Telematics and Informatics Reports最新文献_第10页

A micro-influencer recommender method based on brand features using deep learning approaches 基于品牌特征的深度学习微影响者推荐方法

Telematics and Informatics Reports

Pub Date : 2025-06-14 DOI: 10.1016/j.teler.2025.100222

Mohammad Javad Shayegan, Nasim Kazem Shiroodi

Influencer marketing has become a critical strategy for brands, with micro-influencers playing a pivotal role in shaping consumer behavior on social media. These influencers significantly impact user decision-making; however, identifying the right micro-influencers who align with a brand’s identity and can maximize engagement remains a challenging task. This study aims to develop an effective system to predict which micro-influencers will generate the highest engagement for brands. The proposed method combines ResNet for extracting visual features from Instagram posts and BERT for extracting textual features, demonstrating a high degree of accuracy in predicting engagement rates. Three machine learning algorithms—Linear Regression, Random Forest, and Extreme Gradient Boosting—were employed for model training, with performance evaluated using Root Mean Squared Error (RMSE). Among these, the Random Forest model, enhanced through feature selection and hyperparameter tuning, achieved the lowest error with an RMSE of 1.50 %. Further evaluation using metrics such as AUC, cAUC, and MedR demonstrated that the proposed approach (ResNet + BERT) outperforms other methods across all metrics, achieving an AUC of 81 %, a cAUC of 67 %, and a MedR of 2. The proposed method improved AUC by 12 % and cAUC by 8 %, confirming its effectiveness in identifying high-engagement micro-influencers.

网红营销已经成为品牌的一项关键战略，微网红在塑造社交媒体上的消费者行为方面发挥着关键作用。这些影响者显著影响用户决策；然而，确定合适的微影响者，他们与品牌的身份一致，可以最大限度地提高参与度，仍然是一项具有挑战性的任务。本研究旨在开发一个有效的系统来预测哪些微影响者将为品牌带来最高的参与度。该方法结合了从Instagram帖子中提取视觉特征的ResNet和提取文本特征的BERT，在预测参与率方面显示出很高的准确性。三种机器学习算法——线性回归、随机森林和极端梯度增强——被用于模型训练，并使用均方根误差（RMSE）来评估性能。其中，随机森林模型通过特征选择和超参数调优得到增强，误差最小，RMSE为1.50%。使用AUC、cAUC和MedR等指标的进一步评估表明，所提出的方法（ResNet + BERT）在所有指标上都优于其他方法，实现了81%的AUC、67%的cAUC和2的MedR。该方法将AUC提高了12%，cAUC提高了8%，证实了其在识别高参与度微影响者方面的有效性。

{"title":"A micro-influencer recommender method based on brand features using deep learning approaches","authors":"Mohammad Javad Shayegan, Nasim Kazem Shiroodi","doi":"10.1016/j.teler.2025.100222","DOIUrl":"10.1016/j.teler.2025.100222","url":null,"abstract":"<div><div>Influencer marketing has become a critical strategy for brands, with micro-influencers playing a pivotal role in shaping consumer behavior on social media. These influencers significantly impact user decision-making; however, identifying the right micro-influencers who align with a brand’s identity and can maximize engagement remains a challenging task. This study aims to develop an effective system to predict which micro-influencers will generate the highest engagement for brands. The proposed method combines ResNet for extracting visual features from Instagram posts and BERT for extracting textual features, demonstrating a high degree of accuracy in predicting engagement rates. Three machine learning algorithms—Linear Regression, Random Forest, and Extreme Gradient Boosting—were employed for model training, with performance evaluated using Root Mean Squared Error (RMSE). Among these, the Random Forest model, enhanced through feature selection and hyperparameter tuning, achieved the lowest error with an RMSE of 1.50 %. Further evaluation using metrics such as AUC, cAUC, and MedR demonstrated that the proposed approach (ResNet + BERT) outperforms other methods across all metrics, achieving an AUC of 81 %, a cAUC of 67 %, and a MedR of 2. The proposed method improved AUC by 12 % and cAUC by 8 %, confirming its effectiveness in identifying high-engagement micro-influencers.</div></div>","PeriodicalId":101213,"journal":{"name":"Telematics and Informatics Reports","volume":"19 ","pages":"Article 100222"},"PeriodicalIF":0.0,"publicationDate":"2025-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144321636","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Generative AI risks and resilience: How users adapt to hallucination and privacy challenges 生成人工智能的风险和弹性：用户如何适应幻觉和隐私挑战

Telematics and Informatics Reports

Pub Date : 2025-06-13 DOI: 10.1016/j.teler.2025.100221

Chunsik Lee , Junga Kim , Joon Soo Lim , Donghee Shin

Purpose

This study examines two central risks affecting continued use of generative AI (GenAI)—AI hallucinations and privacy concerns—and explores how protective behaviors serve as adaptive mechanisms to mitigate these risks.

Design/methodology/approach

Drawing on Protection Motivation Theory, the study tests a risk-adaptive GenAI use model using survey data from 789 users recruited via a Prolific panel. Structural equation modeling is employed to analyze direct and moderating effects.

Findings

Privacy concerns negatively influence user attitudes while positively predicting both personal and system-level protective behaviors. Hallucination risk is similarly associated with negative attitudes but positively predicts information verification. Notably, only information verification significantly moderates the link between attitude and continuance intention.

