also underscores the efficacy of leveraging multimodal review context information in fake review detection. The dataset and code implementation are publicly available for further research.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"70 ","pages":"Article 101485"},"PeriodicalIF":5.9,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143376710","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DP-SGD-global-adapt-V2-S: Triad improvements of privacy, accuracy and fairness via step decay noise multiplier and step decay upper clipping threshold","authors":"Sai Venkatesh Chilukoti ,&nbsp;Md Imran Hossen ,&nbsp;Liqun Shan ,&nbsp;Vijay Srinivas Tida ,&nbsp;Mahathir Mohammad Bappy ,&nbsp;Wenmeng Tian ,&nbsp;Xiali Hei","doi":"10.1016/j.elerap.2025.101476","DOIUrl":"10.1016/j.elerap.2025.101476","url":null,"abstract":"<div><div>Differentially Private Stochastic Gradient Descent (DP-SGD) has become a widely used technique for safeguarding sensitive information in deep learning applications. Unfortunately, DP-SGD’s per-sample gradient clipping and uniform noise addition during training can significantly degrade model utility and fairness. We observe that the latest DP-SGD-Global-Adapt’s average gradient norm is the same throughout the training. Even when it is integrated with the existing linear decay noise multiplier, it has little or no advantage. Moreover, we notice that its upper clipping threshold increases exponentially towards the end of training, potentially impacting the model’s convergence. Other algorithms, DP-PSAC, Auto-S, DP-SGD-Global, and DP-F, have utility and fairness that are similar to or worse than DP-SGD, as demonstrated in experiments. To overcome these problems and improve utility and fairness, we developed the DP-SGD-Global-Adapt-V2-S. It has a step-decay noise multiplier and an upper clipping threshold that is also decayed step-wise. DP-SGD-Global-Adapt-V2-S with a privacy budget (<span><math><mi>ϵ</mi></math></span>) of 1 improves accuracy by 0.9795%, 0.6786%, and 4.0130% in MNIST, CIFAR10, and CIFAR100, respectively. It also reduces the privacy cost gap (<span><math><mi>π</mi></math></span>) by 89.8332% and 60.5541% in unbalanced MNIST and Thinwall datasets, respectively. Finally, we develop mathematical expressions to compute the privacy budget using truncated concentrated differential privacy (tCDP) for DP-SGD-Global-Adapt-V2-T and DP-SGD-Global-Adapt-V2-S.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"70 ","pages":"Article 101476"},"PeriodicalIF":5.9,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143332282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The merits of the pre-owned strategy: Price competition in multichannel digital goods distribution","authors":"Yu Bai ,&nbsp;Xiuwu Liao ,&nbsp;Yang Liu","doi":"10.1016/j.elerap.2025.101486","DOIUrl":"10.1016/j.elerap.2025.101486","url":null,"abstract":"<div><div>The provision of pre-owned physical version products by traditional brick-in-store platforms (such as GameStop) was once highly profitable but has rapidly declined with the rise of digital content. As publisher firms sell digital versions directly, they now compete with platforms that sell both new and pre-owned physical versions. This paper conducts an economic analysis on a traditional brick-in-store platform’s pre-owned strategies and a publishing firm’s competition strategies. We examine two key factors that significantly influence competition between pre-owned, digital, and physical versions: distribution contracts (wholesale and agency) and countermeasures (such as exclusive digital content). Generally, pre-owned strategies hurt both the firm and the platform. However, our analysis reveals that pre-owned products benefit platforms under certain condition. From the platform’s perspective, when beneficial, wholesale contract is preferable to agency contracts. Our findings propose an “enticement yet limitation” principle for platform managers when developing pre-owned strategies. From the firm’s perspective, offering extra content as countermeasure is beneficial, helping to mitigate the harmful effects of pre-owned versions and enhancing competitive effectiveness. Our results reveal the operational mechanism behind such content monetization strategy and emphasize the strategic importance of countermeasures when responding to platforms’ pre-owned strategies. Finally, we check the robustness in extensions with considering differences in costs, reservation values and various pre-owned versions. While the pre-owned product has sparked controversy and is generally believed to be detrimental to the industry, our results suggest that content monetization strategy in countermeasure may sometimes be economically beneficial to both firms and platforms. Our findings shed light on the key factors publisher firms and platforms should consider when deairing with pre-owned strategies in the digital content industries.