Electronic Commerce Research and Applications最新文献

{"title":"Talking or typing? The differential impact of voice and text communication on customer churn","authors":"Ziying Mi,&nbsp;Yuxiao Chen,&nbsp;Xiaoling Lu,&nbsp;Feifei Wang","doi":"10.1016/j.elerap.2026.101643","DOIUrl":"10.1016/j.elerap.2026.101643","url":null,"abstract":"<div><div>As market competition intensifies, customer churn has become a critical concern for companies. A key driver of retention is the quality of customer service interactions across different communication modes. Prior research largely focuses on perceptual or experiential outcomes of communication mode differences, but does not examine how the four dimensions of Media Richness Theory (MRT) manifest as objective behavioral features in real-world customer service interactions. It also remains unclear whether these dimensions exert uniform or differential effects on customer renewal, and whether a richer communication medium uniformly benefits customer retention. This study addresses two research questions. RQ1 examines how voice calls and text chats differ across the four dimensions of media richness (i.e., immediate feedback, natural language use, multiple cues, and personal focus) in actual customer service interactions; RQ2 examines how these differences influence customer churn. Using large-scale interaction data from a Chinese online education platform, we develop an objective framework that operationalizes each dimension using behavioral indicators from interaction logs. Results show that voice calls are richer than text chats across all MRT dimensions. However, their effects on churn differ by dimension: feedback immediacy and personal focus reduce churn, natural language use shows no significant effect, and multiple cues exhibit a more nuanced pattern. These findings extend MRT to customer relationship management and offer actionable insights for designing more effective service interactions.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"79 ","pages":"Article 101643"},"PeriodicalIF":6.8,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149140660","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":"How can criterion-rich e-marketplaces recommend without CF? An explainable hybrid RL and entropy-weighted TOPSIS framework","authors":"Özkan Canay,&nbsp;Cem Özkurt","doi":"10.1016/j.elerap.2026.101633","DOIUrl":"10.1016/j.elerap.2026.101633","url":null,"abstract":"<div><div>E-marketplaces may expose rich product criteria while collaborative histories remain sparse, restricted, or unavailable, creating a need for auditable collaborative-filtering-free (CF-free) ranking. This paper proposes Hybrid RL-TOPSIS, which combines a regularized entropy-weighted TOPSIS structural ranker with a tabular one-step action-value learner through a fixed linear fusion. The target-blind primary evaluation uses 50 deterministic, stratified 400-item Amazon India catalogs, full-catalog action support, no target-membership reward, and a constructed target separating observable and profile-specific utility. Hybrid RL-TOPSIS attains F1@7 of 0.331, compared with 0.129 for RL-only and 0.259 for TOPSIS-only. The two prespecified, design-conditional paired gains are 0.202 over RL-only (catalog-resample 95% interval [0.182, 0.221], <span><math><mrow><mi>p</mi><mo>=</mo><mn>3.12</mn><mo>×</mo><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mo>−</mo><mn>25</mn></mrow></msup></mrow></math></span>) and 0.071 over TOPSIS-only ([0.045, 0.098], <span><math><mrow><mi>p</mi><mo>=</mo><mn>4.85</mn><mo>×</mo><mn>1</mn><msup><mrow><mn>0</mn></mrow><mrow><mo>−</mo><mn>6</mn></mrow></msup></mrow></math></span>); NDCG@7 shows the same ordering. At <span><math><mrow><mi>K</mi><mo>=</mo><mn>7</mn></mrow></math></span>, the Hybrid–TOPSIS gap corresponds to approximately 0.50 additional relevant item per seven-slot list and a 27.5% relative F1@7 gain. The previously reported 0.901/0.574/0.259 scores are exactly reproducible under the original target-assisted candidate and reward protocol, but are retained only as historical mechanism-reproduction evidence, not as the target-blind estimate. Future-blind sudden and gradual preference-drift checks preserve a Hybrid-over-RL advantage. Exact fixed-reference TOPSIS Shapley values and realized reward-path decomposition reconstruct all audited scores to numerical precision. On the tested laptop, cached 400-item fusion plus full sorting has median/P95/P99 latency of 53.5/117.6/<span><math><mrow><mn>150.2</mn><mspace></mspace><mi>μ</mi><mi>s</mi></mrow></math></span>. These results support a bounded criterion-rich CF-free niche, not universal or production-log superiority.