Artificial Intelligence Review最新文献

{"title":"Face-based machine learning diagnostics: applications, challenges and opportunities","authors":"Jie Song,&nbsp;Mengqiao He,&nbsp;Xin Zheng,&nbsp;Yuxin Zhang,&nbsp;Cheng Bi,&nbsp;Jinhua Feng,&nbsp;Jiale Du,&nbsp;Hang Li,&nbsp;Bairong Shen","doi":"10.1007/s10462-025-11246-2","DOIUrl":"10.1007/s10462-025-11246-2","url":null,"abstract":"<div><p>Traditional medical diagnostic methods face bottlenecks such as high cost, poor accessibility, and delayed diagnosis in genetic syndromes, neurological disorders, psychiatric disorders, and endocrine disorders. Face-based machine learning (ML) technology provides a new path for early screening of diseases by analyzing facial phenotypes, dynamic expressions, facial skin, and 3D structural abnormalities, and is gradually becoming a clinically assisted screening tool. This paper provides a comprehensive overview of the applications, advances, and challenges of the technology. We summarize the range of diseases for which facial diagnosis is applicable and describe the basic process and related techniques for face-based ML diagnostic systems. In addition, this paper organizes the resources of current publicly available facial medical datasets and clarifies their disease coverage and sample size. Finally, possible future solutions to challenges hindering widespread adoption in clinical practice such as data bias, privacy, interpretability, generalizability, clinical value, and resource constraints are discussed. This review aims to provide researchers with a comprehensive foundation that integrates clinical perspectives, technological insights, and practical resources, to facilitate the development and successful implementation of face-based ML diagnostics in real-world clinical practice.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 8","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11246-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143938333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning model inversion attacks and defenses: a comprehensive survey","authors":"Wencheng Yang,&nbsp;Song Wang,&nbsp;Di Wu,&nbsp;Taotao Cai,&nbsp;Yanming Zhu,&nbsp;Shicheng Wei,&nbsp;Yiying Zhang,&nbsp;Xu Yang,&nbsp;Zhaohui Tang,&nbsp;Yan Li","doi":"10.1007/s10462-025-11248-0","DOIUrl":"10.1007/s10462-025-11248-0","url":null,"abstract":"<div><p>The rapid adoption of deep learning in sensitive domains has brought tremendous benefits. However, this widespread adoption has also given rise to serious vulnerabilities, particularly model inversion (MI) attacks, posing a significant threat to the privacy and integrity of personal data. The increasing prevalence of these attacks in applications such as biometrics, healthcare, and finance has created an urgent need to understand their mechanisms, impacts, and defense methods. This survey aims to fill the gap in the literature by providing a structured and in-depth review of MI attacks and defense strategies. Our contributions include a systematic taxonomy of MI attacks, extensive research on attack techniques and defense mechanisms, and a discussion about the challenges and future research directions in this evolving field. By exploring the technical and ethical implications of MI attacks, this survey aims to offer insights into the impact of AI-powered systems on privacy, security, and trust. In conjunction with this survey, we have developed a comprehensive repository to support research on MI attacks and defenses. The repository includes state-of-the-art research papers, datasets, evaluation metrics, and other resources to meet the needs of both novice and experienced researchers interested in MI attacks and defenses, as well as the broader field of AI security and privacy. The repository will be continuously maintained to ensure its relevance and utility. It is accessible at https://github.com/overgter/Deep-Learning-Model-Inversion-Attacks-and-Defenses.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 8","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11248-0.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143938200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Groundbreaking taxonomy of metaverse characteristics","authors":"Abolghasem Sadeghi-Niaraki,&nbsp;Fatema Rahimi,&nbsp;Nur Alya Emanuelle Binti Azlan,&nbsp;Houbing Song,&nbsp;Farman Ali,&nbsp;Soo-Mi Choi","doi":"10.1007/s10462-025-11243-5","DOIUrl":"10.1007/s10462-025-11243-5","url":null,"abstract":"<div><p>The Metaverse, a dynamic and immersive virtual realm, has captured the imagination of researchers and enthusiasts worldwide. This survey paper aims to introduce a groundbreaking taxonomy for the characteristics of the Metaverse offering a structured and adaptable framework that extends beyond existing categorizations by incorporating dynamic transformations. Unlike prior taxonomies, which often focus on fixed attributes, our approach emphasizes the dynamic evolution of Metaverse characteristics. Through an extensive review of published literature, this study explores key technological, social, economic, and ethical dimensions of the Metaverse. It introduces a process-oriented classification based on 23 distinct characteristics, including immersification, spatiotemporalification, interactification, persistentification, presentification, personification, unification, imaginification, economification, uncertaintification, and credification. By mapping these evolving aspects, we provide a structured and future-proof foundation for understanding the Metaverse’s continuous development. This survey establishes a new standard for comprehensiveness and innovation, shedding light on the diverse facets that have been explored in literature. Through this novel taxonomy, we provide a detailed map of the current landscape and offer insights that pave the way for future research and development in this burgeoning digital frontier.