WIREs Data Mining and Knowledge Discovery最新文献

{"title":"Data as an Asset: Economic Traits, Market Dynamics, and Monetization Challenges","authors":"Petter Kvalvik, Sanjay Misra, Biswajeeban Mishra, Satyen Dash","doi":"10.1002/widm.70072","DOIUrl":"https://doi.org/10.1002/widm.70072","url":null,"abstract":"The digital economy has established data as a fundamental asset, shaping industries, governance, and technological innovation while driving competitive advantage across sectors. The proliferation of data-generating technologies like IoT and social media has transformed data into a strategic resource, particularly benefiting Big Tech companies. This study examines data's dual nature as both asset and commodity, analyzing market structures, monetization challenges, and emerging technology impacts on the data economy. Employing a multivocal literature review (MLR) approach, this research systematically reviewed academic literature and gray sources including white papers and industry reports. Comprehensive searches across five major databases such as ACM digital library, IEEE Xplore, ScienceDirect, SpringerLink, and Wiley Online Library yielded 764 records, with an additional 86 from other sources (i.e., gray-literature), totaling 850 screened studies. Ultimately, 64 studies addressed five key research questions on data's economic characteristics, market structures, monetization challenges, technological impacts, and public-commercial balance frameworks. Analysis reveals data's distinctive economic properties like non-rivalry, non-excludability, and scalability necessitating novel valuation approaches. Market dynamics are significantly influenced by regulatory frameworks like GDPR, while emerging technologies such as AI and blockchain reshape monetization strategies. Key challenges include accurate valuation, data quality assurance, and privacy compliance. Data's economic essence defies conventional valuation frameworks due to its unique properties. AI and blockchain are transforming data markets through enhanced value extraction and secure transactions. However, robust governance frameworks like data commons and trusts are essential for balancing monetization with ethical use, requiring future research on comprehensive valuation models integrating technological advancements with public interest alignment.","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"52 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-02-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147279278","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}

{"title":"Speech Emotion Recognition Using Transfer Learning: A Comparative Study","authors":"Yunus Korkmaz","doi":"10.1002/widm.70073","DOIUrl":"https://doi.org/10.1002/widm.70073","url":null,"abstract":"Speech emotion recognition (SER) is a critical research area at the intersection of affective computing and audio signal processing. Traditional approaches often require extensive manual feature engineering and large labeled datasets, which can limit performance in real‐world applications. Recently, transfer learning with pretrained audio neural networks has gained significant traction for SER, leveraging knowledge from large‐scale audio corpora to improve recognition accuracy and generalization. This review presents a comprehensive overview of transfer learning‐based SER systems, with a particular focus on pretrained audio models such as YAMNet, VGGish, Wav2Vec2, and so on. Two main paradigms were examined in this study. First, feature embedding extraction, where pretrained models serve as fixed feature encoders. Second, fine‐tuning strategies, where model weights are partially or fully updated on emotion‐specific corpora. The article categorizes and compares state‐of‐the‐art studies across multiple datasets, including RAVDESS, EmoDB, IEMOCAP, CREMA‐D, and so on, while discussing commonly used evaluation metrics such as accuracy, precision, recall, F1 score, and AUC. Comparative tables are presented to highlight methodological trends and performance differences between feature‐based and fine‐tuned SER systems. This overview aims to guide researchers in selecting appropriate models and strategies for future SER tasks and to identify open challenges and future research directions in transfer learning for emotion recognition from speech signals. This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item> Technologies &gt; Machine Learning </jats:list-item> <jats:list-item> Technologies &gt; Artificial Intelligence </jats:list-item> <jats:list-item> Algorithmic Development &gt; Multimedia </jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"30 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146215795","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}

{"title":"Federated Learning and 5G/6G-Based Internet of Medical Things (IoMT): Applications, Key Enabling Technologies, Open Issues and Future Research Directions","authors":"Abdul Ahad, Kazi Istiaque Ahmed, Farhan Ullah, Muhammad Aman Sheikh, Mohammad Tahir, Mohammad Hayajneh, Ivan Miguel Pires","doi":"10.1002/widm.70065","DOIUrl":"https://doi.org/10.1002/widm.70065","url":null,"abstract":"The rapid expansion of smart healthcare technologies has created a growing need for systems that are not only intelligent and efficient, but also deeply respectful of patient privacy. As medical data becomes increasingly distributed across wearables, hospital networks, home-based sensors, and mobile applications, traditional centralized approaches struggle to keep pace with evolving security, latency, and interoperability demands. In this review, we explore federated learning (FL) as a promising pathway towards decentralized intelligence, one that allows healthcare institutions and Internet of Medical Things (IoMT) devices to collaborate without sharing sensitive patient data. Supported by emerging 5G and 6G communication technologies, FL has the potential to reshape modern healthcare by enabling real-time analytics, reliable remote monitoring, personalized treatment recommendations, and advanced medical diagnosis. High-bandwidth, low-latency networks provide the connectivity backbone required for FL to function smoothly across diverse medical environments. We examine FL's various forms, its integration into IoMT applications, and the role of enabling technologies such as edge computing, Device-to-device (D2D) communication, Massive Machine Type Communication (mMTC), Blockchain, Software Defined Networking (SDN), Network Function Virtualization (NFV), Digital twins, and Fog computing. At the same time, we acknowledge that this integration is far from straightforward. Challenges such as data heterogeneity, communication overhead, model drift, security risks, resource allocation, and clinical interoperability continue to shape the research landscape. By synthesizing current findings, identifying open issues, and outlining future research directions, this review provides clarity and drives forward research efforts within the integrated fields of AI, networking, and digital healthcare.","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"18 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-02-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146215700","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}