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"79 ","pages":"Article 101633"},"PeriodicalIF":6.8,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148936238","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":"Information sharing strategy in the fresh produce supply chain considering the 3PL’s fresh-keeping effort","authors":"Fan Jiang,&nbsp;Kun Bai,&nbsp;Jianghua Wu","doi":"10.1016/j.elerap.2026.101636","DOIUrl":"10.1016/j.elerap.2026.101636","url":null,"abstract":"<div><div>This paper investigates the horizontal and vertical information sharing strategies and the fresh-keeping effort by constructing a fresh produce supply chain comprising one supplier, one third-party logistics provider (3PL), and two retailers engaged in a Cournot game. We examine how freshness elasticity impacts information sharing strategies and the supply chain’s profit. The results reveal that retailers’ optimal information sharing strategy depends on freshness elasticity. Interestingly, retailers never share demand information directly with the supplier, focusing all their direct vertical information sharing exclusively on the 3PL, while the supplier and the 3PL endogenously transmit the information they receive to each other. Specifically, when freshness elasticity is low, retailers do not share information horizontally, and neither retailer shares demand information with the 3PL. When freshness elasticity is intermediate, retailers share no information horizontally, but both retailers share demand information with the 3PL. When freshness elasticity is high, retailers share information horizontally, and both retailers share demand information with the 3PL. Moreover, when retailers share information with the 3PL and the information is endogenously transmitted to the supplier, the supply chain’s profit can be maximized. We also show that horizontal information sharing strategies and freshness elasticity jointly affect the optimal fresh-keeping effort and supply chain coordination.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"79 ","pages":"Article 101636"},"PeriodicalIF":6.8,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148936502","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":"Optimal ship-from-store strategies in live streaming e-commerce","authors":"Zhengqiang Li,&nbsp;Qingyun Xu,&nbsp;Yongbo Xiao,&nbsp;Xinyue Cai","doi":"10.1016/j.elerap.2026.101642","DOIUrl":"10.1016/j.elerap.2026.101642","url":null,"abstract":"<div><div>To enhance the shopping experience of online consumers, retailers have implemented a variety of strategies. These include offering Ship-from-Store (SFS) services to reduce delivery waiting times and launching the live-streaming channel to enrich the entertainment aspect of shopping. In practice, however, most retailers adopt one of two approaches: either providing SFS exclusively for the traditional online channel (OS strategy) or not offering SFS services at all (NS strategy). Only a small number of retailers extend SFS services to both traditional online and live-streaming channels simultaneously (DS strategy). The underlying reasons for this variation in SFS strategy adoption across different channels remain unclear. This paper develops an analytical model to investigate the optimal SFS strategies and corresponding pricing decisions for retailers operating with and without offline sales channels. Our findings reveal that the optimal SFS strategy is shaped by several key factors, including the unit cost of implementing SFS, the proportion of time-sensitive consumers, and the hassle cost incurred by consumers when purchasing via live-streaming. These insights offer valuable guidance for retailers seeking to optimize their SFS services in a multi-channel environment, particularly in the context of live-streaming commerce.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"79 ","pages":"Article 101642"},"PeriodicalIF":6.8,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149140661","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":"Understanding users’ information disclosure in AI-driven medical consultations: a mixed-method study using SEM and fsQCA within the push-pull-mooring framework","authors":"Qiongwei Ye,&nbsp;Luoyan Meng,&nbsp;Yumei Luo,&nbsp;Liyue Lu,&nbsp;Shuo Liu,&nbsp;Zhilin Pang","doi":"10.1016/j.elerap.2026.101641","DOIUrl":"10.1016/j.elerap.2026.101641","url":null,"abstract":"<div><div>As a pivotal scenario of digital health services, AI-driven medical consultation (AIMC) relies heavily on users’ information disclosure to ensure diagnostic accuracy—yet his decision represents a complex “algorithmic threshold” for patients. This study anchors on the push–pull-mooring (PPM) framework to conceptualize disclosure not as a simple trade-off, but as a behavioral migration from a status quo of defensive non-disclosure to active sharing. Specifically, we identify perceived privacy invasion (PPI) and perceived information control (PIC) as push factors; perceived convenience (PC) and perceived information quality (PIQ) as pull factors; and previous privacy experience (PPE) and personal technological innovativeness (PTI) as mooring factors acting as direct psychological anchors. Drawing on 424 valid responses from a nationwide survey in China, we employed a hybrid approach—structural equation modeling (SEM) for linear effect testing and fuzzy-set qualitative comparative analysis (fsQCA) for configurational exploration. SEM results reveal that PPI, PIC, PIQ, and PTI exert significant direct effects on disclosure intention, while PC and PPE do not, highlighting a “Rationality Filter” where patients prioritize the “precision dividend” over mere functional ease. Complementarily, fsQCA identifies three equifinal configurations leading to high disclosure: system-optimized, low-barrier, and knowledge-empowered trajectories. These findings reveal a substitution effect between individual curiosity and systemic control, suggesting that information quality is the non-negotiable anchor of the AIMC value proposition. This study advances theoretical understanding of privacy decision-making in digital health and offers strategic insights for designing user-centric, trust-mediated service ecosystems.