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"70 ","pages":"Article 101486"},"PeriodicalIF":5.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143387862","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unraveling the anchoring effect of seller’s show on buyer’s show to enhance review helpfulness prediction: A multi-granularity attention network model with multimodal information","authors":"Feifei Wang ,&nbsp;Zeyue Zhang ,&nbsp;Jie Song ,&nbsp;Yixia Yang ,&nbsp;Xiaoling Lu","doi":"10.1016/j.elerap.2025.101484","DOIUrl":"10.1016/j.elerap.2025.101484","url":null,"abstract":"<div><div>The prevalence of multimodal data has become commonplace in e-commerce platforms. Both seller showcases (i.e., the seller’s show) and user-generated content (i.e., the buyer’s show) now incorporate diverse modalities, combining both textual and visual elements. In this work, we aim to unraveling the impact of seller’s show on buyer’s show through the anchoring effect. We narrow our research on the specific problem of review helpfulness prediction and further explore whether the anchoring effect can improve the prediction accuracy of review helpfulness. In pursuit of this goal, we develop the <em>Multi-granularity Attention Network Model based on Anchoring Effect</em> (MAN-AE). This model first extracts the multi-granularity features in both seller’s show and buyer’s show and then accounts for the anchoring effect through a cross-source transformer. Through extensive experiments on an Amazon dataset, we demonstrate the anchoring effect of seller’s show on buyer’s show in enhancing the review helpfulness prediction performance. In comparison with other state-of-the-art models, our model demonstrates significantly superior prediction performance.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"70 ","pages":"Article 101484"},"PeriodicalIF":5.9,"publicationDate":"2025-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143332281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A tale of stores and screens: Unveiling consumer behaviour in omnichannel retailing through the lens of behavioural reasoning","authors":"Neha Sharma ,&nbsp;Emiliano Acquila-Natale ,&nbsp;Nirankush Dutta ,&nbsp;Ángel Hernández-García","doi":"10.1016/j.elerap.2025.101480","DOIUrl":"10.1016/j.elerap.2025.101480","url":null,"abstract":"<div><div>This research examines the mechanisms that foster or deter consumers’ adoption of digital storefronts that traditional brick-and-mortar retailers integrate for omnichannel operations in emerging markets, through the lens of Behavioural Reasoning Theory. Using a mixed-methods approach, the study first identifies specific reasons for and against shopping on the digital platforms of brick-and-mortar retailers through qualitative interviews with retail experts. These findings are then tested quantitatively with a survey of 1392 Indian omnichannel consumers, analysed using partial least squares structural equation modelling and importance-performance matrix analysis.</div><div>The results reveal that perceived product quality and shopping flexibility are the main drivers for adoption, while the attractiveness of alternatives and concerns over delivery timeliness are key deterrents. Additionally, the analysis considers the influence of perceived compatibility and the moderating effect of product type, with electronics and clothing as representative of search and experience goods, respectively. The analysis finds that perceived compatibility exerts a moderate to low effect on consumers’ reasoning and their attitudes, and that deterrents have a stronger negative impact for experience goods in shaping consumers’ attitude towards adopting the digital storefronts of brick-and-mortar retailers. The findings advance Behavioural Reasoning Theory in retail contexts, providing actionable insights for brick-and-mortar retailers to enhance their omnichannel strategies by addressing consumer-specific motivations and barriers.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"70 ","pages":"Article 101480"},"PeriodicalIF":5.9,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143140796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The thrust and drag forces affecting norm violation in live streaming eCommerce","authors":"Yu-Ting Chang-Chien ,&nbsp;Kuang-Ting Cheng ,&nbsp;Jack Shih-Chieh Hsu ,&nbsp;Hsieh-Hong Huang","doi":"10.1016/j.elerap.2025.101478","DOIUrl":"10.1016/j.elerap.2025.101478","url":null,"abstract":"<div><div>Norm violation, in the form of buyers failing to confirm and pay for orders lodged, increases costs for sellers in the live streaming online transaction context in which most sellers are small brands or customer-to-customer sellers. Attempting to violate norm causes cognitive dissonance and consumers may alter their cognition or avoid behavior to reduce negative feeling caused by inner conflict. The former may be done by rationalizing the focal behavior with neutralization technique (thrust force) and the latter is to avoid behavior through maintaining strong cognition formed by commitment (drag force). In addition, we proposed potential measures for the inhibition and enhancement of such thrust and drag forces, respectively. Furthermore, we explored whether the magnitude of these forces is contingent on buyers’ gender. After collecting survey data from 331 buyers, we found that neutralization strengthens, and commitment weakens norm-violating intention. Furthermore, the impact of neutralization was greater for male buyers, and the impact of normative commitment was greater for female buyers. In addition, the four proposed measures (including descriptive norm, community participation, policy communication, and sanction policy) can effectively reduce neutralization and increase commitment.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"70 ","pages":"Article 101478"},"PeriodicalIF":5.9,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143332280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multi-level sparse attentive fusion network integrating hard and soft information for firm-level loan default prediction","authors":"Jingling Ma ,&nbsp;Junqiao Gong ,&nbsp;Gang Wang ,&nbsp;Xuan Zhang","doi":"10.1016/j.elerap.2025.101479","DOIUrl":"10.1016/j.elerap.2025.101479","url":null,"abstract":"<div><div>Firm-level loan default prediction (FLDP) deserves much attention from both academic and industry. Even a small improvement in the accuracy of FLDP could lead to significant savings by reducing credit risk. While previous studies have utilized deep learning models for FLDP task, they failed to well handle the intra-type ambiguity and inter-type interaction simultaneously facing with combined hard and soft information, thus remaining an area of ongoing development. By this perspective, we seek to design a novel Multi-level Sparse Attention (MLSA) based deep learning fusion framework for FLDP, aiming to fully capture default signals conveyed from both hard and soft information. First, multiple types of information are extracted grounded in 5P theory and LAPP theory, ensuring the sufficiency and rationality of the features. Second, Sparse Attentive MLP (SA-MLP) and Sparse Attentive GRU (SA-GRU) module are proposed to handle the intra-type ambiguity embedded in hard and soft information separately. Further, the Attentive Fusion (AF) module including Differential Enhancive module and Common Selective module is proposed to explore inter-type interaction among hard and soft information. Last, we adopt the focal loss function to mitigate the adverse effects of imbalanced data. The proposed MLSA informs future FLDP research about how to fully exploit the value of hard and soft information by considering their intra-type ambiguity and inter-type interaction. Empirical evaluation of the MLSA on a real-world dataset demonstrates its outperformance of state-of-the-art benchmarks in the FLDP task. Our results also contribute to the growing literature on this topic by highlighting the roles of hard and soft information and improving interpretability.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"70 ","pages":"Article 101479"},"PeriodicalIF":5.9,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143140799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Revenue-based personalized product recommendation considering stochastic purchase probability","authors":"Chao Huang,&nbsp;Xi Zhang,&nbsp;Yifan Zhang,&nbsp;Qinghao Hu","doi":"10.1016/j.elerap.2025.101477","DOIUrl":"10.1016/j.elerap.2025.101477","url":null,"abstract":"<div><div>Recommender systems(RS) play a critical role in e-commerce platforms by providing personalized and relevant product suggestions to customers, thereby enhancing their shopping experience and increasing platform revenue. Existing RSs focus on improving accuracy or maximizing user purchase probability when generating recommendations. However, a sole emphasis on accuracy does not ensure the optimization of platform revenue, and recommendations that maximize user purchase probability can also fail to simulate the real purchase behavior of users, which shows strong uncertainty due to external factors. To address these issues, we propose a two-stage personalized product recommendation method based on stochastic purchase probability (PRSPP). In the first stage, we follow previous studies which prioritize user preferences during the recommendation process. A taxonomy-based approach is employed to estimate user preferences at the category level and select candidate products for each user. Subsequently, considering the impact of factors such as product price, sales and category similarity on user utility, a logistic regression model is employed to quantify user preferences for these candidate products and further estimate user purchase probability for them. In the second stage, we aim to optimize the recommendation from the perspective of platform operators, considering user purchases are subject to diverse external factors and exhibit strong uncertainty. We treat purchase probability as a random variable, and a stochastic optimization model with the objective of maximizing platform revenue is formulated. Furthermore, we apply the Sample Average Approximation (SAA) approach to solve the model. Finally, we conduct experiments on Amazon public dataset, and the results present advantages of PRSPP in improving both recommendation accuracy and platform revenue.