The impact of artificial intelligence recommendation on consumer behaviour in the e-commerce industr

「文智助手」全场景办公AI助理AI+OPC全流程实践方案(大赛作品) 阅读详情

Java Python The impact of artificial intelligence recommendation on consumer behaviour in the e-commerce industry

Abstract

In this study, the impact of AI shopping recommendation systems on internet users' purchasing experience and purchase intention is explored, focusing on four key aspects: entertainment, relevance, usability, and user privacy data security. Firstly, AI recommendation systems enhance the shopping experience by providing personalized recommendations, making the shopping process more enjoyable and engaging. Secondly, AI systems can accurately analyze user behavior. and needs, offering highly relevant product recommendations that significantly improve user satisfaction. Thirdly, modern AI recommendation systems are designed to be intuitive and user-friendly, enabling users to easily navigate and interact with the system, thereby enhancing the overall shopping experience. Finally, regarding user privacy and data security, AI systems employ techniques such as data anonymization and encryption to protect user information effectively, fostering greater trust in the system.

User data was gathered through questionnaires and subsequently analyzed using a multiple linear regression model with SPSS 2.0. This model quantified the relationships between the identified key aspects and users' purchasing experience and purchase intention. The analysis demonstrates that by enhancing entertainment, relevance, usability, and user privacy data security, AI shopping recommendation systems significantly improve internet users' purchasing experience and purchase intention.

Table of Contents

CHAPTER 1

1.1 Introduction

With the rise of e-commerce as a growth model, online shopping is gradually changing traditional shopping behavior, offering consumers unprecedented convenience and a vast array of choices at their fingertips. Over the past decade, online retail has grown exponentially, creating new opportunities for businesses and consumers alike. The rise of e-commerce platforms such as Amazon, Alibaba, eBay, and Shopee has facilitated this shift, enabling seamless transactions across borders and time zones (Gupta & Bansal, 2019). This e-commerce marketplace has not only expanded access to goods and services, but has also intensified competition among retailers, prompting them to innovate to attract and retain customers.

One of the key innovations in e-commerce is the implementation of recommender systems. These systems analyze user data to provide personalized product recommendations that enhance the shopping experience. Traditional recommender systems rely on collaborative filtering and content-based filtering methods to predict and recommend items that a user may like based on the user's past behavior. or the behavior. of similar users (Fu & Ma, 2021). Such systems have proven to be effective in increasing user engagement and driving sales because they make the shopping process more intuitive and tailored to individual preferences. Traditional e-commerce marketing recommendation systems usually use consumers' historical consumption and browsing information on the current platform's website to further recommend personalized goods and services, and based on high-tech based on 'Big Data + Artificial Intelligence (AI)', e-commerce platforms mine and analyze a large amount of user data aiming to provide consumers with more accurate personalized goods and services (Alamdari & Navimipour, 2020).

Compared with the traditional e-commerce marketing recommendation system, e-commerce platform. enterprises realize precision marketing through AI intelligent recommendation system, which on the one hand can improve the efficiency of information data processing on the platform. website and consumers' willingness to buy; on the other hand, it can improve consumers' satisfaction and loyalty to the platform. website (Sharma & Lijuan, 2015). Some studies have shown that AI intelligent recommendation can have an important impact on consumer behaviors such as consumers' perceived value and purchase intention, such as affecting consumers' psychological distance to online shopping, consumers' perceived experiential value and purchase intention (Yang, Tang, Men & Zheng, 2021). With the shift from traditional shopping to mobile shopping and the development of marketing thinking such as "human-centered" and "user-first", consumers can shop anytime and anywhere through e-commerce platforms, which strengthens their online impulse purchase intention, among which the influence of personalized intelligent recommendation by AI has a significant impact. The influence of AI's personalized smart recommendations is crucial. Factors affecting consumers' online purchases are manifold, including the need for accurate and detailed product information, as well as the strength of prices and discounts, and the user experience of the website, etc., of which the strongest influence is accurate product information, which is a key factor influencing consumers' consumption (Islam & Daud, 2011). Although there have been studies showing that AI recommendation has an impact on consumers' purchase intention , these studies mainly focus on the impact of AI recommendation system on consumers' information adoption intention and purchase behavior. under traditional marketing, and these studies lack a consumer behavioral perspective to study the process of its impact on consumers' online purchase intention.

Overall, the rise of artificial intelligence provides the possibility to drive the quality online shopping scene, and the functions of the AI recommendation system such as information presentation, system interaction and community influence and its impact on consumer behaviour have influenced consumer behaviour to a certain extent. For the above research, this paper uses SOR theory and consumer behaviour theory to analyse, collects data through questionnaire survey method, and uses spss software to conduct in-depth research with linear regression model in business analysis, specifically analysing in what way AI intelligent recommendation affects consumer's purchasing decision and the overall shopping experience and the difference of consumer's experience and satisfaction in this process.

1.2 Problem Statement

The integration of AI recommender systems in the e-commerce industry has to some extent changed the way consumers interact with online platforms. This is especially true through AI-powered recommender systems. Despite the increasing use of AI recommendations, there are still key issues that need to be addressed, and these key issues include the lack of understanding in current research of how AI recommender systems affect consumer behaviour in e-commerce, particularly in terms of trust, satisfaction and willingness to use. This research will provide insights into the psychological and behavioural impacts of AI systems on consumers, including information presentation, system interactions and community impacts. This helps to develop the potential of AI recommendation systems, so it is essential to understand how these AI recommendations affect consumers from their perspective. A full understanding of this process through research can better optimise their impact on consumer behaviour. This study aims to investigate these key issues and provide actionable insights for e-commerce platforms.

