Hasil untuk "Commerce"

Ting-Peng Liang, Yi-Ting Ho, Yu-Wen Li et al.

T. Šercar

Qujiaheng Zhang, Guagnyue Xu, Fengjie Li

Modern e-commerce search is inherently multimodal: customers make purchase decisions by jointly considering product text and visual informations. However, most industrial retrieval and ranking systems primarily rely on textual information, underutilizing the rich visual signals available in product images. In this work, we study unified text-image fusion for two-tower retrieval models in the e-commerce domain. We demonstrate that domain-specific fine-tuning and two stage alignment between query with product text and image modalities are both crucial for effective multimodal retrieval. Building on these insights, we propose a noval modality fusion network to fuse image and text information and capture cross-modal complementary information. Experiments on large-scale e-commerce datasets validate the effectiveness of the proposed approach.

Luis Antonio Gutiérrez Guanilo, Mir Tafseer Nayeem, Cristian López et al.

Large Language Models (LLMs) have demonstrated exceptional versatility across diverse domains, yet their application in e-commerce remains underexplored due to a lack of domain-specific datasets. To address this gap, we introduce eC-Tab2Text, a novel dataset designed to capture the intricacies of e-commerce, including detailed product attributes and user-specific queries. Leveraging eC-Tab2Text, we focus on text generation from product tables, enabling LLMs to produce high-quality, attribute-specific product reviews from structured tabular data. Fine-tuned models were rigorously evaluated using standard Table2Text metrics, alongside correctness, faithfulness, and fluency assessments. Our results demonstrate substantial improvements in generating contextually accurate reviews, highlighting the transformative potential of tailored datasets and fine-tuning methodologies in optimizing e-commerce workflows. This work highlights the potential of LLMs in e-commerce workflows and the essential role of domain-specific datasets in tailoring them to industry-specific challenges.

Xiaolin Lin, Yibai Li, Xuequn Wang

G. Gregory, L. Ngo, Munib Karavdic

Abstract This study builds on resource based view (RBV) theory by examining the effects of e-commerce on exporting performance. Specifically, a framework is developed and tested to determine the e-commerce resources/capabilities–marketing efficiencies–performance relationship. To explore the impact of e-commerce on exporting, a two-stage methodological approach was employed. Results from 15 depth interviews with exporters were used to gain insight into types of e-commerce resources and capabilities and their impact on export marketing efficiencies and performance. Next, the framework was empirically tested using a sample of 340 exporters. The evidence shows that specialized e-commerce marketing capabilities directly increase a firm's degree of distribution and communication efficiency, which in turn leads to enhanced export venture market performance. Overall, the analyses provide support for the need to incorporate e-commerce constructs into existing RBV theory in export marketing. Theoretical and managerial contributions are discussed and directions for future research are offered.

S. Verkijika

Abstract As smartphone penetration continues to double in Sub-Saharan Africa, many businesses are looking into this channel for conducting their business activities. In Cameroon, all the top e-commerce giants have deployed smartphone applications to facilitate m-commerce activities. However, little is known about the factors that influence m-commerce adoption in the country. As such, this study had as objective to determine the key factors that influence consumer’s adoption of m-commerce applications in Cameroon. Using data from 372 respondents, a modified version of the extended unified theory of acceptance and use of technology (UTAUT2) was validated in the Cameroon context. The findings showed that social influence, facilitating conditions, hedonic motivations, perceived risk and perceived trust were significant predictors of the behavioural intention to adopt m-commerce applications. Also, the results showed that consumers who had a high intention to adopt m-commerce were more likely to recommend the technology to others. For researchers, the study depicts the relevance of extending existing technology acceptance models like the UTUAT2 with appropriate factors in different technological and geographical context. For practitioners, the study identifies customer-specific and environmental factors that m-commerce providers in Cameroon and other regions with similar characteristics could consider when designing and implementing strategies for attracting consumers to use their m-commerce applications.

Jun-Jie Hew, Lai-Ying Leong, G. Tan et al.

