Artificial Intelligence Review最新文献

{"title":"Defending against attacks in deep learning with differential privacy: a survey","authors":"Zhang Xiangfei,&nbsp;Zhang Qingchen","doi":"10.1007/s10462-025-11350-3","DOIUrl":"10.1007/s10462-025-11350-3","url":null,"abstract":"<div><p>Recently, we have witnessed the revolutionary development of deep learning. As the application domain of deep learning has expanded, its privacy risks have attracted attention since deep leaning methods often use private data for training. Some methods for attacking deep learning, such as membership inference attacks, increase the privacy risks of deep learning models. One risk-reducing defensive strategy with great potential is to apply some degree of random perturbation during the training (or other) phase. Therefore, differential privacy, as a privacy protection framework originally designed for publishing data, is widely used to protect the privacy of deep learning models due to its solid mathematical foundation. In this paper, we first introduce several attack methods that threaten deep learning. Then, we systematically review the cross-applications of differential privacy and deep learning to protect deep learning models. We encourage researchers to visually demonstrate the defense effects of their approaches in the literature rather than solely providing rigorous mathematical proofs. In addition to privacy, we also discuss and review the impact of differential privacy on the robustness, overfitting, and fairness of deep neural networks. Finally, we analyze some potential future research directions, highlighting the significant potential for differential privacy to make positive contributions to future deep learning systems.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 11","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11350-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144881013","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":"A Comprehensive Survey on Subspace Clustering: Methods and Applications","authors":"Jianyu Miao,&nbsp;Xiaochan Zhang,&nbsp;Tiejun Yang,&nbsp;Chao Fan,&nbsp;Yingjie Tian,&nbsp;Yong Shi,&nbsp;Mingliang Xu","doi":"10.1007/s10462-025-11349-w","DOIUrl":"10.1007/s10462-025-11349-w","url":null,"abstract":"<div><p>As a pivotal strategy to deal with complicated and high-dimensional data, subspace clustering is to find a set of subspaces of a high-dimensional space and then partition each data point in dataset into the corresponding subspace. This field has witnessed remarkable progress over recent decades, with substantial theoretical advancements and successful applications spanning image processing, genomic analysis and text analysis. However, existing surveys predominantly focus on conventional shallow-structured methods, with few up-to-date reviews on deep-structured methods, i.e., deep neural network-based approaches. In fact, recent years has witnessed the overwhelming success of deep neural network in various fields, including computer vision, natural language processing, subspace clustering. To address this gap, this paper presents a comprehensive review on subspace clustering methods, including conventional shallow-structured and deep neural network based approaches, which systematically analyzes over 150 papers published in peer-reviewed journals and conferences, highlighting the latest research achievements, methods, algorithms and applications. Specifically, we first briefly introduce the basic principles and evolution of subspace clustering. Subsequently, we present an overview of research on subspace clustering, dividing the existing works into two categories: shallow subspace clustering and deep subspace clustering, based on the model architecture. Within each category, we introduce a refined taxonomy distinguishing linear and nonlinear approaches based on data characteristics and subspace structural assumptions. Finally, we discuss the challenges currently faced and future research direction for development in the field of subspace clustering.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 11","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11349-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880945","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":"Automatic radiology report generation with deep learning: a comprehensive review of methods and advances","authors":"Yilin Li,&nbsp;Chao Kong,&nbsp;Guosheng Zhao,&nbsp;Zijian Zhao","doi":"10.1007/s10462-025-11337-0","DOIUrl":"10.1007/s10462-025-11337-0","url":null,"abstract":"<div><p>Automatic report generation refers to the process of generating medical reports from medical images without the need for manual intervention, enabling faster, more consistent, and objective analysis of radiological data. The rapid progress in deep learning, particularly in the fields of computer vision and natural language processing, has significantly improved the efficacy of this approach. By leveraging deep learning techniques, which seamlessly integrate image analysis with natural language generation, these methods have shown promise in interpreting complex medical images and producing