Artificial Intelligence Review最新文献

{"title":"Don’t push the button! Exploring data leakage risks in machine learning and transfer learning","authors":"Andrea Apicella,&nbsp;Francesco Isgrò,&nbsp;Roberto Prevete","doi":"10.1007/s10462-025-11326-3","DOIUrl":"10.1007/s10462-025-11326-3","url":null,"abstract":"<div><p>Machine Learning (ML) has revolutionized various domains, offering predictive capabilities in several areas. However, there is growing evidence in the literature that ML approaches are not always used appropriately, leading to incorrect and sometimes overly optimistic results. One reason for this inappropriate use of ML may be the increasing availability of machine learning tools, leading to what we call the “push the button” approach. While this approach provides convenience, it raises concerns about the reliability of outcomes, leading to challenges such as incorrect performance evaluation. In particular, this paper addresses a critical issue in ML, known as data leakage, where unintended information contaminates the training data, impacting model performance evaluation. Indeed, crucial steps in ML pipeline can be inadvertently overlooked, leading to optimistic performance estimates that may not hold in real-world scenarios. The discrepancy between evaluated and actual performance on new data is a significant concern. In particular, this paper categorizes data leakage in ML, discussing how certain conditions can propagate through the ML approach workflow. Furthermore, it explores the connection between data leakage and the specific task being addressed, investigates its occurrence in Transfer Learning framework, and compares standard inductive ML with transductive ML paradigms. The conclusion summarizes key findings, emphasizing the importance of addressing data leakage for robust and reliable ML applications considering tasks and generalization goals.</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/s10462-025-11326-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880784","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 review on key technologies toward smart healthcare systems based IoT: technical aspects, challenges and future directions","authors":"Muntadher Alsabah,&nbsp;Marwah Abdulrazzaq Naser,&nbsp;A. S. Albahri,&nbsp;O. S. Albahri,&nbsp;A. H. Alamoodi,&nbsp;Sadiq H. Abdulhussain,&nbsp;Laith Alzubaidi","doi":"10.1007/s10462-025-11342-3","DOIUrl":"10.1007/s10462-025-11342-3","url":null,"abstract":"<div><p>The unexpected death of humans due to a lack of medical care is a serious problem. Additionally, the number of elderly people requiring continuous care is increasing. A global aging population poses a challenge to the sustainability of conventional healthcare systems for the future. Simultaneously, recent years have seen remarkable progress in the Internet of Things (IoT) and communication technologies, alongside the growing importance of artificial intelligence (AI) explainability and information fusion. Therefore, developing smart healthcare systems based on IoT and advanced technologies is crucial. This would open up new possibilities for efficient and intelligent medical systems. Hence, it is imperative to present a prospective vision of smart healthcare systems and explore the key technologies that enable the development of these intelligent medical systems. With smart healthcare systems, the future of healthcare can be significantly enhanced, providing higher-quality care, improved treatment, and more efficient patient care. This paper aims to provide a comprehensive review of the key enabling and innovative technologies for smart healthcare systems. To this end, it will cover the primary goals of each technology, the current state of research, potential applications envisioned, associated challenges, and future research directions. This paper is intended to be a valuable resource for researchers and healthcare providers. Ultimately, this paper provides valuable insights for both industry professionals and academic researchers, while also identifying potential new research avenues.</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/s10462-025-11342-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880789","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":"Advancements in multimodal differential evolution: a comprehensive review and future perspectives","authors":"Dikshit Chauhan,&nbsp; Shivani,&nbsp;Donghwi Jung,&nbsp;Anupam Yadav","doi":"10.1007/s10462-025-11314-7","DOIUrl":"10.1007/s10462-025-11314-7","url":null,"abstract":"<div><p>Multi-modal optimization involves identifying multiple global and local optima of a function, offering valuable insights into diverse optimal solutions within the search space. Evolutionary algorithms (EAs) excel at finding various solutions in a single run, providing a distinct advantage over classical optimization techniques that often require multiple restarts without guarantee of obtaining diverse solutions. Among these EAs, differential evolution (DE) stands out as a powerful and versatile optimizer for continuous parameter spaces. DE has shown