{"title":"Connecting the Dots: A Semantic Network Approach to Pediatric iNO Research Discovery","authors":"S. Fernando Gomes, Paulo Cesar Duarte Junior, Shirley Ferraz Crispilho, Hafsa Siddiqui, Jaciara Santos, Kaushik Pal, Aron José Pazin de Andrade, Himanshu Upadhyay","doi":"10.1002/widm.70071","DOIUrl":"https://doi.org/10.1002/widm.70071","url":null,"abstract":"In pediatric cardiac surgery, inhaled nitric oxide (iNO) is commonly used in managing postoperative pulmonary hypertension. However, a comprehensive understanding of its research trajectory remains incomplete. This study addresses that gap by applying advanced computational bibliometric techniques, including lexical network analysis, Latent Dirichlet Allocation (LDA) based topic modeling, Concept Level Entity Extraction (CLEE), and supergraph based semantic mapping, to elucidate the conceptual evolution of iNO research within the clinical domain. We examined 2517 publications sourced from Scopus, Web of Science, and PubMed. A two‐stage deduplication process, initially automated and subsequently refined through manual screening, yielded a curated dataset of 219 unique records. Scopus accounted for approximately 75% of all retrieved entries, a predominance consistent with its broader coverage of pediatric cardiac and perioperative literature, while funding analyses identified China as the primary driver of contemporary iNO scholarship. Lexical co‐occurrence networks revealed stable semantic clusters that remained coherent across robustness checks, and the integrated supergraph captured thematic transitions spanning six historical phases, ranging from early mechanistic explorations to current emphases on patient‐centered outcomes, perioperative optimization and translational readiness. These patterns may help identify where coordinated multicenter research could be most impactful, particularly regarding protocol harmonization, biomarker development and standardized reporting. This work constitutes the first reproducible historical reconstruction of pediatric iNO research using a supergraph based bibliometric framework, providing a descriptive synthesis that highlights knowledge gaps rather than inferring causality, and may support the design of future hypothesis‐driven translational investigations. This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item> Algorithmic Development &gt; Statistics </jats:list-item> <jats:list-item> Application Areas &gt; Science and Technology </jats:list-item> <jats:list-item> Technologies &gt; Machine Learning </jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"11 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-02-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146215701","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}

{"title":"Lung Cancer in the New Era: Trends, Innovations, and Future Recommendations","authors":"Mohd Munazzer Ansari, Shailendra Kumar, Md Belal Bin Heyat, Tariq Umair, Faijan Akhtar, Saba Parveen, Syed Hasan Saeed, Jian Ping Li","doi":"10.1002/widm.70061","DOIUrl":"https://doi.org/10.1002/widm.70061","url":null,"abstract":"Lung cancer remains one of the most serious global health concerns, with patient survival influenced by lifestyle habits, genetic makeup, and environmental conditions. This paper reviews how new advances in artificial intelligence (AI) and machine learning (ML) are changing the way lung cancer is detected, classified, and treated. Modern deep learning (DL) methods, especially convolutional neural networks (CNNs), have shown stronger performance than many traditional diagnostic techniques, particularly when analyzing medical scans such as low‐dose computed tomography (CT) images. These approaches not only support early cancer detection but also enable more precise classification of its subtypes, leading to more reliable outcomes. In addition, AI is increasingly used to design treatment plans tailored to individual patients' needs, taking genetic variations into account. This makes therapies more effective and reduces unnecessary side effects. Despite these advances, several barriers still limit clinical adoption, including differences in available data, legal and regulatory requirements, and privacy‐related ethical issues. By also integrating environmental risk factors, such as long‐term exposure to air pollution, AI systems may more effectively identify high‐risk groups. The review suggests ways to address these barriers and highlights the growing potential of AI to reshape lung cancer care. This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item> Technologies &gt; Computational Intelligence </jats:list-item> <jats:list-item> Application Areas &gt; Health Care </jats:list-item> <jats:list-item> Algorithmic Development &gt; Biological Data Mining </jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"3 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146196612","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}