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 8","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11243-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143938479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Decoupling feature-driven and multimodal fusion attention for clothing-changing person re-identification","authors":"Yongkang Ding,&nbsp;Xiaoyin Wang,&nbsp;Hao Yuan,&nbsp;Meina Qu,&nbsp;Xiangzhou Jian","doi":"10.1007/s10462-025-11250-6","DOIUrl":"10.1007/s10462-025-11250-6","url":null,"abstract":"<div><p>Person Re-Identification (ReID) plays a crucial role in intelligent surveillance, public safety, and intelligent transportation systems. However, clothing variation remains a significant challenge in this field. To address this issue, this paper introduces a method named Decoupling Feature-Driven and Multimodal Fusion Attention for Clothing-Changing Person Re-Identification (DM-ReID). The proposed approach employs a dual-stream feature extraction framework, consisting of a global RGB image feature stream and a clothing-irrelevant feature enhancement stream. These streams respectively capture comprehensive appearance information and identity features independent of clothing. Additionally, two feature fusion strategies are proposed: firstly, an initial fusion of RGB features and clothing-irrelevant features is achieved through the Hadamard product in the mid-network stage to enhance feature complementarity; secondly, a multimodal fusion attention mechanism is integrated at the network’s end to dynamically adjust feature weights, further improving feature representation capabilities. To optimize model performance, a composite loss function combining identity loss and triplet loss is utilized, effectively enhancing the model’s discriminative ability and feature distinctiveness. Experimental results on multiple public datasets, including PRCC, LTCC, and VC-Clothes, demonstrate that DM-ReID surpasses most existing mainstream methods in Rank-1 accuracy and mean Average Precision (mAP) metrics under clothing-changing scenarios. These findings validate the method’s effectiveness and robustness in handling complex clothing variations, highlighting its promising prospects for practical applications.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 8","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11250-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143938249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recent trends on mammogram breast density analysis using deep learning models: neoteric review","authors":"S. Jeba Prasanna Idas,&nbsp;K. Hemalatha,&nbsp;Jayakumar Naveenkumar,&nbsp;T. Joshva Devadas","doi":"10.1007/s10462-025-11232-8","DOIUrl":"10.1007/s10462-025-11232-8","url":null,"abstract":"<div><p>Breast cancer is a globally prevalent and potentially fatal illness affecting women. Timely identification of screening mammography may decrease the occurrence of incorrect positive results and enhance the rate of patient survival. Nevertheless, the density of breast tissue in mammograms can impact the precision and effectiveness of detecting breast cancer. This paper examines the existing body of research on the analysis of breast density in mammograms utilising advanced deep learning models, including convolutional neural networks (CNN), transfer learning (TL), and ensemble learning (EL). Additionally, it examines various datasets and evaluation measures employed in the investigations. The study demonstrates that deep learning models can attain exceptional accuracy in categorising breast density. However, they encounter obstacles such as limited data availability, intricate model structures, and difficulties in interpreting the results. The research asserts that categorising breast density is an essential undertaking in order to enhance the identification and survival rates of breast cancer. Further investigation is warranted to examine the most effective deep learning structures, data augmentation methods, and interpretable models for this undertaking.