Although several studies have explored the impact of AI's personalised recommender systems on consumer decision-making, these studies often lack a comprehensive analysis of consumer trust, satisfaction and willingness to use.Nguyen (2021) points out that most of the existing research focuses on the technical implementations of recommender systems, and there is less research on the consumer interaction experience . Ideally, consumers would have a high level of trust and acceptance of AI recommender systems, leading to increased user engagement and satisfaction (Rodrigues, 2021). However, the potential of AI recommender systems has not yet been fully realised due to a lack of in-depth knowledge of consumer trust and satisfaction.

The Problem Statement of this study is to analyse the impact of AI recommender systems on consumer behaviour by revealing the key factors that influence consumer trust, satisfaction and willingness to use. From the study of consumer behaviour and psychological level, the aspects of information presentation, system interaction and community influence will be analysed in depth, and optimisation strategies will be proposed at the end.

1.3 Research Questions

This study addresses six key research questions to provide a comprehensive understanding of how AI recommendation systems influence consumer behavior. Each question is designed to explore different dimensions of consumer actions and perceptions in the context of these systems.

1. ‘How do consumers perceive AI recommendations?’ This question explores how consumers view various aspects of AI recommendation systems, including their entertainment value, ease of use, and relevance. Understanding these perceptions is crucial for assessing how well the recommendation systems meet consumer expectations. For example, does the entertainment value enhance engagement, and how does the ease of use impact satisfaction? Insights into consumer perceptions help identify factors that drive positive or negative experiences and guide the refinement of recommendation systems to better align with consumer preferences.

2. ‘How do AI recommendations influence consumers' purchase decisions and overall shopping experience?’ This question examines the role of AI recommendations throughout different stages of the shopping process, from information search to evaluation and final decision-making. By understanding how recommendations impact these stages, the study aims to optimize these systems to better support consumers. For instance, how do recommendations shape decision-making, and to what extent do they affect the final purchase choice? This analysis helps in enhancing the design and functionality of recommendation systems to improve consumer decision-making and overall shopping experience.

3. ‘How do the entertainment value and ease of use of AI recommendations impact the shopping experience?’ This question focuses on evaluating whether the entertainment value and ease of use of AI recommendations contribute significantly to the shopping experience. For example, does a recommendation system that provides entertaining content lead to a more enjoyable shopping experience? How critical is ease of use in ensuring that consumers can effectively navigate and utilize the system? Understanding these factors helps in designing recommendation systems that not only engage users but also streamline their shopping processes, enhancing overall satisfaction.

4. ‘How does the relevance of AI recommendations and data security measures affect consumer purchase intentions and trust?’ This question investigates how the relevance of recommendations and the perceived security of user data influence consumer trust and purchase intentions. It examines whether recommendations that closely match consumer preferences increase the likelihood of purchase and how robust data protection measures affect trust. For example, does the relevance of a recommendation impact a consumer’s willingness to make a purchase, and how does data security influence overall trust in the recommendation system? This analysis provides insights into how effective recommendations and strong data protection can enhance consumer trust and drive purchasing decisions.

5. ’What role do relationship satisfaction and trust play in the relationship between AI recommendation features and consumer outcomes?‘ This question explores how intermediary factors such as relationship satisfaction and trust mediate the effects of AI recommendation features on consumer outcomes. Specifically, it looks at how satisfaction with the recommendation experience and trust in the system influence shopping experience and purchase intentions. For instance, does higher satisfaction with the recommendation system correlate with a better shopping experience, and how does trust impact the decision to follow through with a purchase? Understanding these mediating factors offers insights into the psychological dynamics that affect consumer responses to AI recommendations.

By addressing these questions, the study aims to gain a thorough understanding of how AI recommendation systems affect consumer behavior, focusing on enhancing user experience and driving purchasing decisions.

code

Research Question

Q1

How do consumers perceive AI recommendations?

Q2

How do AI recommendations influence consumers' purchase decisions and overall shopping experience?

Q3

How do the entertainment value and ease of use of AI recommendations impact the shopping experience?

Q4

How does the relevance of AI recommendations and data security measures affect consumer purchase intentions and trust?

Q5

What role do relationship satisfaction and trust play in the relationship between AI recommendation features and consumer outcomes?

1.4 Research Objectives

The primary objectives of this research are to explore and analyze the impact of AI recommendation systems on consumer behavior. Specifically, this study aims to achieve the following goals:

1. Evaluate Consumer Perceptions of AI Recommendations: To understand how consumers perceive AI recommendation systems, including their entertainment value, ease of use, and relevance. This objective seeks to assess how these factors influence user satisfaction and acceptance of recommendation systems, providing insights into how well these systems align with consumer expectations.

2.Analyze the Impact of AI Recommendations on Purchase Decisions and Shopping Experience: To examine the role of AI recommendations in various stages of the consumer decision-making process. This objective focuses on understanding how recommendations affect information search, evaluation, and final purchase decisions, as well as their overall impact on the shopping experience. By identifying how recommendations influence these stages, the research aims to optimize recommendation systems to better support consumer decision-making and enhance the shopping experience.

3. Assess the Effects of Entertainment Value and Ease of Use on Shopping Experience: To determine the significance of the entertainment value and ease of use of AI recommendations in shaping the consumer shopping experience. This objective involves evaluating whether these factors contribute positively to user satisfaction and engagement, and how they influence the effectiveness of the recommendation system in facilitating a smooth shopping process.

4. Investigate the Influence of Relevance and Data Security on Purchase Intentions and Trust: To explore how the relevance of AI recommendations and the perceived security of user data impact consumer trust and purchase intentions. This objective aims to understand the relationship between the accuracy of recommendations, the robustness of data protection measures, and their effects on consumer trust and willingness to make purchases.

5. Examine the Role of Relationship Satisfaction and Trust in Mediating the Effects of AI Recommendations: To analyze how intermediary variables such as relationship satisfaction and trust mediate the relationship between AI recommendation features and consumer outcomes. This objective seeks to understand how these factors influence the shopping experience and purchase intentions, offering insights into the psychological mechanisms that affect consumer responses to AI recommendations.