significant success in multi-modal optimization by utilizing its population-based search to promote the formation of multiple stable subpopulations, each targeting different optima. Recent advancements in DE for multi-modal optimization have focused on niching methods, parameter adaptation, hybridization with other algorithms, including machine learning, and applications across various domains. Given these developments, it is an opportune moment to present a critical review of the latest literature and identify key future research directions. This paper offers a comprehensive overview of recent DE advancements in multimodal optimization, including methods for handling multiple optima, hybridization with EAs, and machine learning, and highlights a range of real-world applications. Additionally, the paper outlines a set of compelling open problems and future research issues from multiple perspectives.</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/s10462-025-11314-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144880869","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":"An efficient model for diabetic detection using heuristic approach based serial cascaded convolutional ensemble network","authors":"Santosh Kumar Bejugam,&nbsp;Jyothi Vankara","doi":"10.1007/s10462-025-11334-3","DOIUrl":"10.1007/s10462-025-11334-3","url":null,"abstract":"<div><p>Diabetes is a chronic pathology that poses significant risks to people. If diabetes is not properly diagnosed and treated, it may contribute to serious health problems. Delayed diagnosis causes many health issues and leads to numerous deaths every year. So, researchers have developed efficient diabetes detection systems for the early detection of this pathology. However, the existing model raises serious issues about the security and privacy of private medical information, and it requires rigorous safety precautions to prevent intrusions and unapproved access. In addition, the unclear characteristics of existing models cause difficulty in healthcare facilities. Thus, the advanced deep learning-based diabetic detection model was designed in this work to overcome these challenges. Also, it aims to detect diabetics and helps to prevent the progression of diabetes in patients. At first, the required data is gathered from the online data source and then fed to the optimal feature selection phase. Here, the features and weight are optimally selected using the Fitness-based Billiards-Inspired Optimization (FBIO) algorithm. This process helps the model to focus on the most impactful information within the data. Further, the obtained optimal weighted feature is passed to the Serial Cascaded Convolutional Ensemble Network (SCCEN) for detection. Here, the SCCEN model serially cascades techniques such as Convolutional Autoencoder (CAE), “1-dimensional Convolutional Neural Network” (1DCNN), and “Convolutional Long Short-Term Memory” (ConvLSTM). This process helps to improve the detection accuracy. Finally, the designed approach’s effectiveness is analyzed by comparing its performance with existing techniques. The suggested approach’s accuracy for dataset-1 is 97.4%, dataset-2 is 97.31%, and dataset-3 is 96.69%, which is higher than the conventional techniques and optimization algorithms. Thus, the result proved that the introduced framework can detect diabetics in premature stages and help the patient to take suitable treatment.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 10","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11334-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144832268","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 review and critical analysis of multimodal datasets for emotional AI","authors":"Sadam Al-Azani,&nbsp;El-Sayed M. El-Alfy","doi":"10.1007/s10462-025-11271-1","DOIUrl":"10.1007/s10462-025-11271-1","url":null,"abstract":"<div><p>With the increasing interest in digital technologies, emotion recognition plays an important role in several applications such as healthcare computer-aided diagnosis, social media analysis, opinion mining and recommendation systems, understanding human behavior and interaction in workplaces, effective communication and linguistic analysis, and cognitive human–machine interaction. This field is receiving a growing interest in recent years. In this paper, we present a thorough review of emotional artificial intelligence through identification and in-depth analysis of existing multimodal datasets along with their related research directions and methodologies. It establishes essential requirements for the development of a multimodal dataset and outlines challenges spanning its entire lifecycle, from recording to deployment. Moreover, a taxonomy of various categories and applications is introduced based on the key characteristics of various multimodal datasets. Finally, the paper concludes with discussions and insights into future directions and prospects for standard schemes to facilitate the efficient development of reliable and reusable benchmark datasets that can help researchers and developers advance this field.