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 8","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11232-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143929928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine and deep learning for personality traits detection: a comprehensive survey and open research challenges","authors":"Anam Naz,&nbsp;Hikmat Ullah Khan,&nbsp;Amal Bukhari,&nbsp;Bader Alshemaimri,&nbsp;Ali Daud,&nbsp;Muhammad Ramzan","doi":"10.1007/s10462-025-11245-3","DOIUrl":"10.1007/s10462-025-11245-3","url":null,"abstract":"<div><p>Natural language processing (NLP), a prominent research domain of Artificial Intelligence (AI), analyzes users’ generated content on social media for various purposes like sentiment analysis, text summarization, chatbots, fake news detection, etc. Recent advancements in NLP have helped for analysis of human behavior analysis and predicting various human personality traits. Understanding personality traits has long been a fundamental pursuit in psychology and cognitive sciences due to its vast applications for understanding from individuals to social dynamics. Due to online social platforms where people express their views, experiences and comments, NLP is applied for users’ behavior and personality analysis, which is helpful in defining marketing strategies, consumers’ behavior analysis, team building, etc. This research study provides a comprehensive overview of existing methodologies, applications, and challenges in the field of personality traits detection using shallow machine learning, ensemble learning and deep learning. To conduct this study, recent research publications relevant to NLP for this new but emerging research domain are reviewed. The background knowledge of personality models of various nature is discussed for better domain understanding. The study encompasses machine learning and deep learning models with thorough analysis of traditional and innovative techniques including ensemble learning and transformer-based models in chronological order highlighting the trend analysis showing evolution of application of advanced methods. The review also presents and compares the widely used datasets which may guide the researchers for selection of datasets in future studies. Performance evaluation metrics have been discussed which are used in the relevant literature. Furthermore, it explores the application of research of personality traits detection in various domains highlighting its significance. We have also carried out extensive empirical analysis using conventional textual to advanced deep embedding features and applying machine learning, ensemble learning and deep learning algorithms. Finally, before conclusion, the review highlights the open research issues and challenges as potential future directions for the researchers.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 8","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11245-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143925442","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rethinking and recomputing the value of machine learning models","authors":"Burcu Sayin,&nbsp;Jie Yang,&nbsp;Xinyue Chen,&nbsp;Andrea Passerini,&nbsp;Fabio Casati","doi":"10.1007/s10462-025-11242-6","DOIUrl":"10.1007/s10462-025-11242-6","url":null,"abstract":"<div><p>In this paper, we argue that the prevailing approach to training and evaluating machine learning models often fails to consider their real-world application within organizational or societal contexts, where they are intended to create beneficial value for people. We propose a shift in perspective, redefining model assessment and selection to emphasize integration into workflows that combine machine predictions with human expertise, particularly in scenarios requiring human intervention for low-confidence predictions. Traditional metrics like accuracy and f-score fail to capture the beneficial value of models in such hybrid settings. To address this, we introduce a simple yet theoretically sound “value” metric that incorporates task-specific costs for correct predictions, errors, and rejections, offering a practical framework for real-world evaluation. Through extensive experiments, we show that existing metrics fail to capture real-world needs, often leading to suboptimal choices in terms of value when used to rank classifiers. Furthermore, we emphasize the critical role of calibration in determining model value, showing that simple, well-calibrated models can often outperform more complex models that are challenging to calibrate.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 8","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11242-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143919180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advanced unsupervised learning: a comprehensive overview of multi-view clustering techniques","authors":"Abdelmalik Moujahid,&nbsp;Fadi Dornaika","doi":"10.1007/s10462-025-11240-8","DOIUrl":"10.1007/s10462-025-11240-8","url":null,"abstract":"<div><p>Machine learning techniques face numerous challenges to achieve optimal performance. These include computational constraints, the limitations of single-view learning algorithms and the complexity of processing large datasets from different domains, sources or views. In this context, multi-view clustering (MVC), a class of unsupervised multi-view learning, emerges as a powerful approach to overcome these challenges. MVC compensates for the shortcomings of single-view methods and provides a richer data representation