By addressing these objectives, this research aims to provide a comprehensive understanding of how AI recommendation systems affect consumer behavior, with the goal of improving the design and effectiveness of these systems to enhance user experience and drive purchasing decisions.

1.5 Significance of the Study

This study offers a valuable contribution to the theoretical understanding of consumer behavior. within the realm of AI-driven recommendation systems. By examining how AI recommendations influence consumer perceptions and actions, the research enhances existing knowledge about consumer decision-making processes and technology adoption. The study provides insights into how factors such as entertainment value, ease of use, relevance, and data security impact consumer trust and purchase intentions. Additionally, by investigating intermediary variables like relationship satisfaction and trust, the study helps to clarify the mechanisms through which AI recommendations affect consumer behavior. These contributions are important for refining theoretical models of consumer behavior. and technology acceptance, particularly in the context of advanced recommendation technologies.

Practically, this study offers useful insights for businesses aiming to improve AI recommendation systems and enhance customer experiences. Understanding how different aspects of AI recommendations—such as their relevance, ease of use, and data security—affect consumer behavior. can help businesses design more effective recommendation strategies. The findings provide guidance on which factors are most important for building consumer trust and increasing purchase intentions. By highlighting the role of entertainment value and ease of use in shaping the shopping experience, the study suggests ways to tailor recommendation systems to better meet consumer needs. Furthermore, the insights into how relationship satisfaction and trust mediate the effects of AI recommendations can inform. strategies for improving customer interactions and fostering loyalty. Overall, this research supports practical efforts to develop recommendation systems that align with consumer expectations and enhance their overall shopping experience.

In conclusion, this study aims to bridge the gap between theoretical knowledge and practical applications, provide a holistic view of the impact of AI recommendation systems on consumer behaviour, and offer strategies to improve their effectiveness and user acceptance.

1.7 Organization of Chapters

This paper is titled "The impact of artificial intelligence recommendation on consumer behavior. in the e-commerce industry". The full text is divided into five chapters as follows:

Chapter 1: Introduction. It mainly explains the research background, Problem Statement, Research Questions, Research Objectives, Significance of the Study, etc.

Chapter 2: Literature Reviews. It mainly sorts out the literature from the following six key parts. And explains the key definitions. And analyzes the importance and impact of AI recommendation from the international and corporate perspectives. This chapter proposes the theoretical basis of this paper, namely SOR theory and consumer behavior. theory. On these basis, the variables of this study are refined, the relationship between variables and the Conceptual Framework of the study are determined.

Chapter 3: Research Methodology. It mainly explains the research methods of this study, including questionnaire design and data collection techniques.

Chapter 4: Data Analysis. This chapter mainly introduces the analysis methods used to interpret the collected data. We first evaluate the normality of the data to ensure the validity of the statistical test, then reveal the relationship between variables through factor analysis, and confirm the consistency of the measurement tool through reliability analysis. This chapter also introduces the profile of the respondents through tables and charts. Finally, this study combines data analysis with the research objectives to evaluate the extent to which the research results have achieved the research objectives.

Chapter 5: Research Conclusions and Recommendations. This chapter will synthesize the research results and explore their impact. This chapter first discusses, creatively combines the research results with the research objectives, and proposes explanations. On these basis, this chapter will propose feasible suggestions for improving the quality and efficiency of AI recommendations, thereby improving consumers' user experience. This chapter will describe the limitations of the current research and propose directions for future research. Finally, the main research results and their overall significance are summarized.

CHAPTER 2- Lit Review

2.1  Introduction

In contemporary electronic commerce, AI recommendation systems have become crucial in shaping consumer behavior. These systems use advanced algorithms to offer personalized product suggestions, which significantly impact various stages of the consumer decision-making process (Jannac The impact of artificial intelligence recommendation on consumer behaviour in the e-commerce industryHaskell h & Adomavicius, 2016). To systematically study the impact of AI recommendations on consumer behavior, this section organizes existing literature and analyzes it from six key perspectives.

2.1.1 Consumer Perceptions

Accurate personalised recommendations in e-commerce platforms can efficiently improve the shopping efficiency of consumers. Correctly personalised recommendations can bring a sense of surprise and satisfaction to consumers, which can effectively enhance the enjoyment and positive experience of shopping, thus optimising the consumer shopping experience.Yoon and Lee (2021) state that AI recommendation systems significantly influence consumer behaviour and satisfaction, indicating positive consumer perceptions of these technologies. In this process, personalised systems will not only recommend products based on consumers' historical behaviour, but also recommend new categories of products based on similar users' preferences, which can also increase the range of choices and enrich the consumer's shopping experience In the field of recommender systems, there is a growing interest in understanding the interactions between various aspects of behavioural research, intelligent systems, and decision support systems. And AI-powered technologies play a key role in enhancing the capabilities of these aspects, leading to more effective recommender systems (Konstan, 2012). For example, intelligent systems powered by AI can process large datasets to provide accurate and personalised recommendations based on user behaviour, preferences and decision-making processes. International Business Machines (IBM) released its third biennial consumer research report titled Revolutionising Retail with AI: Customers Won't Wait. The report is the result of a survey of 20,000 global consumers in 26 countries on their digital habits and AI usage. One of the key findings of the report is that consumers are dissatisfied with the retail experience. Only 9 per cent of respondents were satisfied with shopping in brick-and-mortar shops, while only 14 per cent were satisfied with shopping online. Interestingly, consumers surveyed showed a strong interest in using AI technology to enhance the shopping experience. 59% of respondents said they would like to use AI apps, and 80% of respondents who had not used AI for shopping said they were interested in trying it (Armonk, 2024). How AI assistants could help shoppers in the future, Luq Niazi, Global Managing Partner, IBM Consulting and Industry and Global Consumer Sector Leader, said IBM's 2024 Consumer Study found that more than half of consumers would like to use an AI assistant while shopping, but there is currently a significant gap between the capabilities of these AI tools and consumers' expectations (Ali, 2024).