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 10","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11271-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144832267","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":"Do you actually need an LLM? Rethinking language models for customer reviews analysis","authors":"Yang Xiao,&nbsp;Yunke Li,&nbsp;Shaoyujie Chen,&nbsp;Hayden Barker,&nbsp;Ryan Rad","doi":"10.1007/s10462-025-11308-5","DOIUrl":"10.1007/s10462-025-11308-5","url":null,"abstract":"<div><p>LLarge language models (LLMs) demonstrate strong natural language processing capabilities but come with significant computational costs, raising questions about their practical utility compared to small language models (SLMs). This study systematically compares SLMs (DistilBERT, ELECTRA) and LLMs (Flan-T5, Flan-UL2) on two customer review analysis tasks: sentiment polarity classification and product correlation analysis. Our results show that while LLMs outperform in sentiment classification, they do so at a much higher computational cost, whereas fine-tuned SLMs excel in domain-specific correlation analysis with greater efficiency. To balance accuracy and cost, we propose a context-enhanced hybrid (CE-Hybrid) model, which refines traditional hybrid methods by enriching LLM input with SLM-generated insights, reducing redundant computation while maintaining accuracy. Our findings quantify the trade-offs between model performance and resource efficiency, offering actionable insights for businesses to optimize AI deployment. These results have significant implications for real-world applications such as e-commerce, customer service automation, and business analytics.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 10","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11308-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145142975","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":"Enhancing CCTA image quality: a review of deep learning approaches for advanced artifact correction and denoising","authors":"Mohanad Alkhodari,&nbsp;Eman Alefisha,&nbsp;Herbert F. Jelinek,&nbsp;Ahmed Kaabneh,&nbsp;Panos Liatsis","doi":"10.1007/s10462-025-11311-w","DOIUrl":"10.1007/s10462-025-11311-w","url":null,"abstract":"<div><p>Cardiac imaging is vital for diagnosing coronary artery disease (CAD), with coronary computed tomography angiography (CCTA) being commonly used to evaluate coronary vessels for stenosis, calcification, and atherosclerosis. However, CCTA images often suffer from artifacts like beam hardening, scatter, and noise, degrading image quality and obscuring anatomical details, leading to diagnostic uncertainty. Conventional post-processing techniques, such as filtered back projection and iterative reconstruction, have limited effectiveness in correcting these artifacts, posing challenges in CCTA, where precise visualization of coronary arteries is crucial. Artifacts can blur vessel boundaries, obscure calcified plaques, and misrepresent stenosis severity, potentially leading to misdiagnosis and suboptimal clinical decisions. Recent advancements in computational imaging, particularly deep learning algorithms, offer clinical benefits for artifact reduction in CCTA. Deep learning models, such as convolutional neural networks (CNNs), outperform traditional methods by effectively de-noising and correcting artifacts through learning complex patterns from large datasets. These models adapt to the non-linear, heterogeneous nature of artifacts, enhancing image clarity and diagnostic reliability. Improved image quality in CCTA enables better visualization of coronary arteries, aiding in accurate assessment of stenosis and calcification. This review highlights deep learning approaches for artifact correction in CCTA, emphasizing their potential to improve CAD diagnosis.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 10","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11311-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145142678","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 dual-driven MAGDM method based on single-valued neutrosophic credibility numbers Einstein variable extended power geometric aggregation operator and SPA-MARCOS","authors":"Pingqing Liu,&nbsp;Junxin Shen,&nbsp;Peng Zhang","doi":"10.1007/s10462-025-11299-3","DOIUrl":"10.1007/s10462-025-11299-3","url":null,"abstract":"<div><p>Traditional multi-attribute group decision-making (MAGDM) methods primarily rely on expert knowledge-driven information, often overlooking the value of objective data in decision-making processes. To address this gap, this paper proposes a novel dual-driven MAGDM method that incorporates knowledge-driven information, expressed through single-valued neutrosophic credibility numbers (SvNCNs), and data-driven information, represented by exact numbers (ENs). The primary innovations of this method include the development of the variable extended power geometric (VEPG) operator, which effectively aggregates knowledge-driven information from SvNCNs. The SvNCN Einstein variable extended power geometric (SvNCNEVEPG) operator