and effective solutions for a variety of unsupervised learning tasks. In contrast to traditional single-view approaches, the semantically rich nature of multi-view data increases its practical utility despite its inherent complexity. This survey makes a threefold contribution: (1) a systematic categorization of multi-view clustering methods into well-defined groups, including co-training, co-regularization, subspace, deep learning, kernel-based, anchor-based, and graph-based strategies; (2) an in-depth analysis of their respective strengths, weaknesses, and practical challenges, such as scalability and incomplete data; and (3) a forward-looking discussion of emerging trends, interdisciplinary applications, and future directions in MVC research. This study represents an extensive workload, encompassing the review of over 140 foundational and recent publications, the development of comparative insights on integration strategies such as early fusion, late fusion, and joint learning, and the structured investigation of practical use cases in the areas of healthcare, multimedia, and social network analysis. By integrating these efforts, this work aims to fill existing gaps in MVC research and provide actionable insights for the advancement of the field.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 8","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11240-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143919100","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Generative AI revolution in cybersecurity: a comprehensive review of threat intelligence and operations","authors":"Mueen Uddin,&nbsp;Muhammad Saad Irshad,&nbsp;Irfan Ali Kandhro,&nbsp;Fuhid Alanazi,&nbsp;Fahad Ahmed,&nbsp;Muhammad Maaz,&nbsp;Saddam Hussain,&nbsp;Syed Sajid Ullah","doi":"10.1007/s10462-025-11219-5","DOIUrl":"10.1007/s10462-025-11219-5","url":null,"abstract":"<div><p>Cyber threats are increasingly frequent in today’s world, posing challenges for organizations and individuals to protect their data from cybercriminals. On the other hand, Generative Artificial Intelligence (GAI) technology offers an efficient way to automatically address these issues with the help of AI models and algorithms. It can work on more critical security aspects where human intervention is required and handle everyday threat situations autonomously. This research paper explores GAI in enhancing cybersecurity by leveraging AI Models and algorithms. GAI can autonomously address common security issues, detect novel threats, and augment human intervention in critical security aspects. Moreover, this research study also highlights autonomous security enhancements, improved security posture against emerging threats, anomaly detection, and threat response. Besides this, we have discussed the GAI limitations, such as occasional incorrect results, expensive training, and the potential for misuse by malicious actors for illegal activities. This research study also provides valuable insights into the balanced adoption of GAI in cybersecurity, ensuring effective threat migration without compromising system integrity.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 8","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11219-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143919101","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-object tracking review: retrospective and emerging trend","authors":"Zhiyu Guan,&nbsp;Zhaofa Wang,&nbsp;Gan Zhang,&nbsp;Luwei Li,&nbsp;Miaomiao Zhang,&nbsp;Zhiping Shi,&nbsp;Na Jiang","doi":"10.1007/s10462-025-11212-y","DOIUrl":"10.1007/s10462-025-11212-y","url":null,"abstract":"<div><p>Multi-object tracking (MOT) is a critical task involving detecting and continuously tracking multiple objects within a video sequence. It is widely used in various fields, such as autonomous driving and intelligent security. In recent years, deep learning architectures have effectively promoted the development of MOT. However, this task poses significant challenges regarding accuracy due to occlusion/truncation, light variation, camera movement. Researchers have proposed many methods to address these issues to reduce trajectory fragmentation, identity switches, and missing targets. To better understand these advancements, it is essential to categorize the approaches based on their methodologies. This article reviewed the recent development of MOT, divided into Tracking by Detection (TBD) and End-to-End (E2E). By introducing and comparing the two types of tracking algorithms, readers can quickly understand the current development status of MOT. Meanwhile, this review summarizes the links to open-source code of excellent algorithms and common benchmark datasets in the appendix. And provide a unified MOT toolkit that includes evaluation and visualization at https://github.com/guanzhiyu817/MOT-tools. In addition, this review discusses the future directions of MOT, specifically cross-modal reasoning.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 8","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11212-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143919103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}