highly accurate textual descriptions. In this paper, we provide a thorough review of various deep learning models and techniques employed for generating radiological reports, with a focus on chest X-ray images as a representative case. We propose a unified encoder-decoder framework that consists of an image encoder for extracting feature representations from medical images, a language decoder for generating textual reports, and enhancement components designed to refine model performance. Through a comprehensive comparison of existing state-of-the-art methods on the widely utilized MIMIC-CXR dataset, we highlight the innovative contributions made by recent advancements in the field. Furthermore, we discuss the current challenges and identify potential research directions for future advancements in this field.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 11","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11337-0.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144881012","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":"A novel ensemble learning-based soft measurement method for rod-pumping system efficiency","authors":"Biao Ma,&nbsp;Shimin Dong","doi":"10.1007/s10462-025-11343-2","DOIUrl":"10.1007/s10462-025-11343-2","url":null,"abstract":"<div><p>Accurate prediction of rod-pumping system efficiency is crucial for evaluating the performance of such systems. Currently, the efficiency of rod-pumping systems is primarily estimated using mechanistic models. With the continuous advancement of information technology and the improvement of oilfield databases, some researchers have employed single neural networks for prediction. However, single neural networks often suffer from low prediction accuracy and poor robustness to noise. To solve this problem, we propose a new integrated learning-based soft measurement of the efficiency of rod pumping systems. Firstly, we proposed five soft measurement methods for rod pumping system efficiency: BiGRU-BiLSTM-CrossAttention, BiRNN-BiGRU-KAN, CNN-BiGRU-KAN, BiLSTM-BiGRU-KAN, and BiLSTM-Transformer-KAN. Then, using these five methods as base learners and FNN as the meta-learner, we constructed a novel rod pumping system efficiency soft measurement method based on the Stacking ensemble learning framework. The hyperparameters were optimized using a multi-strategy integrated Crayfish optimization algorithm, and the model was validated using 5-fold cross-validation. To verify the accuracy of the proposed soft measurement method, we applied it to 10,250 real oil wells for calculation and conducted a comparative analysis with baseline models. The results demonstrate that the proposed soft measurement method can effectively predict the efficiency of rod pumping systems.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 11","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11343-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880946","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":"Artificial intelligence (AI) and machine learning (ML) in procurement and purchasing decision-support (DS): a taxonomic literature review and research opportunities","authors":"Dursun Balkan,&nbsp;Goknur Arzu Akyuz","doi":"10.1007/s10462-025-11336-1","DOIUrl":"10.1007/s10462-025-11336-1","url":null,"abstract":"<div><p>Artificial intelligence (AI), machine learning (ML) and decision-support (DS) are gaining increasing interest with widening adoption. This article investigates the enabler role of AI and ML for providing decision-support in procurement&amp;purchasing domain. The study follows a systematic review approach via taxonomic analysis. Comprehensive analysis and discussions are provided for: (a) the relevance and applicability of AI and ML in procurement&amp;purchasing decision-support; (b) functionalities/processes for which they are utilized; (c) related methodologies; and (d) implementation benefits as well as challenges. Findings reveal that procurement&amp;purchasing area holds significant potential in terms of AI-ML applications for decision-support almost every related sub-process. This study is original by offering a process-oriented approach to the research domain; providing unique clustering and classification; and presenting detailed analyses via unique taxonomy tables with respect to approach, topic, focus, context and methodologies of the literature items reviewed. The study offers further research opportunities and has significant potential to provide managerial insights by the identified sectoral applications, benefits and challenges.