Abstract Drivers of social commerce usage has been the focus of scholars in recent years, but mobile social media users' resistance behavior towards mobile social commerce has been in the darkness and therefore worth torched lights on. With the data collected from mobile social media users who have no experience in mobile social commerce, Artificial Neural Network analysis was engaged to capture both linear and nonlinear relationships in a research model that consists of innovation barriers and privacy concern. Surprisingly, all resistances positively correlated with usage intention, except for image barrier, which appeared to be the most influencing resistance. Several explanations were offered for such outcomes. The possible coexistence of resistance behavior and usage intention resembles the fitting justification. Mobile social media users intend to embrace mobile social commerce; however, their intentions have been held up by their perceptions on innovation barriers and privacy concern. Based upon these outcomes, this study has reaffirmed the coexistence of resistances and usage intention, as well as the “privacy paradox” phenomenon. These discoveries are believed to have contributed to the existing literature. Practitioners are then advised to act accordingly to these findings, and several methods on catalyzing mobile social media users' adoption decision were suggested.

Kangming Xu, Huiming Zhou, Haotian Zheng et al.

With the rapid evolution of the Internet and the exponential proliferation of information, users encounter information overload and the conundrum of choice. Personalized recommendation systems play a pivotal role in alleviating this burden by aiding users in filtering and selecting information tailored to their preferences and requirements. Such systems not only enhance user experience and satisfaction but also furnish opportunities for businesses and platforms to augment user engagement, sales, and advertising efficacy.This paper undertakes a comparative analysis between the operational mechanisms of traditional e-commerce commodity classification systems and personalized recommendation systems. It delineates the significance and application of personalized recommendation systems across e-commerce, content information, and media domains. Furthermore, it delves into the challenges confronting personalized recommendation systems in e-commerce, including data privacy, algorithmic bias, scalability, and the cold start problem. Strategies to address these challenges are elucidated.Subsequently, the paper outlines a personalized recommendation system leveraging the BERT model and nearest neighbor algorithm, specifically tailored to address the exigencies of the eBay e-commerce platform. The efficacy of this recommendation system is substantiated through manual evaluation, and a practical application operational guide and structured output recommendation results are furnished to ensure the system's operability and scalability.

Bo Peng, Xinyi Ling, Ziru Chen et al.

With tremendous efforts on developing effective e-commerce models, conventional e-commerce models show limited success in generalist e-commerce modeling, and suffer from unsatisfactory performance on new users and new products - a typical out-of-domain generalization challenge. Meanwhile, large language models (LLMs) demonstrate outstanding performance in generalist modeling and out-of-domain generalizability in many fields. Toward fully unleashing their power for e-commerce, in this paper, we construct ECInstruct, the first open-sourced, large-scale, and high-quality benchmark instruction dataset for e-commerce. Leveraging ECInstruct, we develop eCeLLM, a series of e-commerce LLMs, by instruction-tuning general-purpose LLMs. Our comprehensive experiments and evaluation demonstrate that eCeLLM models substantially outperform baseline models, including the most advanced GPT-4, and the state-of-the-art task-specific models in in-domain evaluation. Moreover, eCeLLM exhibits excellent generalizability to out-of-domain settings, including unseen products and unseen instructions, highlighting its superiority as a generalist e-commerce model. Both the ECInstruct dataset and the eCeLLM models show great potential in empowering versatile and effective LLMs for e-commerce. ECInstruct and eCeLLM models are publicly accessible through https://ninglab.github.io/eCeLLM.

Michał Malinowski

This paper presents a study on the implementation of the author's Algorithm of Recommendation Sessions (ARS) in an operational e-commerce information system and analyses the basic parameters of the resulting recommendation system. It begins with a synthetic overview of recommendation systems, followed by a presentation of the proprietary ARS algorithm, which is based on recommendation sessions. A mathematical model of the recommendation session, constructed using graph and network theory, serves as the input for the ARS algorithm. This paper also explores graph structure representation methods and the implementation of a G graph (representing a set of recommendation sessions) in a relational database using the SQL standard. The ARS algorithm was implemented in a working e-commerce information system, leading to the development of a fully functional recommendation system adaptable to various e-commerce IT systems. The effectiveness of the algorithm is demonstrated by research on the recommendation system's parameters presented in the final section of the paper.