is also introduced, and its mathematical properties are rigorously proven, offering an advanced approach to handling extreme values. To resolve ambiguity and uncertainty in decision analysis, a new subjective weight determination method, SvNCN-PIPRECIA, is introduced, complemented by an objective entropy-based weighting method. These are integrated into a new combined weight determination model using the Uninorm operator, enhancing the accuracy and reliability of the decision-making process. The dual-driven MAGDM method, combining knowledge-driven and data-driven information, improves decision-making comprehensiveness and precision through the dual-driven SPA-Entropy-PIPRECIA-MARCOS approach. The proposed methodology is validated through a case study on evaluating data trading platforms (DTPs), where sensitivity analysis of parameters and a comparative study with existing methods demonstrate the flexibility and scientific robustness of the approach.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 10","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11299-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145142609","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 lightweight few-shot learning model for crop pest and disease identification","authors":"Linsen Wei,&nbsp;Jingjun Tang,&nbsp;Jinxiu Chen,&nbsp;Carine Pierrette Mukamakuza,&nbsp;Defu Zhang,&nbsp;Tong Zhang","doi":"10.1007/s10462-025-11323-6","DOIUrl":"10.1007/s10462-025-11323-6","url":null,"abstract":"<div><p>Production quality is directly related to the economic development of agriculture. However, the growth of crops is susceptible to pest and disease infestations, which can negatively affect agricultural yields. Therefore, adopting efficient pest and disease identification methods is of the utmost importance. This paper proposes a lightweight few-shot learning model for crop pest and disease identification. The model utilizes a lightweight backbone network and incorporates adaptive spatial feature fusion to aggregate multi-scale features, thus avoiding feature redundancy and interference between multi-scale features. Additionally, a lightweight and efficient attention module is introduced to further explore the salient information in images from both channel and spatial dimensions. Experimental results demonstrate that, compared to the state-of-the-art methods in the field, the model achieved an average recognition accuracy improvement of 0.41% under the 10-shot setting on the PlantVillage dataset and improvements of 4.03% and 2.47% under the 5-shot and 10-shot settings, respectively, on the PlantDoc dataset. Furthermore, the model achieved a 1.46% increase in overall average recognition accuracy on the IP102 dataset, while also showing strong generalization capabilities on locally collected datasets.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 10","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11323-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145145567","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-based methods for protein structure prediction: a survey","authors":"Zhi-Hui Zhan,&nbsp;Jun Hong,&nbsp;Jian-Yu Li,&nbsp;Cheng Wang,&nbsp;Langchong He,&nbsp;Zongben Xu,&nbsp;Jun Zhang","doi":"10.1007/s10462-025-11325-4","DOIUrl":"10.1007/s10462-025-11325-4","url":null,"abstract":"<div><p>Protein structure prediction (PSP) is a meaningful problem that has drawn worldwide attention, where various artificial intelligence (AI) techniques, such as evolutionary computation (EC)-based and neural networks (NNs)-based methods, have been applied to PSP and have obtained promising results in recent years. Considering the rapid and significant advances of AI-based methods for PSP, it is vital to make a survey on this progress to summarize the existing research experience and to provide guidelines for further development of related research fields. With these aims, a broad survey of AI-based methods for solving PSP problems is provided in this article. First, EC-based PSP methods are reviewed, which are organized by three key steps involved in using EC-based methods for PSP. Second, NNs-based PSP methods are reviewed. More specifically, typical NNs-based methods to predict protein structural features are described and state-of-the-art NNs-based methods with end-to-end architecture and attention mechanism are reviewed. Third, the accuracy, interpretability, accessibility, and ethical challenges of AI-based methods are discussed. Last, the future directions including hybrid AI paradigm, protein language models, and the prediction of protein complexes and biomolecular interactions are given, and the conclusion is drawn. This survey is expected to draw attention, raise discussions, and inspire new ideas in the wonderful interdisciplinary field of biology and AI.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 10","pages":""},"PeriodicalIF":13.9,"publicationDate":"2025-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11325-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145145576","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}