{"title":"Edge Computing in Healthcare Using Machine Learning: A Systematic Literature Review","authors":"Amir Mashmool, Giorgio Delzanno, Hamid Saadatfar, Aakash Ahmad, Rainer Koschke, Roohallah Alizadehsani, U. Rajendra Acharya, Daniele D'Agostino","doi":"10.1002/widm.70069","DOIUrl":"https://doi.org/10.1002/widm.70069","url":null,"abstract":"Healthcare is rapidly evolving with the integration of machine learning (ML) and edge computing, which enables real‐time data processing and improved patient care. Edge computing plays a critical role by reducing latency and enhancing data privacy, especially in patient monitoring systems. However, limitations such as device resource constraints and security issues persist. This study presents a systematic literature review (SLR) on using ML and edge computing in healthcare, identifying key benefits, challenges, and research trends. This SLR aimed to identify key benefits, challenges, and current research trends. We sourced relevant studies from databases such as IEEE Xplore, ScienceDirect, ACM Digital Library, and so forth. We applied inclusion and exclusion criteria. We also used the snowballing technique to find more relevant studies by checking selected papers' reference lists, ensuring we did not miss any important ones. Finally, 37 papers were selected and analyzed for their methodologies, algorithms, tools, frameworks, data sources, limitations, motivations, and challenges. Findings show a broad use of ML methods such as support vector machines, clustering, and deep learning, with a strong emphasis on data privacy and model performance; many studies employed federated learning and privacy‐preserving techniques to support real‐time decision‐making. Overall, ML and edge computing integration promise to transform healthcare, though challenges remain. Future research should address resource limitations, enhance ML models for edge environments, and develop standardized protocols. This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item> Application Areas &gt; Health Care </jats:list-item> <jats:list-item> Technologies &gt; Machine Learning </jats:list-item> <jats:list-item> Technologies &gt; Cloud Computing </jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"10 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146169799","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}

{"title":"The Evolution of Digital Image Forgery Detection: A Comprehensive Bibliometric Review, Trends, and Future Prospects","authors":"Rupinder Kaur, Raman Kumar, Gagandeep Kaur, M. Chethan, Sehijpal Singh, Anurag Sinha, Pulkit Kumar, Riyam Adnan Hasan, Zainab Ahmed Abass","doi":"10.1002/widm.70066","DOIUrl":"https://doi.org/10.1002/widm.70066","url":null,"abstract":"Ensuring the authenticity of digital images is essential in forensic investigations, media, and scientific research, where these images serve as critical evidence. This necessity leads to the development of digital image forgery detection (DIF). This study reviewed DIF articles from 2005 to 2024. It performs a comprehensive evaluation and bibliometric analysis of DIF methodologies, aiming to uncover trends, technological advancements, and thematic progressions. The study utilized Scopus data to illustrate key DIF methodologies, citation trends, and thematic changes. It offers fresh insights by showcasing the rising prevalence of deep learning‐based DIF techniques post‐2018 and the emergence of hybrid models that integrate traditional and AI‐driven methods to bolster detection robustness and precision. The review highlights that China and the United States are leading the field, with significant contributions from institutions such as the South China University of Technology and the State University of New York at Albany. The bibliometric analysis reveals three key trends: a marked increase in deep learning‐based DIF methods since 2018, indicating a shift away from traditional feature‐based techniques; strengthening collaboration between industry and academia, especially in China and the US, fueling significant advancements; and a heightened focus on real‐world forgeries, such as deepfakes, emphasizing the necessity for more adaptable detection tools. It highlights challenges like the lack of substantial and varied benchmark datasets. This comprehensive study also suggests enhancing DIF accuracy and applicability across different domains. This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item> Algorithmic Development &gt; Multimedia </jats:list-item> <jats:list-item> Technologies &gt; Computational Intelligence </jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"120 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146134171","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}