Originality

The study extends Protection Motivation Theory to the GenAI context by developing and validating new constructs—hallucination risk and system-level privacy-protective behaviors. It reveals how different types of risk trigger distinct behavioral adaptations, highlighting the dual role of risk as both a deterrent and catalyst in GenAI adoption.

Practical implications

The study offers insights that risk perceptions may not deter continued use of GenAI when users perform risk protective behaviors. By highlighting which protective behaviors enhance continued use, the findings inform risk-mitigation strategies for developers, educators, and regulators.

本研究探讨了影响持续使用生成式人工智能（GenAI）的两个主要风险——人工智能幻觉和隐私问题，并探讨了保护行为如何作为自适应机制来减轻这些风险。设计/方法/方法：根据保护动机理论，本研究通过一个多产小组招募的789名用户的调查数据，测试了一种风险适应性GenAI使用模型。采用结构方程模型分析直接效应和调节效应。发现隐私问题对用户态度有负面影响，而对个人和系统级保护行为有积极预测。幻觉风险同样与消极态度相关，但对信息验证有积极的预测作用。值得注意的是，只有信息验证能显著调节态度与继续倾向之间的联系。该研究通过开发和验证幻觉风险和系统级隐私保护行为的新结构，将保护动机理论扩展到基因环境。它揭示了不同类型的风险如何引发不同的行为适应，突出了风险在基因采用过程中既是威慑又是催化剂的双重作用。实际意义该研究提供的见解是，当用户采取风险保护行为时，风险感知可能不会阻止继续使用GenAI。通过强调哪些保护行为可以增强持续使用，研究结果为开发商、教育工作者和监管机构提供了降低风险的策略。

{"title":"Generative AI risks and resilience: How users adapt to hallucination and privacy challenges","authors":"Chunsik Lee , Junga Kim , Joon Soo Lim , Donghee Shin","doi":"10.1016/j.teler.2025.100221","DOIUrl":"10.1016/j.teler.2025.100221","url":null,"abstract":"<div><h3>Purpose</h3><div>This study examines two central risks affecting continued use of generative AI (GenAI)—AI hallucinations and privacy concerns—and explores how protective behaviors serve as adaptive mechanisms to mitigate these risks.</div></div><div><h3>Design/methodology/approach</h3><div>Drawing on Protection Motivation Theory, the study tests a risk-adaptive GenAI use model using survey data from 789 users recruited via a Prolific panel. Structural equation modeling is employed to analyze direct and moderating effects.</div></div><div><h3>Findings</h3><div>Privacy concerns negatively influence user attitudes while positively predicting both personal and system-level protective behaviors. Hallucination risk is similarly associated with negative attitudes but positively predicts information verification. Notably, only information verification significantly moderates the link between attitude and continuance intention.</div></div><div><h3>Originality</h3><div>The study extends Protection Motivation Theory to the GenAI context by developing and validating new constructs—hallucination risk and system-level privacy-protective behaviors. It reveals how different types of risk trigger distinct behavioral adaptations, highlighting the dual role of risk as both a deterrent and catalyst in GenAI adoption.</div></div><div><h3>Practical implications</h3><div>The study offers insights that risk perceptions may not deter continued use of GenAI when users perform risk protective behaviors. By highlighting which protective behaviors enhance continued use, the findings inform risk-mitigation strategies for developers, educators, and regulators.</div></div>","PeriodicalId":101213,"journal":{"name":"Telematics and Informatics Reports","volume":"19 ","pages":"Article 100221"},"PeriodicalIF":0.0,"publicationDate":"2025-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144365818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The influence of motivational factors on online purchase intention utilizing AI-enabled portals: A hybrid SEM-ANN approach 动机因素对利用人工智能门户网站在线购买意愿的影响：一个混合SEM-ANN方法

Telematics and Informatics Reports

Pub Date : 2025-06-01 DOI: 10.1016/j.teler.2025.100219

Sanjoy Kumar Roy , Murshedul Arafin , Jubayer Ahmed

In the current era of digital transformation, businesses increasingly utilize AI-enabled platforms to promote online purchasing behavior. Understanding consumers' purchase intentions within AI-driven environments has become essential. Therefore, this investigation aims to estimate the motivating factors that impact the intention of university students to make online purchases utilizing AI-enabled settings. For that, this research expanded the theory of planned behavior (TPB) and technology acceptance model (TAM) by integrating two additional factors, namely personal faith (PF) and perceived value (PV). Data were collected from 540 undergraduate students using purposive and convenience sampling methods through a structured questionnaire. Unlike earlier research that primarily used structural equation modeling (SEM) analysis, this investigation employed a hybrid methodology that integrates SEM with artificial neural network (ANN) assessment to evaluate real-world data. The outcomes demonstrated that attitude (PA), subjective norms (SN), ease of use (EU), perceived usefulness (PU), PF, and PV significantly correlated with undergraduates' online PI employing AI-enabled sites. On the contrary, perceived behavior control (PBC) is not significantly associated with PI. The ANN findings revealed that EU had the most significant impact on the online PI of undergraduate students with 97.21 % normalized importance, followed by PF (74.51 %), PU (70.37 %), PV (63.06 %), PA (63.05 %), SN (50.58 %), and PBC (22.22 %). The study provides valuable recommendations for businesses on effectively using AI-enabled platforms for customized advertising and promotional campaigns.