2.1.2 The impact of AI recommendations on the shopping decision-making

A large number of studies have explored the impact of AI recommender systems on consumer purchasing decisions and behaviours, and Kumar's (2024) study found that personalised recommendations can significantly increase conversion rates and average order values, and that tailored suggestions not only enhance user engagement and loyalty, but also have varying degrees of impact across different product categories. This finding underscores the potential of AI recommender systems to drive consumer behaviour.AI recommender systems are effective in stimulating consumers' desire to buy by increasing the frequency and amount of spending through accurate recommendations. When consumers browse a certain type of product, the related products recommended by the system may trigger their behaviour to make additional purchases, thus increasing the platform's sales and profits (Pathak & Garfinkel, 2010). However, there may be a risk of system algorithmic bias and manipulation in the recommendation process. Platforms may recommend more profitable products rather than products that are truly appropriate for consumer needs (Marcellis, 2022). Such manipulation may harm consumers' interests and trust in e-commerce platforms.The impact of AI personalisation on search may also be seen in other ways. For example, when searching for unknown product names, consumers sometimes do not know the name of the product they want to buy, and eliminating them one by one would waste a lot of time or even abandon the desire to buy the product. A convenient search method combines voice input and image recognition technology to perform. image matching by inputting detailed attributes of the product to find other images that are similar or related to the input image. the AI system analyses the features and metadata of the image to be able to recommend products that are related to the input image and provide a link to purchase. This image recognition-based recommendation system is widely used in Southeast Asia due to the fact that consumers in the region have a high interest in fashion trends and tend to look for products recommended by online celebrities through images. Research has shown that image recognition technology can not only enhance the accuracy of recommendation systems, but also significantly improve consumers' shopping experience (Zhang et al., 2021). In this context, the introduction of image search technology effectively meets Southeast Asian consumers' demand for quick access to Netflix products, enabling them to find their favourite fashion items more conveniently (Chen & Xu, 2022).

2.1.3 Entertainment Value & Ease of Use

The entertainment value and ease of use of AI recommendation systems are critical factors influencing consumer behavior. Research has shown that the entertainment value of recommendations, which includes the enjoyment and engagement derived from the recommendation process, significantly affects consumer satisfaction and engagement (Hassanein & Head, 2007). Consumers are more likely to engage with and respond positively to recommendation systems that provide an enjoyable and stimulating experience, enhancing their overall shopping experience (Liu et al., 2020). This enjoyment, however, is often secondary to the system’s ease of use, which remains a core determinant of user satisfaction. Ease of use refers to the simplicity and intuitiveness of the recommendation interface, which facilitates user interactions and contributes to a smoother shopping process (Davis, 1989). The perceived ease of use has been consistently linked to higher user satisfaction and increased likelihood of system adoption (Venkatesh & Bala, 2008). While entertainment value can enhance engagement and make the shopping experience more pleasant, ease of use is fundamental in ensuring that consumers can efficiently navigate and utilize recommendation systems without frustration. Thus, while entertainment value adds to the appeal of AI recommendations, ease of use is crucial for maintaining a positive and functional consumer interaction with the system.

2.1.4 Impact of Accuracy and Relevance of AI Recommendations

The impact of recommendation accuracy and relevance on consumer trust has also received widespread attention. Rohden (2023) found that highly accurate recommendations enhance consumer satisfaction and trust, while recommendation relevance is a key factor in maintaining consumer engagement and trust. On the one hand accurate and relevant recommendations not only save consumers' time and energy, but also increase consumers' repurchase rate of products and users' stickiness to the platform. (Zhang & Zhao, 2018). This suggests that improving the accuracy and relevance of recommendation systems is an important way to enhance consumer trust. When the accuracy of recommendations is improved, the sales of the platform. will be improved accordingly. The model must be localised and must be familiar with local brands as well as words, and as the accuracy of the AI recommendations increases, the sales of the platform. also increase. For example, Lazada has developed a "graphic multimodal recognition" model that recognises more than 97% of Southeast Asian brands. The system can accurately search by uploading photos of celebrities wearing Tommy Hilfiger white polo shirts without incorrectly recommending Abercrombie & Fitch white polos, demonstrating high recommendation accuracy. After solving the first two key issues, Lazada took the lead in the Southeast Asian market and saw a five-fold increase in the number of users of its relevant features in just six months. Research has shown that the accuracy of recommendation systems has a direct impact on consumer satisfaction and the platform's sales performance (Jiang & Li, 2023). Therefore, by improving recommendation accuracy, Lazada not only enhanced user experience, but also significantly boosted sales growth (Smith & Chen, 2022).

2.1.5 Differences in Consumer Behavior

Studies have shown significant differences in the response of different demographics and user profiles to AI recommendation systems.Singh's (2021) study shows that the process of AI recommendation involves the system analysing the user's demographics such as gender, age and other demographics, and making more accurate matches for the user through data analysis. The study suggests that younger consumers are more receptive to AI recommendations than older groups, whereas younger consumers may have a greater preference for fashion and tech products, whereas older ones may have a greater preference for health science lifestyle. products. Therefore different demographic information can also imply different needs and preferences, and by implementing precise marketing strategies for different groups of people can improve the marketing effectiveness and conversion rate of e-commerce businesses (Saluja, Soth & Pawar 2023). Socio-economic factors and previous online shopping experiences also play an important role in shaping consumer responses, for example, high-income consumers may be inclined to purchase luxury or premium goods, whereas low-income consumers may be more focused on value for money (Kotler & Keller, 2016). Consumers with higher levels of education are usually more knowledgeable about product information and brands, and they may be more cautious and rational in their purchase decisions (Baker & Churchill, 2018). Research suggests that positive shopping experiences can enhance consumers' trust and satisfaction with shopping platforms, while negative shopping experiences may lead to customer attrition (Hsu, Chuang, & Hsu, 2014). Experienced consumers are more likely to choose familiar platforms for shopping because they have higher trust in the service quality and product reliability of these platforms (Venkatesh & Bala, 2008). These findings reveal the diverse user needs that recommender systems need to consider in their design and application.