{"title":"AI Price Tags and Privacy: When Your Data Sets Your Price","authors":"Varda Mone, Abhishek Thommandru, Fazilov Farkhod Maratovich, Khudaykulov Feruzbek Khurramovich, Abzalova Khurshida Mirziyatovna","doi":"10.1002/widm.70070","DOIUrl":"https://doi.org/10.1002/widm.70070","url":null,"abstract":"This study examines personalized algorithmic pricing and consumer protection across three major jurisdictions the United States, European Union, and India analyzing how artificial intelligence‐driven pricing systems challenge traditional regulatory frameworks and threaten consumer autonomy. The research adopts a comparative methodology combining doctrinal legal analysis with empirical examination of enforcement patterns, scrutinizing recent regulatory developments including the EU's Digital Services Act, the US Department of Justice's RealPage litigation, and India's Consumer Protection Act amendments. The central argument demonstrates that transparency‐only approaches prove fundamentally inadequate in addressing algorithmic filter bubbles and market concentration. Evidence from India's fast‐commerce sector reveals sophisticated discrimination patterns, including device‐based pricing differentials and usage‐pattern exploitation, while “hub‐and‐spoke conspiracies” enable algorithmic collusion without explicit coordination between competitors. Key findings of study that existing legal frameworks, designed for pre‐digital markets, cannot effectively address technologically sophisticated forms of consumer harm and market manipulation. The study identifies critical gaps in jurisdictional approaches: India's reactive consumer protection model, the EU's proactive transparency requirements, and the US's antitrust‐centric enforcement. The research proposes moving beyond disclosure paradigms toward “information enrichment” mandates requiring platforms to actively diversify algorithmic recommendations, coupled with user‐controlled choice architectures and structural market reforms. These interventions, aligned with fundamental rights principles requiring states to serve as ultimate guarantors of diversity offering pathways for regulatory frameworks that balance technological innovation with consumer welfare and market competition. This article is categorized under: <jats:list list-type=\"simple\"> <jats:list-item> Commercial, Legal, and Ethical Issues &gt; Legal Issues </jats:list-item> <jats:list-item> Commercial, Legal, and Ethical Issues &gt; Ethical Considerations </jats:list-item> <jats:list-item> Commercial, Legal, and Ethical Issues &gt; Security and Privacy </jats:list-item> </jats:list>","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"9 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146121910","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}

{"title":"Advanced Deep Learning Frameworks for Cross-Modal Information Retrieval: A Comprehensive Review of Techniques, Challenges, and Future Directions","authors":"Aamir Khan, Nisha Chandran S., D. R. Gangodkar","doi":"10.1002/widm.70055","DOIUrl":"https://doi.org/10.1002/widm.70055","url":null,"abstract":"A cross-modal information retrieval (CMIR) has emerged as a pivotal research area, enabling efficient retrieval across diverse data with multiple modalities. With the production of multimodal data, advanced deep learning frameworks have demonstrated significant promise in aligning and mapping heterogeneous data representations into a unified latent space. This review explores the revolution of advanced deep learning techniques in CMIR, highlighting key advancements, methodology, and challenges, especially focusing on intelligent frameworks that leverage architectures such as convolutional neural networks (CNNs), recurrent neural networks (RNNs), Transformers, and generative adversarial networks (GANs), for enhancing semantic alignment and retrieval accuracy. It also discusses challenges such as modality, imbalance, cross-representation, and inter-permeability with other modalities, providing insight into emerging trends such as multi-model, generative AI, autoencoders, and large-scale, pretrained models, by synthesizing recent advancements and identifying research gaps. This review paper aims to provide a foundation for future exploration in intelligent CMIR systems; the findings underscore the transformative latent of advanced deep learning frameworks in addressing the growing demand for accurate and scalable CMIR solutions.","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"53 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146110183","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}

{"title":"A Review of Deep Learning and Large Language Models for Cold Start Problem in Recommender Systems","authors":"Chenlong Liu, Daguang Jiang, Yi Cai, Hui Li","doi":"10.1002/widm.70068","DOIUrl":"https://doi.org/10.1002/widm.70068","url":null,"abstract":"Recommender systems are essential for information filtering but often suffer from the cold start problem caused by limited interaction data. Recent advances in deep learning (DL) and large language models (LLMs) have shown promise, yet systematic analysis of their effectiveness remains scarce. To address this gap, we introduce a paradigm-driven taxonomy that categorizes solutions by their primary source of information: content, structure, transfer, and generation. Within this framework, DL methods have matured in leveraging content and structural information from interaction logs and multimodal data, while LLMs demonstrate advantages in text-rich and data-sparse environments through transfer-based paradigms that exploit semantic understanding and pre-trained knowledge. Furthermore, emerging generative approaches show potential for synthesizing data or relations to alleviate information scarcity. No universal solution exists; effectiveness depends on the dominant paradigm of a given scenario as well as data availability and computational cost. Combining DL and LLM offers substantial opportunities, including enhanced feature representation, data augmentation, and hybrid pipelines. However, research gaps persist, particularly the lack of standardized evaluation metrics and limited exploration of integration strategies. Addressing these challenges through a paradigm-aware perspective could significantly improve the robustness and adaptability of the cold-start recommendation in diverse contexts.","PeriodicalId":501013,"journal":{"name":"WIREs Data Mining and Knowledge Discovery","volume":"79 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146121909","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}