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"79 ","pages":"Article 101641"},"PeriodicalIF":6.8,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149140613","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":"Activating patient continuity of care through nonverbal signal: The role of physicians’ voice in online depression treatment","authors":"Wenjie Pan,&nbsp;Jianqiang Wang,&nbsp;Xiaokang Wang,&nbsp;Zhiya Zuo,&nbsp;Fei Xiao","doi":"10.1016/j.elerap.2026.101645","DOIUrl":"10.1016/j.elerap.2026.101645","url":null,"abstract":"<div><div>Depression treatment requires continuity of care (CoC), yet patient adherence remains challenging. As a critical form of healthcare e-commerce, online medical consultations (OMCs) have emerged as a promising avenue to facilitate CoC for patients with depression. However, the predominantly text-based nature of OMCs may constrain the transmission of nonverbal cues that are essential for mental health care. This study examines how voice messaging in OMCs affects CoC among patients with depression. Guided by signaling theory, we collected secondary data from a Chinese OMC platform and employed fixed-effects models to examine the effect of voice message count (signal intensity) on CoC and the moderating effects of physicians’ voice characteristics (signal clarity), including pitch, pause, and speed. We find that more voice messages are associated with higher CoC, consistent with voice functioning as a signal of physician effort and availability. Signal clarity moderates this effect: higher vocal pitch, fewer pauses, and slower speed each amplify the impact of voice count on CoC, indicating that signal intensity translates into patient adherence more effectively when signal clarity is high. These results extend signaling theory by identifying the interplay between signal intensity and signal clarity in shaping receiver responses in the OMC context and offer insights for leveraging voice to strengthen patient CoC in digital mental health care and promote platform sustainability in digital health e-commerce.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"79 ","pages":"Article 101645"},"PeriodicalIF":6.8,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149140658","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":"Decision-making and coordination of live commerce supply chain considering dynamic viewership: The role of live streamer","authors":"Xiaran Zhang,&nbsp;Zheng Zhang,&nbsp;Xiaoyan Sun,&nbsp;Xin Cui","doi":"10.1016/j.elerap.2026.101616","DOIUrl":"10.1016/j.elerap.2026.101616","url":null,"abstract":"<div><div>Live commerce continuously acquires live viewers and effectively converts them into purchases. Live streamers effectively facilitate this process. This study innovates by employing differential equations to model the dynamic fluctuations of live streaming viewership, and analyze how the dynamics of live streaming viewership influence the efforts of the brand owner and the performance of the live streamer, and propose a “reciprocal subsidy” contract to achieve system coordination in live commerce. The results show that as live streamers’ core competencies (core fan cohort and viewer-to-buyer conversion rate) are strengthened, the operational efficiency of live commerce improves. Centralized decision-making significantly outperforms decentralized decision-making in optimizing system performance. While the “reciprocal subsidy” contract enables coordination, its viability depends on constraining the commission ratio within a specific range acceptable to both the live streamer and the brand owner. Furthermore, the applicability of the contract diminishes as the core fan cohort expands. This study provides actionable insights for managing live commerce operations under dynamic live streaming viewership.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"78 ","pages":"Article 101616"},"PeriodicalIF":6.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148566831","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":"Detecting fraudulent entities in e-commerce transaction networks via cross-view contrastive graph learning","authors":"Wenyao Zhu,&nbsp;Xing Chen","doi":"10.1016/j.elerap.2026.101615","DOIUrl":"10.1016/j.elerap.2026.101615","url":null,"abstract":"<div><div>The rapid growth of e-commerce has been accompanied by increasingly sophisticated fraud, costing online platforms over $48 billion annually. Fraudulent entities – fake review accounts, payment fraud rings, and money laundering operations – often exploit the gap between carefully constructed profile attributes and their actual transaction behavior. Detecting such fraud in large-scale online transaction networks is challenging due to the scarcity of labeled fraud cases and the complex, multi-faceted nature of fraudulent behavior. We propose CSGAD, a semi-supervised graph-based fraud detection framework that models the inconsistency between two complementary views of entity behavior in e-commerce networks: (1) the <em>structural view</em>, encoding transaction connectivity patterns such as who transacts with whom and through which intermediaries, and (2) the <em>semantic view</em>, encoding account profile attributes and transaction characteristics. Our key insight is that legitimate e-commerce participants exhibit consistency between their network behavior and profile attributes, whereas fraudulent entities show characteristic cross-view divergence. CSGAD employs dual specialized encoders – a transaction pattern encoder