在当前的数字化转型时代，企业越来越多地利用人工智能平台来促进在线购买行为。在人工智能驱动的环境中，了解消费者的购买意图变得至关重要。因此，本调查旨在估计影响大学生利用人工智能设置进行在线购物意愿的激励因素。为此，本研究通过整合个人信念（PF）和感知价值（PV）这两个附加因素，对计划行为理论（TPB）和技术接受模型（TAM）进行了扩展。采用有目的、方便抽样的方法，对540名大学生进行结构化问卷调查。与早期主要使用结构方程模型（SEM）分析的研究不同，该研究采用了将结构方程模型与人工神经网络（ANN）评估相结合的混合方法来评估真实世界的数据。结果表明，态度（PA）、主观规范（SN）、易用性（EU）、感知有用性（PU）、PF和PV与大学生使用人工智能网站的在线PI显著相关。相反，感知行为控制（PBC）与PI没有显著相关。ANN结果显示，欧盟对大学生网络自我评价的影响最为显著，其标准化重要性为97.21%，其次是PF（74.51%）、PU（70.37%）、PV（63.06%）、PA（63.05%）、SN（50.58%）和PBC（22.22%）。该研究为企业有效利用人工智能平台进行定制广告和促销活动提供了有价值的建议。

{"title":"The influence of motivational factors on online purchase intention utilizing AI-enabled portals: A hybrid SEM-ANN approach","authors":"Sanjoy Kumar Roy , Murshedul Arafin , Jubayer Ahmed","doi":"10.1016/j.teler.2025.100219","DOIUrl":"10.1016/j.teler.2025.100219","url":null,"abstract":"<div><div>In the current era of digital transformation, businesses increasingly utilize AI-enabled platforms to promote online purchasing behavior. Understanding consumers' purchase intentions within AI-driven environments has become essential. Therefore, this investigation aims to estimate the motivating factors that impact the intention of university students to make online purchases utilizing AI-enabled settings. For that, this research expanded the theory of planned behavior (TPB) and technology acceptance model (TAM) by integrating two additional factors, namely personal faith (PF) and perceived value (PV). Data were collected from 540 undergraduate students using purposive and convenience sampling methods through a structured questionnaire. Unlike earlier research that primarily used structural equation modeling (SEM) analysis, this investigation employed a hybrid methodology that integrates SEM with artificial neural network (ANN) assessment to evaluate real-world data. The outcomes demonstrated that attitude (PA), subjective norms (SN), ease of use (EU), perceived usefulness (PU), PF, and PV significantly correlated with undergraduates' online PI employing AI-enabled sites. On the contrary, perceived behavior control (PBC) is not significantly associated with PI. The ANN findings revealed that EU had the most significant impact on the online PI of undergraduate students with 97.21 % normalized importance, followed by PF (74.51 %), PU (70.37 %), PV (63.06 %), PA (63.05 %), SN (50.58 %), and PBC (22.22 %). The study provides valuable recommendations for businesses on effectively using AI-enabled platforms for customized advertising and promotional campaigns.</div></div>","PeriodicalId":101213,"journal":{"name":"Telematics and Informatics Reports","volume":"18 ","pages":"Article 100219"},"PeriodicalIF":0.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144230009","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A four-wave longitudinal study on attitudes toward the use of AI in different domains—The self-determination theory and locus of control perspective 不同领域对人工智能使用态度的四波纵向研究——自我决定理论和控制点视角

Telematics and Informatics Reports

Pub Date : 2025-06-01 DOI: 10.1016/j.teler.2025.100220

Anica Cvetkovic , Iina Savolainen , Mayu Koike , Atte Oksanen

Artificial intelligence (AI) has gained a foothold in diverse domains and it is being used by both specialists and private individuals. The aim of this study was to analyze people’s attitudes toward the use of AI across 12 domains—medicine, nursing and caregiving, teaching and education, traffic, urban planning, building and real estate technology, defense forces, information security, job recruitment, dating services, culture and art, and political decision-making. We used the self-determination theory (SDT) and locus of control (LOC) as theoretical frameworks. This study was based on longitudinal four-wave data collected in the period 2021–2024 from participants in Finland (N = 1,226, 4,904 observations). The main independent variables were the three basic psychological needs—autonomy, competence, and relatedness—and their technological fulfillment. Our analyses were based on multilevel between–within regression models. Technological relatedness revealed both between-person effects (differences among individuals) and within-person effects (differences over time). Technological autonomy revealed significant between- and within-person effects only in specific domains. Technological competence was not a significant factor in our models. We found that a higher internal LOC was associated with positive attitudes toward AI employed in the fields of medicine and the defense forces. Our findings imply that when people experience connectedness while interacting with AI, they are more likely to view this technology in a positive manner, while experiencing autonomy and having a higher internal LOC is significant only in specific domains. These findings should be considered when designing AI tools and developing AI integration strategies and use policies.