2.1.6 Consumer acceptance of privacy and data security

Users are also evolving. They are becoming proactive, knowledgeable, and prioritise privacy. As a result, sellers may need to make efforts to build trust with these users. The impact of privacy and data security issues on consumer acceptance of AI recommendation systems cannot be ignored. Personalised recommendation systems rely on the collection and analysis of large amounts of consumer data, which raises serious privacy and data security concerns. The main issues include false information, fraud, and copyright infringement, and consumers may be concerned about the misuse or disclosure of their personal information, which reduces their trust in recommender systems.Wang (2021) points out that consumers are wary of data use and storage, and that addressing these issues through strong data protection measures and clear communication can enhance consumer trust and acceptance. Singh (2024) further emphasises the importance of transparency and consumer control over data, which is crucial in fostering positive consumer attitudes towards AI systems. Therefore unsupervised learning in the e-commerce domain is very important. In e-commerce AI recommendation systems, machine learning relies on data analysis of items and orders, which usually requires an exhaustive labelling system. However, many orders are virtually unlabelled. For example, Taobao, a subsidiary of China's AliPay, uses AI techniques to categorise orders for verification and labels these orders through unsupervised learning, which is subsequently fed into the recommender system for intelligent recommendations (Zhang & Zhang, 2021). This approach not only improves the accuracy of recommendations, but also demonstrates the positive role of unsupervised learning in enhancing consumer acceptance of privacy and data security (Li, 2022).

2.1.7 Mediating Roles

In exploring the impact of AI recommendation systems on consumer behavior, relationship satisfaction and trust play crucial mediating roles. Relationship satisfaction, which reflects the overall contentment of consumers with their interaction with the recommendation system, has been shown to significantly influence both shopping experience and purchase intentions (Morgan & Hunt, 1994). Satisfied consumers are more likely to develop positive attitudes towards the recommendation system and exhibit higher engagement levels, which can enhance their overall shopping experience and increase their likelihood of making a purchase (Hennig-Thurau, Gwinner, Walsh, & Gremler, 2004). Similarly, trust in the recommendation system is a fundamental determinant of consumer behavior. Trust encompasses the confidence consumers have in the accuracy of recommendations and the protection of their personal data. Research indicates that higher levels of trust lead to increased acceptance of recommendations and greater willingness to share personal information (Sirdeshmukh, Singh, & Sabol, 2002). When consumers trust that the recommendation system will deliver relevant suggestions and safeguard their privacy, they are more likely to act on those recommendations and maintain a positive shopping experience (Chen & Zhang, 2019). Thus, both relationship satisfaction and trust are essential in shaping how consumers interact with AI recommendation systems, influencing their overall engagement and purchasing behavior.

These literatures provide a solid theoretical foundation for this study, and by exploring the multifaceted effects of AI recommendation systems on consumer behaviour, they provide important insights and directions for further research.

2.2 key definitions

When studying the impact of artificial intelligence (AI)-driven recommender systems on consumer behaviour in e-commerce, it is crucial to clarify key terms, features and attributes. These definitions will provide the basis for subsequent analyses and discussions.

AI-Driven Recommendation Systems: AI recommendation systems use machine learning algorithms and data analytics to provide personalised recommendations based on user behaviour and preferences. Its key features include personalisation, dynamic adaptation and data-driven. Recommendation systems analyse user data (e.g., browsing history, purchase history, etc.) and use algorithms (e.g., collaborative filtering and content filtering) to optimise recommendations (Ricci, Rokach, & Shapira, 2015).

Consumer Perceptions: Consumer perceptions refer to their overall view of the AI recommendation system, including trust, satisfaction, and ease of use. Trust reflects confidence in the reliability of the system, satisfaction is the degree of satisfaction with the quality of recommendations, and ease of use focuses on the convenience of use (Hsu, Chuang, & Hsu, 2014). The system's interface design and feedback mechanisms directly affect these perceptions (Srinivasan & Moorman, 2005).

Consumer Purchasing Decisions: Consumer purchasing decisions cover the processes of information search, choice evaluation, and final purchase. Factors such as personalised recommendations, product reviews, brand reputation, price and promotions all influence decisions (Kotler & Keller, 2016). The conversion rate and order value of recommender systems are key metrics to assess their impact (Chechi & Malerba, 2021).

Recommendation Accuracy and Relevance: Recommendation accuracy and relevance refers to how well the recommended content matches consumer needs and interests. High-quality data and algorithms are critical to improving recommendation accuracy and relevance (Jannach & Adomavicius, 2016). The matching ability of the algorithm and the quality of the data directly affects the effectiveness of recommendations (Smith & Linden, 2017).

Privacy and Data Security: Privacy and data security involves the protection of consumers' personal data, including data encryption, transparency, and user control. Data protection prevents unauthorised access, transparency provides information about data usage, and user control allows consumers to manage data sharing (Solove, 2021). The use of data encryption and clear privacy policies help to safeguard consumer privacy and choice (Culnan & Bies, 2003).

By clarifying these key definitions, characteristics and attributes, the impact of AI-driven recommender systems on consumer behaviour in e-commerce can be systematically understood and analysed, providing a solid theoretical foundation for further research.