Jane Rossouw

Background: Life Orientation educators hold great responsibility for the well-being of their students, which can be supported through imparting sexuality education. However, the absence of formal training for this subject may have negative consequences in fulfilling professional duties. Objectives: This article intends to foreground how Life Orientation educators impart sexuality education to their students, exploring aspects of their personal attitudes and comfort in imparting education related to sexuality and queerness. Methods: This qualitative study consisted of five Life Orientation educators in the Gauteng Province to understand their approaches to impart sexuality education to their students and the influence of their personal upbringings. The research was thematically analysed through a systems theory framework. Results: The results emphasise how personal religious beliefs impact sexuality education’s delivery and educators’ discretion in implementing the curriculum. The ambiguity of the curriculum and diverse teaching backgrounds also contribute to avoidance of topics like sexuality education and queer identities. Moreover, the non-examinable nature of these topics, combined with subjective interpretations of age-appropriateness, further marginalise them. Conclusion: This article calls for awareness of the consequences of religious convictions and subjective perceptions of age-appropriateness of educators on the delivery of sexuality and queer education. Contribution: This study contributes by highlighting challenges faced by Life Orientation educators in creating inclusive environments when personal religious beliefs conflict with comprehensive sexuality education. It enhances understanding of areas for improvement in training and subject knowledge to ensure educators affirm diverse identities and impart sexuality education effectively.

Rogério Gesta Leal

O presente trabalho objetiva abordar os cenários e riscos inerentes à segurança pública e à inteligência artificial, assim como iniciativas para controlá-los. Para tanto, elegemos como objetivos específicos: (i) demarcar as relações entre sociedade do conhecimento e sociedade da vigilância; (ii) as reações institucionais e políticas à sociedade de vigilância; (iii) propor premissas viabilizadoras de politicas de segurança pública democráticas para o uso de novas tecnologias com o uso de IA no âmbito da segurança pública. Pretendemos utilizar neste trabalho o método dedutivo, testando nossas hipóteses com os fundamentos que passam a ser declinados. Utilizaremos para tanto técnica de pesquisa com documentação indireta, nomeadamente bibliográfica.

Liying Zhou, Weiquan Wang, Jingjun Xu et al.

Abstract Consumers abandon their online purchases at an e-commerce website partly due to the lack of information transparency of the website. We identify the antecedents of consumers’ perceived information transparency of an e-commerce website and its effects on consumers’ online purchase intention. We collected data through a scenario-based survey conducted in a laboratory setting. We found that (1) product transparency, vendor transparency, and transaction transparency significantly influence perceived information transparency; (2) perceived information transparency significantly increases consumers’ online purchase intention; and (3) perceived risk partially mediates the effects of perceived information transparency on purchase intention.