人工智能（AI）已经在各个领域站稳了脚跟，专家和个人都在使用它。这项研究的目的是分析人们对12个领域使用人工智能的态度，这些领域包括医学、护理和护理、教学和教育、交通、城市规划、建筑和房地产技术、国防力量、信息安全、招聘、约会服务、文化和艺术以及政治决策。我们以自我决定理论（SDT）和控制点（LOC）作为理论框架。本研究基于芬兰参与者在2021-2024年期间收集的纵向四波数据（N = 1226,4904次观察）。主要的自变量是三个基本的心理需求——自主性、能力和相关性——以及它们的技术实现。我们的分析基于多水平内间回归模型。技术相关性揭示了人与人之间的影响（个体之间的差异）和人与人之间的影响（随时间的差异）。技术自主仅在特定领域显示出显著的人际和人际效应。在我们的模型中，技术能力并不是一个重要的因素。我们发现，较高的内部LOC与对医学和国防领域使用人工智能的积极态度有关。我们的研究结果表明，当人们在与人工智能互动时体验到联系时，他们更有可能以积极的态度看待这项技术，而体验自主性和拥有更高的内部LOC仅在特定领域具有重要意义。在设计人工智能工具和制定人工智能集成策略和使用政策时，应考虑这些发现。

{"title":"A four-wave longitudinal study on attitudes toward the use of AI in different domains—The self-determination theory and locus of control perspective","authors":"Anica Cvetkovic , Iina Savolainen , Mayu Koike , Atte Oksanen","doi":"10.1016/j.teler.2025.100220","DOIUrl":"10.1016/j.teler.2025.100220","url":null,"abstract":"<div><div>Artificial intelligence (AI) has gained a foothold in diverse domains and it is being used by both specialists and private individuals. The aim of this study was to analyze people’s attitudes toward the use of AI across 12 domains—medicine, nursing and caregiving, teaching and education, traffic, urban planning, building and real estate technology, defense forces, information security, job recruitment, dating services, culture and art, and political decision-making. We used the self-determination theory (SDT) and locus of control (LOC) as theoretical frameworks. This study was based on longitudinal four-wave data collected in the period 2021–2024 from participants in Finland (N = 1,226, 4,904 observations). The main independent variables were the three basic psychological needs—autonomy, competence, and relatedness—and their technological fulfillment. Our analyses were based on multilevel between–within regression models. Technological relatedness revealed both between-person effects (differences among individuals) and within-person effects (differences over time). Technological autonomy revealed significant between- and within-person effects only in specific domains. Technological competence was not a significant factor in our models. We found that a higher internal LOC was associated with positive attitudes toward AI employed in the fields of medicine and the defense forces. Our findings imply that when people experience connectedness while interacting with AI, they are more likely to view this technology in a positive manner, while experiencing autonomy and having a higher internal LOC is significant only in specific domains. These findings should be considered when designing AI tools and developing AI integration strategies and use policies.</div></div>","PeriodicalId":101213,"journal":{"name":"Telematics and Informatics Reports","volume":"18 ","pages":"Article 100220"},"PeriodicalIF":0.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144222650","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Mapping misinformation gatekeepers in non-western contexts: A computational network analysis of fake news on X (Twitter) using Gephi 绘制非西方背景下的错误信息看门人：使用Gephi对X （Twitter）上的假新闻进行计算网络分析

Telematics and Informatics Reports

Pub Date : 2025-06-01 DOI: 10.1016/j.teler.2025.100214

Moumita Roy, Maria DeRoche

This study investigates the gatekeeping of fake news dissemination on X (formerly Twitter) in non-Western contexts (data from India and Bangladesh), identifying influential actors in the online fake news ecosystem. Grounded in cascade theory, the research applies network analysis to examine how fake news spreads in these countries, where political and religious misinformation is prevalent. Using data from 10,001 X posts from April 2024 that reference the term "fake news," the study analyzes key accounts based on betweenness centrality, eigenvector centrality, and modularity class. Employing both computational methods and qualitative text analysis, the study visualizes network structures, relationships, and topic trends. Findings reveal that individual personal accounts, rather than large institutional or media accounts, dominate fake news conversations. While larger accounts have broader reach, they are not as influential in controlling information flow. The primary topics driving fake news discussions are political and religious, with a clear focus on the upcoming elections in both countries. This study contributes to both theory and practice by offering insights into the misinformation landscape in non-Western contexts and suggesting future research directions, including surveys and platform-specific studies. Additionally, it provides practical implications for policymakers seeking to combat misinformation through enhanced digital literacy and regulatory frameworks.

本研究调查了非西方背景下（来自印度和孟加拉国的数据）X（前身为Twitter）上假新闻传播的把关情况，确定了在线假新闻生态系统中有影响力的参与者。该研究以级联理论为基础，运用网络分析来研究假新闻是如何在这些政治和宗教错误信息普遍存在的国家传播的。该研究使用了从2024年4月起引用“假新闻”一词的100001个帖子的数据，基于中间性中心性、特征向量中心性和模块化类分析了关键账户。该研究采用计算方法和定性文本分析，将网络结构、关系和主题趋势可视化。调查结果显示，主导假新闻对话的是个人账户，而不是大型机构或媒体账户。虽然大型账户的影响范围更广，但它们在控制信息流方面的影响力较小。推动假新闻讨论的主要话题是政治和宗教，显然关注的是这两个国家即将举行的选举。本研究通过深入了解非西方背景下的错误信息格局，并提出未来的研究方向，包括调查和特定平台的研究，对理论和实践都有贡献。此外，它还为寻求通过加强数字素养和监管框架打击错误信息的政策制定者提供了实际意义。