2.3 Global perspectives

Omni-channel marketing is gaining traction as customers expect a seamless service experience across multiple channels. However, this poses a challenge for brands to ensure consistency of service.Zendesk reports that 35 per cent of customers expect customer service across multiple channels (Zendesk, 2022). In this context, the application of AI recommender systems is particularly important to help e-commerce companies to improve the intelligence of their recommender systems and enhance operational portability, research capabilities and collaboration. By combining academic research with real-world applications, AI recommender systems offer e-commerce businesses the opportunity to increase turnover in the online environment (Smith & Linden, 2017). AI recommendation engines increase user satisfaction and brand loyalty by delivering a consistent and personalised experience across multiple channels (e.g., social media, websites and mobile applications). They use user data (e.g., search history and purchase history) to provide personalised suggestions, for example, when a user searches for a product on a website, the recommendation engine can recommend related items through channels such as email. Research has shown that personalised content can significantly increase brand loyalty, with Microsoft reporting that 96% of customers believe that great customer service and experience is critical to brand loyalty (Microsoft, 2021).

AI recommendation systems are widely used in Asia, especially in China. Chinese e-commerce giants such as Alibaba and Jingdong use advanced AI technology to provide highly personalised shopping experiences (Wang & Zhang, 2022). Research shows that Asian consumers are generally more receptive to AI technology, but cultural differences affect their responses. Chinese consumers typically trust systems based on social recommendations, while Japanese consumers focus more on the accuracy and personalisation of recommendations (Li, 2021). In developing countries, although the penetration of AI recommendation systems is low, these markets show great potential for growth as the acceptance of digital shopping increases (Chen & Xu, 2020). However, limitations in infrastructure and level of technological development are the main challenges.Kumar and Singh's (2022) study points out that despite the weaker technological infrastructure, AI recommender systems have a huge potential for growth in these markets as internet penetration increases and mobile devices are widely used.

Taken together, research on AI recommender systems from a global perspective reveals diversity across regions in terms of technology acceptance, privacy protection, and cultural differences. These differences not only influence the implementation strategies of AI technologies, but also provide a rich background and case studies for further research. Through in-depth analysis of the global market, the application of AI recommender systems in different environments can be better understood and optimised to provide theoretical support and practical guidance for promoting the development of global e-commerce.

2.4 zooming in

In order to gain a deeper understanding of the application of AI-powered recommender systems and their impact in global markets, it is necessary to focus on specific markets, and this study analyses case studies of two of the largest e-commerce platforms in East Asia. These case studies not only help to reveal the diverse applications of AI recommender systems in different cultures and market environments, but also provide valuable experiences and lessons learned to help understand how to optimise AI technology to enhance consumer experience and satisfaction.

Lazada

Lazada specialises in providing tailored shopping experiences through data-driven personalisation. The platform. uses customer data analytics to optimise product recommendations and marketing campaigns to increase conversion rates. According to Lazada's report, the introduction of personalised recommendation systems has led to an increase in conversion rates by about 10% (Lazada, 2023). This is due to the recommendation system's ability to accurately match users based on their behaviours and preferences, improving shopping relevance and user satisfaction. Its omni-channel strategy strengthens brand image and engages consumers by integrating multiple channels (Lazada, 2022).AI technology plays a key role in Lazada's recommender system by analysing user behaviours and purchasing history to provide personalised suggestions, thus increasing customer engagement and brand loyalty (Liu, 2021). In addition, Lazada has introduced augmented reality (AR) technologies such as virtual trial tools to further enhance user experience (Lazada, 2022).

Taobao (Alibaba)

Taobao is transforming the consumer shopping journey by embedding artificial intelligence deeply into its platform. One notable innovation is Taobao Ask, a generative AI shopping assistant to be launched in 2023. This is an AI-powered market insight tool that helps merchants analyse consumer demand, industry trends and competitor strategies (Zhang, 2024). The tool allows for real-time adjustments in areas such as product mix and pricing to keep merchants competitive. Ask helps users by answering product-related questions, providing detailed product recommendations, and offering multimedia content such as videos and live streams. The tool is particularly popular during shopping festivals such as 11.11, where it helps users find the best deals and make informed purchasing decisions.AI tools generate product descriptions and marketing content based on popular keywords and consumer insights (Zeng, 2023). This automation not only saves time, but also increases product visibility and conversion rates.

AI innovations played a crucial role in mega shopping events such as the 11.11 Global Shopping Festival, where Taobao's AI tools were used more than 1.5 billion times, helping the company earn about $450 billion (Rashidin &Gang, 2021). The platform's focus on AI-driven innovations not only enhances the consumer experience, but also significantly improves the operational efficiency of merchants, enabling them to thrive in the highly competitive Chinese e-commerce market         

船载 OpenHarmony AIS 航海工作台:告警与防碰撞预警系统 船载OpenHarmony AIS航海工作台的告警与防碰撞系统,集成十余类告警(超速、偏航、会遇、SOS等),通过网格空间索引实现高效邻近判定,保障实时性。告警管理器统一调度规则、去重、通知与日志,支持悬浮球、语音、跑马灯、弹窗多渠道触达,确保航行安全。SOS紧急通道独立运行,避免干扰。系统设计强调性能优化与多场景适配,为航海安全提供核心支撑。 阅读详情

相关推荐

AI 应用开发:四维架构落地实战——以 HR Agent 为例拆解感知·推理·记忆·执行【意图路由、规划编排、长期记忆、工具执行、面试题】

本文基于四维架构(感知·推理·记忆·执行),以HR Agent为例,系统拆解大模型应用从“聊天”到“生产可用”的落地路径。通过Spring AI实现意图路由、规划编排、RAG检索与工具调用,结合真实踩坑复盘,揭示架构化设计对安全、成本与可维护性的关键价值,为AI Agent工程化提供可复用的实战框架。

王大师企业官方博客 1325

springboot李庄古镇农产品销售系统 毕业设计----附源码66743

系统将用户、商家、管理员三种角色分为农产商品推荐、客服咨询、多级商城浏览、购物车和订单流转、地址簿维护、售后申请、店铺信息配置、商品上架下架、配送状态更新、可视化报表、用户权限审核等具体的业务。按照软件工程规范依次完成需求调研、总体架构设计、前后端功能模块划分、数据库关系建模、编码实现和测试验证。