J. Oláh, Nicodemus Kitukutha, Hossam Haddad et al.

The Internet revolution has led to the advancement of online business all over the world. The environmental, social, and economic aspects are significant to the e-commerce sector, on both the retailer and consumer sides. It cannot be over-emphasized how important the sustainability of e-commerce in all three dimensions is. E-commerce will allow consumers to shop online easily, at any hour of the day, using secure payment systems; furthermore, trust in retailers’ websites is of paramount importance to consumers. This calls our attention to the gap in previous studies, and consequently, the purpose of this study is to fill the gap, to ensure sustainable e-commerce in three dimensions; environmental, social, and economic. The question and aim under investigation are: How to integrate three dimensions into e-commerce to ensure that sustainability is achieved now and for future generations, while thriving as an industry? Collaboration is required, and all stakeholders in the virtual market must take appropriate responsibility. The methodology adopted is a review of previous studies done on each individual dimension of sustainability, since no joint studies have been carried out and integrated into the same literature framework. Furthermore, a case study involving companies in Kenya and Jordan is used in order to collect empirical data. The findings of the study show that: First, integration is essential for the sustainability of e-commerce in its three dimensions; second, trade-offs must be taken in the various dimensions in order for companies to realize sustainable e-commerce. This will go in hand with the realization of the maximum benefits of integrating the three dimensions in e-commerce to make it more sustainable. In conclusion, by applying these aspects of sustainability in e-commerce, it is clear that everyone wins. This is achieved by improving and safeguarding the quality of life by protecting the environment, preserving natural resources, and maintaining and sustaining the economy. The implications of the study are that, in order to make e-commerce more sustainable, to make decisions and take action, social/environmental/economic aspects must be considered as a fundamental element, and must be treated as a group and not separately as in previous studies. In this way, we can realize greater benefits, not only in online business sustainability, but also in policy-making and environmental protection, while companies will create economic value as well as avoiding labor unrest.

Jingting Fan, Lixin Tang, Weiming Zhu et al.

Domestic trade costs reduce aggregate welfare and result in worse access to consumption goods in small and remote cities. As a new trade technology, e-commerce can increase inter-city trade and alleviate spatial consumption inequality because it (1) eliminates the fixed cost of market entry, and (2) reduces the effects of distance on trade costs. Using unique data from China's leading e-commerce platform, we provide evidence consistent with these two features: online trade is less hindered by distance relative to offline trade, and residents from smaller and more remote cities spend a larger fraction of their income online. We then build a multi-region general equilibrium model to quantify the impacts of e-commerce on domestic trade and welfare. We find that although it partially crowds out inter-city trade originally taking place offline, the emergence of e-commerce increases the aggregate domestic trade. The welfare gains from e-commerce are 1.6% on average, and are about 30% larger for cities in the smallest population and market potential quintiles.

Omer A Gibreel, Dhari A. AlOtaibi, J. Altmann

Lai-Ying Leong, N. Jaafar, Ainin Sulaiman

Yangning Li, Shirong Ma, Xiaobin Wang et al.

Recently, instruction-following Large Language Models (LLMs) , represented by ChatGPT, have exhibited exceptional performance in general Natural Language Processing (NLP) tasks. However, the unique characteristics of E-commerce data pose significant challenges to general LLMs. An LLM tailored specifically for E-commerce scenarios, possessing robust cross-dataset/task generalization capabilities, is a pressing necessity. To solve this issue, in this work, we proposed the first e-commerce instruction dataset EcomInstruct, with a total of 2.5 million instruction data. EcomInstruct scales up the data size and task diversity by constructing atomic tasks with E-commerce basic data types, such as product information, user reviews. Atomic tasks are defined as intermediate tasks implicitly involved in solving a final task, which we also call Chain-of-Task tasks. We developed EcomGPT with different parameter scales by training the backbone model BLOOMZ with the EcomInstruct. Benefiting from the fundamental semantic understanding capabilities acquired from the Chain-of-Task tasks, EcomGPT exhibits excellent zero-shot generalization capabilities. Extensive experiments and human evaluations demonstrate that EcomGPT outperforms ChatGPT in term of cross-dataset/task generalization on E-commerce tasks.

Yang Peng, Changzheng Liu, Wei Shen

Customer-centric marketing campaigns generate a large portion of e-commerce website traffic for Walmart. As the scale of customer data grows larger, expanding the marketing audience to reach more customers is becoming more critical for e-commerce companies to drive business growth and bring more value to customers. In this paper, we present a scalable and efficient system to expand targeted audience of marketing campaigns, which can handle hundreds of millions of customers. We use a deep learning based embedding model to represent customers and an approximate nearest neighbor search method to quickly find lookalike customers of interest. The model can deal with various business interests by constructing interpretable and meaningful customer similarity metrics. We conduct extensive experiments to demonstrate the great performance of our system and customer embedding model.