{"title":"Mapping misinformation gatekeepers in non-western contexts: A computational network analysis of fake news on X (Twitter) using Gephi","authors":"Moumita Roy, Maria DeRoche","doi":"10.1016/j.teler.2025.100214","DOIUrl":"10.1016/j.teler.2025.100214","url":null,"abstract":"<div><div>This study investigates the gatekeeping of fake news dissemination on X (formerly Twitter) in non-Western contexts (data from India and Bangladesh), identifying influential actors in the online fake news ecosystem. Grounded in cascade theory, the research applies network analysis to examine how fake news spreads in these countries, where political and religious misinformation is prevalent. Using data from 10,001 X posts from April 2024 that reference the term \"fake news,\" the study analyzes key accounts based on betweenness centrality, eigenvector centrality, and modularity class. Employing both computational methods and qualitative text analysis, the study visualizes network structures, relationships, and topic trends. Findings reveal that individual personal accounts, rather than large institutional or media accounts, dominate fake news conversations. While larger accounts have broader reach, they are not as influential in controlling information flow. The primary topics driving fake news discussions are political and religious, with a clear focus on the upcoming elections in both countries. This study contributes to both theory and practice by offering insights into the misinformation landscape in non-Western contexts and suggesting future research directions, including surveys and platform-specific studies. Additionally, it provides practical implications for policymakers seeking to combat misinformation through enhanced digital literacy and regulatory frameworks.</div></div>","PeriodicalId":101213,"journal":{"name":"Telematics and Informatics Reports","volume":"18 ","pages":"Article 100214"},"PeriodicalIF":0.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144178019","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Bridging the gray digital divide: A cross-cultural qualitative study on digital inclusion and healthy aging in Germany, Japan, and Thailand 弥合灰色数字鸿沟：德国、日本和泰国数字包容与健康老龄化的跨文化定性研究

Telematics and Informatics Reports

Pub Date : 2025-06-01 DOI: 10.1016/j.teler.2025.100218

I-Chun Chen , Anuchit Phanumartwiwath

This study explores the global challenge of digital inclusion for healthy aging through a cross-cultural analysis of Germany, Japan, and Thailand. Using semi-structured interviews with twenty-four stakeholders and thematic analysis via ATLAS.ti 9.0 software, we identify three core themes: “Digital Literacy and Barriers to Inclusion,” “Facilitators of Digital Inclusion,” and “Community-Based Health and Technology Support.” Findings reveal that socio-cultural, psychological, and infrastructural factors shape distinct regional approaches: Germany's community-based 'Zentrum Plus' centers, Japan's volunteer networks, and Thailand's government-led initiatives. These case studies demonstrate the necessity of culturally sensitive strategies to bridge the “gray digital divide.” By examining how digital media cultures (i.e., technology adoption patterns shaped by sociocultural norms) mediate psychosocial outcomes (e.g., self-efficacy, social connectivity) in aging populations, the study contributes to interdisciplinary research at the intersection of gerontechnology, media studies, and public health. Specifically, we advance two frameworks: 1) a cultural relativity model of digital literacy acquisition and 2) a psychosocial impact matrix linking technology access to mental well-being metrics. Policy recommendations include public-private partnerships, volunteer-driven literacy programs, and rural digital infrastructure investment.

本研究通过对德国、日本和泰国的跨文化分析，探讨了数字包容促进健康老龄化的全球挑战。采用与24个利益相关者的半结构化访谈和ATLAS的专题分析。在9.0软件中，我们确定了三个核心主题：“数字素养和包容的障碍”、“数字包容的促进者”和“基于社区的卫生和技术支持”。研究结果表明，社会文化、心理和基础设施因素形成了不同的区域方法：德国以社区为基础的“中心+”中心，日本的志愿者网络，以及泰国政府主导的倡议。这些案例研究表明，有必要采取文化敏感策略来弥合“灰色数字鸿沟”。通过研究数字媒体文化（即由社会文化规范形成的技术采用模式）如何调节老龄化人口的心理社会结果（如自我效能感、社会联系），该研究有助于老年技术、媒体研究和公共卫生交叉领域的跨学科研究。具体而言，我们提出了两个框架：1)数字素养习得的文化相关性模型和2)将技术获取与心理健康指标联系起来的社会心理影响矩阵。政策建议包括公私伙伴关系、志愿者驱动的扫盲计划和农村数字基础设施投资。