VXSJ881的博客 170

激光SLAM总结——Fast LIO / Fast LIO2 / Faster LIO

激光SLAM总结——Fast LIO / Fast LIO2 / Faster LIO 在之前的工作中有接触过LOAM,最近在工作中又接触到Faster LIO相关的工作,于是想着对Fast LIO / Fast LIO2 / Faster LIO这一系列工作进行一个简单的总结,以加深自己对激光SLAM算法的理解,之前总结过的一些和激光SLAM算法相关的博客还有,感兴趣的同学可以一起学习讨论: 激光SLAM总结——VLOAM / LIMO算法解析 激光SLAM总结——LIO-Mapping / LIOM /

weixin_44580210的博客 1万+

AI 问数真的能替代 BI 报表吗?

AI问数赋能自然语言查询,提升临时分析效率,但无法替代传统报表。报表适用于稳定输出与持续监控,AI则擅长探索性提问。企业用好AI问数,关键在于夯实数据模型、统一指标口径、健全权限体系与结果校验机制。合理组合报表、仪表板与AI,才能实现高效、准确的业务洞察。

葡萄城技术团队博客 341

AD620单电源应变片测量电路分析

笔者最近在用AD620做应变片的压力检测电路,正好AD620的datasheet中有对于应变片压力测量的应用电路,参考该电路进行设计及实验后发现该电路存在些许问题,并对其进行了优化改进。本文纯属个人理解,供大家参考,欢迎批评指正。 AD620应用电路分析及优化 应用电路分析 AD620手册中原电路如下(加了部分标注以便分析): 该电路中,前半部分比较明显,为经典的AD620放大电路结构:四个应变...

拙诚电子工作室 1万+

蓝桥杯FPGA赛道积分赛

代码如下: 按键驱动 代码如下 按键驱动 代码如下: 顶层文件 按键驱动 数码管驱动 声明:本代码借鉴网络资料,如有侵权和错误请联系作者删除。我是从寒假开始备赛的,由于我基础不行,备赛过程中也是混混呃呃的,备赛效率不高,总体来说备赛不认真,有点拖拉,但是最后我也坚持备赛到了今天,在备赛过程中,即复习了旧的东西,也学到了新的东西,总结起来还是有点收获的,后天北京、天津、河北地区就要进行省赛了,我会全力以赴,不管结果怎么样,尽力就好,最后祝大家都能

weixin_69804913的博客 2005

ChatGPT、Perplexity、Gemini 与 AI Overviews 引用对比:谁的引用更可信

AI 引擎的引用机制差异很大:ChatGPT 以行内引用加 Sources 列表呈现来源,Perplexity 几乎逐句可溯源,Gemini 在回答中内嵌相关来源链接,AI Overviews 则把多源聚合答案与来源列表一起放在搜索结果顶部。本文从引用透明度、来源偏好、可干预性 3 个维度对比这 4 个引擎,并给出 GEO 的 3 点启示,帮企业在多引擎环境下提升被引用的概率。

淡然一笑 的专栏 279

【SABRE 3D, SABRE 3DxT,LAM设备】2025年9月14日-Clamshell 拆卸与安装

Clamshell 拆卸与安装。

weixin_46735260的博客 1695

【从零理解 Spring AI · 12】MCP 的两种角色:做客户端还是做服务端

上一篇讲了 MCP 的概念:标准协议、三大角色、M+N 替代 M×N。

nice_lcj520的博客 616

交换机救命命令手册:华为 & 思科平台最全运维指令速查表

很多时候,真正拉你一把的不是“高深理论”,而是你手里有没有一份靠得住的命令速查表。这篇文章,希望能成为你“进可调试,退可排障”的一份救命卡片。

HuaweiLab_Peach的博客 1519

GJB 5000B二级-II实施基础

本文主要描述GJB 5000B二级II实施基础与GJB 5000A的变化情况,以及过程域各条款的实施要点

Abel的博客 2589

抽象主义人工智能和生命体AI的内在联系

彭军辉提出“抽象主义人工智能”作为第四大AI范式,批判符号主义、连接主义、行为主义剥离生命本体的误区,确立“智能是现象,生命是本体”的核心公理。主张原生智能必须依附具备存续诉求的生命本体,提出“生命体AI理论”,构建七层演化阶梯与自闭环工程框架,强调智能不可跳过底层存续机制直接模拟语言。该体系为生成式AI提供科学边界,推动从“仿真表象”向“原生智能”的范式跃迁,开辟硅基人工生命新赛道。

抽象主义人工智能和生命体AI 208

【BUG已解决】ImportError: cannot import name ‘OpenAI‘ from ‘openai‘ 解决方案

的核心是版本不匹配用确认当前版本如果是 0.x 版本 →升级到 1.x升级后代码语法要同步改为团队协作项目务必在中锁定版本号,避免"我这能跑,你那不能跑"的问题对接国产模型时,只需替换base_url参数,SDK 调用方式完全一致。

一键难忘的博客 1万+

OpenAI Control Room 架构拆解:给企业 AI Agent 加一套可审计的急停系统

OpenAI 公开了用于追踪和停止研究模型工作负载的 Control Room。本文拆解 job_id、推理前鉴权、35 秒撤销窗口与 Fail Closed 设计,并给出企业 AI Agent 急停控制面的数据模型、FastAPI 参考实现、缓存撤销策略和故障演练清单。

jerry764的博客 345

MFC C++ 中开发 PLC 通信客户端

OPC UA | 工业标准化通信 | `open62541`(开源OPC UA库)或 `UA-.NET` 封装 | 跨平台、安全可靠,支持复杂数据类型 |- 连接句柄(`modbus_t*`/`UA_Client*`/`TS7Client*`)不可跨线程共享,需为每个线程创建独立连接。Modbus 是工业常用协议,分 **RTU**(串口)和 **TCP**(以太网)两种模式,以下以 Modbus TCP 为例。s=Tag`),复杂类型需解析扩展对象。