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 11","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11336-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880749","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 for cyber threat intelligence: applications, challenges, and analysis of real-world case studies","authors":"Prasasthy Balasubramanian,&nbsp;Sonali Liyana,&nbsp;Hamsini Sankaran,&nbsp;Shambavi Sivaramakrishnan,&nbsp;Sruthi Pusuluri,&nbsp;Susanna Pirttikangas,&nbsp;Ella Peltonen","doi":"10.1007/s10462-025-11338-z","DOIUrl":"10.1007/s10462-025-11338-z","url":null,"abstract":"<div><p>This paper presents a comprehensive survey of the applications, challenges, and limitations of Generative AI (GenAI) in enhancing threat intelligence within cybersecurity, supported by real-world case studies. We examine a wide range of data sources in Cyber Threat Intelligence (CTI), including security reports, blogs, social media, network traffic, malware samples, dark web data, and threat intelligence platforms (TIPs). This survey provides a full reference for integrating GenAI into CTI. We discuss various GenAI models such as Large Language Models (LLMs) and Deep Generative Models (DGMs) like Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), and Diffusion Models, explaining their roles in detecting and addressing complex cyber threats. The survey highlights key applications in areas such as malware detection, network traffic analysis, phishing detection, threat actor attribution, and social engineering defense. We also explore critical challenges in deploying GenAI, including data privacy, security concerns, and the need for interpretable and transparent models. As regulations like the European Commission’s AI Act emerge, ensuring trustworthy AI solutions is becoming more crucial. Real-world case studies, such as the impact of the WannaCry ransomware, the rise of deepfakes, and AI-driven social engineering, demonstrate both the potential and current limitations of GenAI in CTI. Our goal is to provide foundational insights and strategic direction for advancing GenAI’s role in future cybersecurity frameworks, emphasizing the importance of innovation, adaptability, and ongoing learning to enhance resilience against evolving cyber threats. Ultimately, this survey offers critical insights into how GenAI can shape the future of cybersecurity by addressing key challenges and providing actionable guidance for effective implementation.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 11","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11338-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880787","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 for intrusion detection in emerging technologies: a comprehensive survey and new perspectives","authors":"Euclides Carlos Pinto Neto,&nbsp;Shahrear Iqbal,&nbsp;Scott Buffett,&nbsp;Madeena Sultana,&nbsp;Adrian Taylor","doi":"10.1007/s10462-025-11346-z","DOIUrl":"10.1007/s10462-025-11346-z","url":null,"abstract":"<div><p>Intrusion Detection Systems (IDS) can help cybersecurity analysts detect malicious activities in computational environments. Recently, Deep Learning (DL) methods in IDS have demonstrated notable performance, revealing new underlying cybersecurity patterns in systems’ operations. Conversely, issues such as low performance in real systems, high false positive rates, and lack of explainability hinder its real-world deployment. In addition, the adoption of many new emerging technologies, such as cloud, edge computing, and the Internet of Things (IoT) introduces new forms of vulnerabilities. Therefore, the improvement of intrusion detection in emerging technologies depends on the clear definitions of challenging security problems and the limitations of existing solutions. The main goal of this research is to conduct a literature review of DL solutions for intrusion detection in emerging technologies to understand the state-of-the-art solutions and their limitations. Specifically, we conduct a comprehensive review of IDS-based automated threat defense methods, with the objective of identifying the landscape of, and opportunities for, incorporating DL methods into IDS. To accomplish this, a thorough review of IDS methods is conducted for multiple platforms and technologies, focusing on the use of common DL techniques. To expand on the study, several widely used IDS datasets are evaluated to assess their ability to train DL models and support researchers in understanding their characteristics and limitations. The analysis of attack vectors in emerging technologies is conducted, enabling an in-depth evaluation of security solutions in the future. Our findings show many clear opportunities for future research, including addressing the gap between solutions for controlled/simulated environments versus real systems, overcoming trustworthiness issues, including lack of explainability, and further exploring operationalization issues such as deployable solutions and continuous detection. Our analysis highlights that the operationalization of DL for intrusion detection in emerging technologies represents a key challenge to be addressed in the next few years.