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"70 ","pages":"Article 101477"},"PeriodicalIF":5.9,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143140798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Insurance policy and pricing decisions in online food delivery market with consumer ratings","authors":"Minjian Liu ,&nbsp;Shaofu Du ,&nbsp;Tengfei Nie ,&nbsp;Yangguang Zhu ,&nbsp;Qi Dong","doi":"10.1016/j.elerap.2024.101474","DOIUrl":"10.1016/j.elerap.2024.101474","url":null,"abstract":"<div><div>Recent advances in mobile technology and the rise of aggregator apps have led many consumers to purchase products through online food delivery platforms, and the effects of COVID-19 have accelerated the trend. Consumers often observe other consumers’ ratings before purchasing to reduce the product’s taste and the delivery service uncertainties. However, the information accuracy of the ratings can be affected by the seller’s marketing strategies, such as the pricing and delivery insurance policy. Considering consumers’ ratings, firstly, we develop two-period models to examine how consumer uncertainty and different delivery insurance policies impact the seller and consumers. We then determine the seller’s optimal insurance policy and pricing decision. We find that the probability of delayed delivery has a nonmonotonic effect on the product’s optimal price. Counterintuitively, our analysis shows that a Free Insurance (FI) policy allows the seller to benefit from a greater delayed delivery probability by influencing consumer ratings. We show that FI is not always beneficial for the seller and consumers; it may reduce the seller’s profit and the total consumer surplus simultaneously. Surprisingly, the seller can always charge a lower price yet earn more profit by adopting a Paid Insurance (PI) policy. Furthermore, when the seller strategically chooses its price and insurance policy (e.g., No Insurance, FI, or PI), it can achieve a win-win situation; and the seller should always support insurance, i.e., adopt FI or offer PI, even though neither FI nor PI dominates under all conditions.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"70 ","pages":"Article 101474"},"PeriodicalIF":5.9,"publicationDate":"2025-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143140797","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Collaboration or Encroachment? The Content Provider’s Strategic Content Provision Strategy","authors":"Siyu Du,&nbsp;Xu Wang","doi":"10.1016/j.elerap.2024.101470","DOIUrl":"10.1016/j.elerap.2024.101470","url":null,"abstract":"<div><div>With the rapid growth of the streaming industry, an increasing number of Super Content Providers (SCPs), such as Disney, Warner Bros., and NBC Universal, have launched their own streaming platforms, entering the downstream streaming market. However, some content providers, like Sony, still collaborate with Netflix. Therefore, this paper employs game theory to investigate the optimal content provision strategy for SCPs: collaboration or encroachment. We derive some interesting findings. First, contrary to intuition, we find that whether under the collaboration or encroachment strategy, improving the quality of a player’s own film library (the SCP or the incumbent platform) does not always result in higher profit for that player (first hurts then benefits the player). Second, for both the platform and the SCP, an increase in the scale of their film libraries does not automatically lead to higher profits. Third, when the revenue share is positively correlated with the scale of the film library, a small-scale library strengthens the motivation to encroach; conversely, when the library scale does not influence the revenue share, a large-scale library increases the motivation to encroach. Lastly, while the encroachment strategy can be a win–win situation for the incumbent platform and the SCP, it hurts consumer surplus. In contrast, the collaboration strategy can achieve a win–win–win situation under some conditions. Our study helps to explain the market practices and provides valuable guidelines for SCPs’ content provision strategies.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"69 ","pages":"Article 101470"},"PeriodicalIF":5.9,"publicationDate":"2025-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143133273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"管理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}