{"title":"Bridging the gray digital divide: A cross-cultural qualitative study on digital inclusion and healthy aging in Germany, Japan, and Thailand","authors":"I-Chun Chen , Anuchit Phanumartwiwath","doi":"10.1016/j.teler.2025.100218","DOIUrl":"10.1016/j.teler.2025.100218","url":null,"abstract":"<div><div>This study explores the global challenge of digital inclusion for healthy aging through a cross-cultural analysis of Germany, Japan, and Thailand. Using semi-structured interviews with twenty-four stakeholders and thematic analysis via ATLAS.ti 9.0 software, we identify three core themes: “Digital Literacy and Barriers to Inclusion,” “Facilitators of Digital Inclusion,” and “Community-Based Health and Technology Support.” Findings reveal that socio-cultural, psychological, and infrastructural factors shape distinct regional approaches: Germany's community-based 'Zentrum Plus' centers, Japan's volunteer networks, and Thailand's government-led initiatives. These case studies demonstrate the necessity of culturally sensitive strategies to bridge the “gray digital divide.” By examining how digital media cultures (i.e., technology adoption patterns shaped by sociocultural norms) mediate psychosocial outcomes (e.g., self-efficacy, social connectivity) in aging populations, the study contributes to interdisciplinary research at the intersection of gerontechnology, media studies, and public health. Specifically, we advance two frameworks: 1) a cultural relativity model of digital literacy acquisition and 2) a psychosocial impact matrix linking technology access to mental well-being metrics. Policy recommendations include public-private partnerships, volunteer-driven literacy programs, and rural digital infrastructure investment.</div></div>","PeriodicalId":101213,"journal":{"name":"Telematics and Informatics Reports","volume":"18 ","pages":"Article 100218"},"PeriodicalIF":0.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144222651","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Virtual self care: Using virtual reality to support adolescent mental health and wellbeing 虚拟自我保健：利用虚拟现实支持青少年心理健康和幸福

Telematics and Informatics Reports

Pub Date : 2025-06-01 DOI: 10.1016/j.teler.2025.100217

Caleb J. Fahey

This research looked at how virtual reality (VR) could be used as a tool to help improve mental health among adolescents. Given the rising mental health issues in this age group, this research set out to see how effective VR could be in tackling these problems. The study used a mixed-methods approach, which included quantitative tests in the initial screening phase and qualitative interviews during the testing of VR. It aimed to understand how VR interventions affected stress relief, emotional control, and overall mental health in adolescent users. By looking closely at the personal experiences during VR use, this research wanted to explain how VR might provide therapeutic benefits for these young people. The findings highlight the potential for using VR in mental health support for adolescents and provide useful ideas on how to improve VR settings to better support mental well-being in schools.

这项研究着眼于如何将虚拟现实（VR）作为一种工具来帮助改善青少年的心理健康。鉴于这个年龄段的心理健康问题日益严重，这项研究旨在了解VR在解决这些问题方面的效果。该研究采用混合方法，包括初始筛选阶段的定量测试和VR测试期间的定性访谈。它旨在了解虚拟现实干预如何影响青少年用户的压力缓解、情绪控制和整体心理健康。通过仔细观察虚拟现实使用过程中的个人经历，这项研究想要解释虚拟现实如何为这些年轻人提供治疗益处。研究结果强调了在青少年心理健康支持中使用VR的潜力，并就如何改善VR环境以更好地支持学校的心理健康提供了有用的想法。

引用次数: 0

Consumers’ Vulnerability and E-commerce purchase behavior: A serial mediation model 消费者脆弱性与电子商务购买行为：一个序列中介模型

Telematics and Informatics Reports

Pub Date : 2025-06-01 DOI: 10.1016/j.teler.2025.100215

Thu Thuy Nguyen, Hong Tam Nguyen, Anh Vu Nguyen, Phuong Thao Tran, Hong Nhung Mai, Thi Van Anh Pham

The current study investigates the influence of consumers’ vulnerability on E-commerce purchase behavior, and how Uncertainty and Perceived behavioral control can affect these relationships. While contemporary studies have investigated the effects of consumers’ vulnerability on e-commerce purchase intentions with mixed results, the research formulates a model that explains specifically how two dimensions of consumers’ vulnerability (lack of skills, lack of knowledge), perceived behavioral control, and uncertainty intricately affect each other in determining customers’ e-commerce purchase behavior. The quantitative study uses data from a questionnaire survey with 364 valid responses. Structural equation modeling (SEM) with 5000 bootstrapping replications was employed to test formulated hypotheses. The results indicate that consumers’ vulnerability of capability negatively impacts customers’ perceived behavioral control and positively affects the perceived Uncertainty of E-commerce. In addition, Consumers’ Uncertainty and Perceived behavioral control serve as serial mediators in the relationship between consumers’ vulnerability and e-commerce purchase behavior. These findings offer some new theoretical and practical implications and insights into consumer decision-making under vulnerability and provide actionable guidance for designing more inclusive and trustworthy e-commerce environments.

本研究探讨消费者的脆弱性对电子商务购买行为的影响，以及不确定性和感知行为控制如何影响这些关系。虽然当代研究调查了消费者脆弱性对电子商务购买意愿的影响，结果好坏参半，但本研究制定了一个模型，具体解释了消费者脆弱性（缺乏技能、缺乏知识）、感知行为控制和不确定性这两个维度如何错综复杂地相互影响，以决定消费者的电子商务购买行为。定量研究采用问卷调查的数据，有364个有效回复。采用结构方程模型（SEM）和5000个自举重复来检验制定的假设。结果表明，消费者能力脆弱性对消费者感知行为控制产生负向影响，对消费者感知电子商务不确定性产生正向影响。此外，消费者的不确定性和感知行为控制在消费者脆弱性与电子商务购买行为的关系中起着串行中介作用。这些研究结果为脆弱性下的消费者决策提供了一些新的理论和实践启示，并为设计更具包容性和可信赖性的电子商务环境提供了可操作的指导。