based on Graph Convolutional Networks and an account profile encoder with controlled Laplacian smoothing – linked through an InfoNCE contrastive objective that maximizes cross-view mutual information. Extensive experiments on eight real-world datasets, including Amazon product review networks (13K＋ entities; collected 2014–2015), the T-Finance online payment platform (39K accounts; collected 2019–2020), the Elliptic Bitcoin transaction network (203K transactions; January–September 2019), and the DGraph digital commerce network (3.7M users; collected 2020–2021), demonstrate state-of-the-art fraud detection. Notably, CSGAD achieves 3.9% and 3.6% AUROC improvements on T-Finance and Elliptic over the strongest baseline (SpaceGNN), while requiring only 1%–5% labeled fraud cases. A detailed case study on Amazon review networks confirms that CSGAD identifies coordinated fake review clusters by detecting profile-behavior inconsistencies that conventional methods miss. We further discuss deployment considerations, scalability, and practical implications for e-commerce platform operators.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"78 ","pages":"Article 101615"},"PeriodicalIF":6.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148566909","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":"Streaming platforms vs. theaters: Navigating the strategic choices of film distributors in channel selection","authors":"Chi Zhou,&nbsp;Shujian Liu,&nbsp;Dianyao Kang,&nbsp;Zongxi Li,&nbsp;Yongjun Liu","doi":"10.1016/j.elerap.2026.101623","DOIUrl":"10.1016/j.elerap.2026.101623","url":null,"abstract":"<div><div>The increasing prevalence of streaming platforms is transforming the film distribution landscape, compelling distributors to reassess their channel strategies and consumers to modify their viewing habits. This transformation has garnered considerable attention in both the film industry and academia, especially concerning the determination of optimal release channels for films. Therefore, this study explores the strategic considerations of film distributors through the development of a game-theoretic model that integrates the effects of market expansion, channel differentiation, and diminishing consumer utility on decision-making processes. This paper examines the profitability and social welfare of the film distributor, streaming platform, and theater under different strategies. Through a comprehensive analysis, this study aims to reveal the critical factors that influence the profits and social welfare of all parties under different strategies Our findings suggest that under specific conditions, prioritizing the theatrical release of films or releasing them exclusively on streaming platforms can lead to higher profitability for distributors. Additionally, we investigate the effects of delayed release times and channel differences on the equilibrium strategies of streaming platforms and theaters. Our analysis demonstrates that the hassle cost coefficient significantly influences the profitability of both streaming platforms and theaters across different strategies. Ultimately, we observe that prices and demand in both streaming and theatrical channels are affected by consumer utility loss due to delayed purchases. This study establishes a comprehensive theoretical framework to guide strategic decision-making for film distributors, streaming platforms, and theaters. Additionally, it offers valuable insights into the development of the film industry, serving as an essential reference for further research and strategic planning.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"78 ","pages":"Article 101623"},"PeriodicalIF":6.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148566791","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":"Competing or serving? platform role frames and third-party sellers’ pricing responses","authors":"Fengyuan Liu,&nbsp;Menghan Duan,&nbsp;Yongmin Zhu,&nbsp;Cheng Zhang,&nbsp;Jie Liu","doi":"10.1016/j.elerap.2026.101621","DOIUrl":"10.1016/j.elerap.2026.101621","url":null,"abstract":"<div><div>Direct price competition between online platforms and third-party sellers has intensified as platforms introduce similar first-party products. While prior research focuses on price-cutting dynamics, we examine a less explored scenario: how sellers respond when platform owners raise their prices. Using large-scale transaction data from a leading pharmaceutical e-commerce platform in China, we analyze third-party sellers’ pricing responses to platform-initiated price increases. We argue that prior interactions create rigid role frames through which sellers interpret the platform owner, thereby shaping their pricing responses. The empirical results show that sellers with greater reliance on platform logistics are more likely to follow price increases, whereas sellers with higher product overlap are less likely to align upward. Seller size further moderates these effects, with larger sellers exhibiting weaker pricing responses. This study contributes to the literature on platform governance and strategic pricing while offering practical implications for pricing management in hybrid platform ecosystems.</div></div>","PeriodicalId":50541,"journal":{"name":"Electronic Commerce Research and Applications","volume":"78 ","pages":"Article 101621"},"PeriodicalIF":6.8,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148566789","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}