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 11","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11346-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880788","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":"Video diffusion generation: comprehensive review and open problems","authors":"Wenping Ma,&nbsp;Xiaoting Yang,&nbsp;Licheng Jiao,&nbsp;Lingling Li,&nbsp;Xu Liu,&nbsp;Fang Liu,&nbsp;Puhua Chen,&nbsp;Yuting Yang,&nbsp;Mengru Ma,&nbsp;Long Sun,&nbsp;Ruohan Zhang,&nbsp;Xueli Geng,&nbsp;Yuwei Guo,&nbsp;Shuyuan Yang,&nbsp;Zhixi Feng","doi":"10.1007/s10462-025-11331-6","DOIUrl":"10.1007/s10462-025-11331-6","url":null,"abstract":"<div><p>Video generation has become an increasingly important component of AI-generated content (AIGC), owing to its rich semantic expressiveness and growing application potential. Among various generative paradigms, diffusion models have recently gained prominence due to their strong controllability, competitive visual quality, and compatibility with multimodal inputs. However, most existing surveys provide limited coverage of diffusion-based video generation, often lacking systematic analysis and comprehensive comparisons. To address this gap, this paper presents a thorough and structured review of diffusion models for video generation. We first outline the theoretical foundations and core architectures of diffusion models, and then the key design principles of representative methods for video generation were introduced. We propose a unified taxonomy that categorizes over two hundred methods, analyzing their key characteristics, strengths, and limitations. In addition, we compared the performance of classical methods and summarized commonly used datasets and evaluation metrics in this field for ease of model benchmarking and selection. Finally, we discuss open problems and future research directions, aiming to provide a valuable reference for both academic research and practical development.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 11","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11331-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880785","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":"AI-mediated healthcare and trust. A trust-construct and trust-factor framework for empirical research","authors":"Marcos Alonso,&nbsp;Aníbal M. Astobiza,&nbsp;Ramón Ortega Lozano","doi":"10.1007/s10462-025-11306-7","DOIUrl":"10.1007/s10462-025-11306-7","url":null,"abstract":"<div><p>Application of Artificial Intelligence (AI) in healthcare is growing exponentially, and its use is expected to continue expanding in the coming years. However, lack of trust in AI systems remains a significant barrier to their widespread adoption. This article analyzes the problem of trust, its various features and its application in AI-mediated healthcare. We first review the literature on trust and trust in technology to detect which theoretical constructs are essential to trust. We then identify the factors that we consider fundamental for a rich and complex comprehension of trust in AI-mediated healthcare. We finally propose a trust-factor framework that could be used for empirical research on AI-mediated healthcare and its practical implementation.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 11","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11306-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880786","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":"Applications of deep learning in tea quality monitoring: a review","authors":"Tao Wu,&nbsp;Lei Zhou,&nbsp;Yiying Zhao,&nbsp;Hengnian Qi,&nbsp;Yuanyuan Pu,&nbsp;Chu Zhang,&nbsp;Yufei Liu","doi":"10.1007/s10462-025-11335-2","DOIUrl":"10.1007/s10462-025-11335-2","url":null,"abstract":"<div><p>Tea is a popular beverage which can offer numerous benefits to human health and support the local economy. There is an increasing demand for accurate and rapid tea quality evaluation methods to ensure that the quality and safety of tea products meet the customers’ expectations. Advanced sensing technologies in combination with deep learning (DL) offer significant opportunities to enhance the efficiency and accuracy for tea quality evaluation. This review aims to summarize the application of DL technologies for tea quality assessment in three stages: cultivation, tea processing, and product evaluation. Various state-of-the-art sensing technologies (e.g., computer vision, spectroscopy, electronic nose and tongue) have been used to collect key data (images, spectral signals, aroma profiles) from tea samples. By utilizing DL models, researchers are able to analyze a wide range of tea quality attributes, including tea variety, geographical origin, quality grade, fermentation stage, adulteration level, and chemical composition. The findings from this review indicate that DL, with its end-to-end analytical capability and strong generalization performance, can serve as a powerful tool to support various sensing technologies for accurate tea quality detection. However, several challenges remain, such as limited sample availability for data training, difficulties for fusing data from multiple sources, and lack of interpretability of DL models. To this end, this review proposes potential solutions and future studies to address these issues, providing practical considerations for tea industry to effectively uptake new technologies and to support the development of the tea industry.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 11","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11335-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880790","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}