{"title":"Consumers’ Vulnerability and E-commerce purchase behavior: A serial mediation model","authors":"Thu Thuy Nguyen, Hong Tam Nguyen, Anh Vu Nguyen, Phuong Thao Tran, Hong Nhung Mai, Thi Van Anh Pham","doi":"10.1016/j.teler.2025.100215","DOIUrl":"10.1016/j.teler.2025.100215","url":null,"abstract":"<div><div>The current study investigates the influence of consumers’ vulnerability on E-commerce purchase behavior, and how Uncertainty and Perceived behavioral control can affect these relationships. While contemporary studies have investigated the effects of consumers’ vulnerability on e-commerce purchase intentions with mixed results, the research formulates a model that explains specifically how two dimensions of consumers’ vulnerability (lack of skills, lack of knowledge), perceived behavioral control, and uncertainty intricately affect each other in determining customers’ e-commerce purchase behavior. The quantitative study uses data from a questionnaire survey with 364 valid responses. Structural equation modeling (SEM) with 5000 bootstrapping replications was employed to test formulated hypotheses. The results indicate that consumers’ vulnerability of capability negatively impacts customers’ perceived behavioral control and positively affects the perceived Uncertainty of E-commerce. In addition, Consumers’ Uncertainty and Perceived behavioral control serve as serial mediators in the relationship between consumers’ vulnerability and e-commerce purchase behavior. These findings offer some new theoretical and practical implications and insights into consumer decision-making under vulnerability and provide actionable guidance for designing more inclusive and trustworthy e-commerce environments.</div></div>","PeriodicalId":101213,"journal":{"name":"Telematics and Informatics Reports","volume":"18 ","pages":"Article 100215"},"PeriodicalIF":0.0,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144240662","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DDoS attack detection using CNN-BiLSTM with attention mechanism 基于关注机制的CNN-BiLSTM DDoS攻击检测

Telematics and Informatics Reports

Pub Date : 2025-05-26 DOI: 10.1016/j.teler.2025.100211

Ashfaq Ahmad Najar , S. Manohar Naik

In today’s interconnected world, nearly every aspect of modern life relies on internet-based services. However, the widespread adoption of these services has led to a significant increase in cyberattacks, particularly Distributed Denial of Service (DDoS) attacks—which pose a major threat to the stability of critical infrastructure. Therefore, timely and accurate detection of DDoS attacks is essential for cybersecurity. Although deep learning has shown promise in intrusion detection, existing systems often struggle with multiclass DDoS classification, class imbalance, and adaptability to evolving traffic patterns—especially in resource-constrained IoT and IIoT environments. To address these challenges, this study proposes a novel lightweight hybrid deep learning model that integrates Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory (BiLSTM) networks, and an attention mechanism to enhance feature extraction and improve classification performance. The model is evaluated on two benchmark datasets, CICDDoS2019 and Edge-IIoT, achieving classification accuracies of 99.78% and 98.84%, and ROC-AUC scores of 100% and 99.98%, respectively. Comparative analysis with baseline models and prior studies highlights the model’s robustness, efficiency, and suitability for real-time deployment in diverse and constrained environments.

在当今互联互通的世界中，现代生活的几乎每个方面都依赖于基于互联网的服务。然而，这些服务的广泛采用导致了网络攻击的显著增加，特别是分布式拒绝服务（DDoS）攻击，这对关键基础设施的稳定性构成了重大威胁。因此，及时准确地检测DDoS攻击对网络安全至关重要。尽管深度学习在入侵检测方面已经显示出前景，但现有系统经常与多类DDoS分类、类不平衡以及对不断变化的流量模式的适应性作斗争，特别是在资源受限的物联网和工业物联网环境中。为了解决这些挑战，本研究提出了一种新的轻量级混合深度学习模型，该模型集成了卷积神经网络（CNN）、双向长短期记忆（BiLSTM）网络和注意机制，以增强特征提取并提高分类性能。在CICDDoS2019和Edge-IIoT两个基准数据集上对该模型进行了评估，分类准确率分别为99.78%和98.84%，ROC-AUC得分分别为100%和99.98%。与基线模型和先前研究的对比分析突出了该模型的鲁棒性、效率以及在各种受限环境下实时部署的适用性。

{"title":"DDoS attack detection using CNN-BiLSTM with attention mechanism","authors":"Ashfaq Ahmad Najar , S. Manohar Naik","doi":"10.1016/j.teler.2025.100211","DOIUrl":"10.1016/j.teler.2025.100211","url":null,"abstract":"<div><div>In today’s interconnected world, nearly every aspect of modern life relies on internet-based services. However, the widespread adoption of these services has led to a significant increase in cyberattacks, particularly Distributed Denial of Service (DDoS) attacks—which pose a major threat to the stability of critical infrastructure. Therefore, timely and accurate detection of DDoS attacks is essential for cybersecurity. Although deep learning has shown promise in intrusion detection, existing systems often struggle with multiclass DDoS classification, class imbalance, and adaptability to evolving traffic patterns—especially in resource-constrained IoT and IIoT environments. To address these challenges, this study proposes a novel lightweight hybrid deep learning model that integrates Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory (BiLSTM) networks, and an attention mechanism to enhance feature extraction and improve classification performance. The model is evaluated on two benchmark datasets, CICDDoS2019 and Edge-IIoT, achieving classification accuracies of 99.78% and 98.84%, and ROC-AUC scores of 100% and 99.98%, respectively. Comparative analysis with baseline models and prior studies highlights the model’s robustness, efficiency, and suitability for real-time deployment in diverse and constrained environments.</div></div>","PeriodicalId":101213,"journal":{"name":"Telematics and Informatics Reports","volume":"18 ","pages":"Article 100211"},"PeriodicalIF":0.0,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144146830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