寒寒豆豆软件公司的博客 140

visio安装包

Microsoft Office Visio 是一款专业的制图软件。利用经过更新的形状、协作工具和数据链接图表,创建专业级图表以简化复杂的信息。

OpenClaw 深度调查:把 AI 助手做成你自己的控制平面

对 openclaw/openclaw 的完整技术尽调:Gateway 架构、Agent 运行时、工具与沙箱、30 个消息通道、181 个插件、安全信任边界、工程代价与采用建议。17 章 / 10 张图,全部数字标注口径,数据截至 2026-10-31。

厚积薄发 351

IDEA的使用指南

IDEA基于java的使用手册,可以最快速的上手,在使用中不断的丰富自己的使用技巧.

AI赋能智慧港口建设|造形家快速生成港口全要素三维场景

随着智慧港口、港口数字孪生项目的持续推进,港口三维场景的需求日益旺盛。传统港口三维建模,往往需要航拍倾斜摄影采集数据,再人工搭建吊机、岸桥、船舶、集装箱等大量工业设施模型,项目周期长、人力成本高,修改迭代繁琐,制约港口可视化项目落地效率。国产AI三维建模工具,依托卫星影像 AI 识别能力,无需航拍作业,一键快速搭建港口完整三维场景,为港口数字孪生、方案汇报、投标演示提供高效的场景解决方案。

m0_69539167的博客 87

WINCC7.3破解文件

WINCC7.3硬件够破解文件,有需要的下载!WINCC7.3硬件够破解文件,有需要的下载!

从设备增点到治理闭环:老旧小区物业数智化落地实践

系统连接合同审批、银行流水、收费记录与项目台账,执行到期提醒、收入核验、异常标色和月度对账,让每一笔资金来源明确、去向可查。针对设施老旧、诉求响应慢、公共收益不透明等问题,津南以6类小区为试点场景,统一建设物业数字化底座,关联楼栋、单元、房屋、业主、空间、设施设备、工单、公共收益等核心数据。系统采用统一编码管理,将摄像头、路灯、井盖、道路、绿化等设施纳入资产台账,解决纸质记录更新滞后、故障责任难追溯的问题。基于历史工单建立分类模型和知识库,辅助识别照明、给排水、电梯、环境秩序、秩序维护等问题;

wuyeshuzhihua的博客 163

企业微信iPad协议详解:接口接入、实时消息同步与语音转码实践

深入解析企业微信iPad协议的技术架构与开发思路,涵盖API接口接入、长连接通信、实时消息同步、Silk语音转码、异常重连及高并发优化,帮助开发者了解企业微信自动化系统的设计与实现。

欢迎来到祖安! 514

Product Hunt 热榜 2026-11-13:Plurai Simulation、YouArt Desktop App、Omnigraph

2026-11-13 Product Hunt 新产品精选:Plurai Simulation、YouArt Desktop App、Omnigraph。汇总 15 款产品的中文介绍、功能关键词和产品链接。

LaughingZhu的博客 618

【AI Agent面试系列】AI Agent可观测、评测、安全相关的知识点

之前的文章介绍了和,本文主要介绍AI Agent的测评、可观测等知识。

408

Docling 原理解读:是什么、为什么、怎么样

从论文视角拆解 Docling 的整体架构、DoclingDocument 数据模型、PDF 预处理、版面分析、OCR、TableFormer 双解码器、阅读顺序与装配流程,并解释版面框与 OCR 框的区别。

PLANTTHESON的博客 402

WorkBuddy 实战:一句话需求 → 七屏可交互原型,户外徒步 App 的 AI 设计全流程

WorkBuddy 实战:一句话需求 → 七屏可交互原型,户外徒步 App 的 AI 设计全流程,跟着博主一起来还原整个过程。

夜郎King的博客 5165

从认识人工智能到理解机器学习:我的 AI 入门第一课

正式开始学习人工智能之前,我对 AI 的认识其实比较模糊:知道 ChatGPT、知道大模型,也知道现在很多岗位都在谈 AI,但如果有人问我“人工智能到底是什么”“机器是怎么学会识别一只猫的”,我未必能从头到尾讲清楚。上午的课程更偏向行业全景:人工智能的发展、研究方向、NLP、岗位分工、学习路线和 AI Coding 工具。下午的课程更偏向技术基础:机器学习的原理、有监督学习、深度学习、模型训练和评估方法。

chh563的博客 572

企业宣传片发前如何批量排查并遮挡视频里的敏感信息

做企业视频的人都会碰到一个具体问题:实拍画面里混进了不该出现的信息。办公室背景里有贴着客户信息的白板,采访对象的授权书边缘露出一串身份证号码,街采素材里路人面孔和车牌号直接入镜。人眼在时间线上扫视,会漏掉快速闪过的帧、画面边缘的小字、以及暗光区域里的证件号码。用抹影AI完成车牌、人脸和证件区域批量检测后,又手动补充了两个时间段的屏幕遮罩,最终导出时画面已无合规隐患。重点关注快速运动、遮挡、低光照条件下的目标,这些是漏检的高发区域。导入视频后选择要检测的目标类型(人脸、车牌、证件号、二维码等)。

Mysssone的博客 182
上一篇: CSC 261/461 Introduction to Databases Project Milestone 1
下一篇: PSYC5180 Quantitative Methods I Homework #1 Descriptive&Regression Quant I F24
wx_codinghelp
博客等级 码龄2年 691粉丝 84原创
评论
成就一亿技术人!
拼手气红包6.0元
还能输入1000个字符
 
 条评论被折叠 查看
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付:10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值