How gamification influences e-commerce behavior in hardcore vs. casual gamers? 游戏化如何影响硬核和休闲玩家的电子商务行为？

Telematics and Informatics Reports

Pub Date : 2025-05-15 DOI: 10.1016/j.teler.2025.100212

Junialdi Dwijaputra , Diana Purwitasari , Ellya Zulaikha

In recent years, e-commerce transactions have experienced a positive growth trend, with gamification emerging as a widely adopted feature to enhance user engagement and business outcomes. Despite its increasing implementation, the effectiveness of gamification in e-commerce requires further investigation. Grounded in the Theory of Planned Behavior (TPB), this study examines how attitudes, subjective norms, and perceived behavioral control influence user behavior in gamified e-commerce environments. Specifically, this research explores the impact of gamification design on user acquisition, engagement, and purchasing behaviors by considering different player types. To bridge existing research gaps, this study proposes several hypotheses: player types positively affect gamification design; gamification design positively influences acquisition, engagement, and purchase in expected business behavior; and gamification design, influenced by player type, positively impacts acquisition, engagement, and purchase behaviors. Using data collected from 302 e-commerce users in Indonesia, this study employs structural equation modeling (SEM) to analyze the relationships between gamification design, user behavior, and business performance. Additionally, a multi-group analysis differentiating hardcore and casual player types is conducted to explore variations in acquisition, engagement, and purchasing behaviors. The empirical findings confirm that gamification strategies tailored to specific player types significantly enhance e-commerce business objectives. Specifically, hardcore players exhibit a stronger response to gamification design in terms of acquisition, engagement, and purchase behaviors than casual players. These insights highlight the critical role of player-centric gamification strategies in optimizing business performance and user experience.

近年来，电子商务交易经历了一个积极的增长趋势，游戏化成为一个广泛采用的功能，以提高用户参与度和业务成果。尽管游戏化在电子商务中的应用越来越多，但其有效性还有待进一步研究。本研究以计划行为理论（Theory of Planned Behavior， TPB）为基础，探讨在游戏化电子商务环境中，态度、主观规范和感知行为控制如何影响用户行为。具体来说，本研究通过考虑不同的玩家类型，探讨了游戏化设计对用户获取、用户粘性和购买行为的影响。为了弥补现有的研究差距，本研究提出了几个假设：玩家类型会积极影响游戏化设计；游戏化设计积极影响用户获取、用户粘性和预期商业行为中的购买行为；受玩家类型影响的游戏化设计会对用户获取、用户粘性和购买行为产生积极影响。本研究采用结构方程模型（SEM）分析了游戏化设计、用户行为和商业绩效之间的关系，并收集了印尼302名电子商务用户的数据。此外，我们还进行了多群体分析，区分硬核和休闲玩家类型，以探索用户获取、用户粘性和购买行为的差异。实证结果证实，为特定玩家类型量身定制的游戏化策略显著提高了电子商务业务目标。具体来说，硬核玩家在获取、粘性和购买行为方面对游戏化设计的反应比休闲玩家更强烈。这些见解强调了以玩家为中心的游戏化策略在优化业务绩效和用户体验方面的关键作用。

{"title":"How gamification influences e-commerce behavior in hardcore vs. casual gamers?","authors":"Junialdi Dwijaputra , Diana Purwitasari , Ellya Zulaikha","doi":"10.1016/j.teler.2025.100212","DOIUrl":"10.1016/j.teler.2025.100212","url":null,"abstract":"<div><div>In recent years, e-commerce transactions have experienced a positive growth trend, with gamification emerging as a widely adopted feature to enhance user engagement and business outcomes. Despite its increasing implementation, the effectiveness of gamification in e-commerce requires further investigation. Grounded in the Theory of Planned Behavior (TPB), this study examines how attitudes, subjective norms, and perceived behavioral control influence user behavior in gamified e-commerce environments. Specifically, this research explores the impact of gamification design on user acquisition, engagement, and purchasing behaviors by considering different player types. To bridge existing research gaps, this study proposes several hypotheses: player types positively affect gamification design; gamification design positively influences acquisition, engagement, and purchase in expected business behavior; and gamification design, influenced by player type, positively impacts acquisition, engagement, and purchase behaviors. Using data collected from 302 e-commerce users in Indonesia, this study employs structural equation modeling (SEM) to analyze the relationships between gamification design, user behavior, and business performance. Additionally, a multi-group analysis differentiating hardcore and casual player types is conducted to explore variations in acquisition, engagement, and purchasing behaviors. The empirical findings confirm that gamification strategies tailored to specific player types significantly enhance e-commerce business objectives. Specifically, hardcore players exhibit a stronger response to gamification design in terms of acquisition, engagement, and purchase behaviors than casual players. These insights highlight the critical role of player-centric gamification strategies in optimizing business performance and user experience.</div></div>","PeriodicalId":101213,"journal":{"name":"Telematics and Informatics Reports","volume":"18 ","pages":"Article 100212"},"PeriodicalIF":0.0,"publicationDate":"2025-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144137901","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0