Artificial Intelligence Review最新文献

{"title":"Modeling and effect analysis of machining parameters for surface roughness and specific energy consumption during TC18 machining using deep reinforcement learning and neural networks","authors":"Juan Lu,&nbsp;Huailong Mu,&nbsp;Haibin Ouyang,&nbsp;Zhenkun Zhang,&nbsp;Weiping Ding","doi":"10.1007/s10462-025-11178-x","DOIUrl":"10.1007/s10462-025-11178-x","url":null,"abstract":"<div><p>Under the impetus of green manufacturing and a low-carbon economy, the critical challenge lies in reducing energy consumption while maintaining machining quality. Against this background, this paper presents the method of modeling and effect analysis for surface roughness and specific energy consumption during TC18 machining using Deep Reinforcement Learning and Neural Networks. In this method, to reduce the experiment cost, multilayer-layer design (MLD) for computer simulation is applied to design a physical experiment, and to improve modeling accuracy, backpropagation neural network (BPNN) optimized by Double deep Q network algorithm (DDQN) is utilized to develop the prediction models of surface roughness (<i>Ra</i>) and specific energy consumption of cutting (<i>E</i>_<i>sec</i>). Finaly, the synergistic influence of cutting parameters on <i>Ra</i> and <i>E</i>_<i>sec</i> is analyzed based on the prediction models of <i>Ra</i> and <i>E</i>_<i>sec</i> built by MLD and DDQN-BPNN. The effectiveness and low cost of MLD and the excellent prediction performance of DDQN-BPNN are verified by comparisons of optimized BPNNs using common heuristic optimization algorithms through the milling experiment of TC18. These technologies provide effective solutions for modeling and factor impact of target features in machining field, and research results provides an effective guidance for the selection of milling parameters of TC18 to reduce the specific energy consumption of cutting under ensuring or improving machining quality.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 7","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11178-x.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821958","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-driven job scheduling in cloud computing: a comprehensive review","authors":"Yousef Sanjalawe,&nbsp;Salam Al-E’mari,&nbsp;Salam Fraihat,&nbsp;Sharif Makhadmeh","doi":"10.1007/s10462-025-11208-8","DOIUrl":"10.1007/s10462-025-11208-8","url":null,"abstract":"<div><p>The demand for efficient job scheduling in cloud computing has grown significantly with the rise of dynamic and heterogeneous cloud environments. While effective in simpler systems, traditional scheduling algorithms fail to meet the complex requirements of modern cloud infrastructures. These limitations motivate the need for AI-driven solutions that offer adaptability, scalability, and energy efficiency. This paper comprehensively reviews AI-based job scheduling techniques, addressing several key research gaps in current approaches. The existing methods face challenges such as resource heterogeneity, energy consumption, and real-time adaptability in multi-cloud systems. Accordingly, the support of AI-based job scheduling in cloud computing is summarized here toward machine learning, optimization techniques, heuristic techniques, and hybrid AI models. This paper pointedly underlines the strengths and weaknesses of various approaches through deep comparative analysis and focuses on how AI will overcome traditional algorithm shortcomings. Is worth noticing that several important improvements this kind of AI-driven model provides, for example, in resource allocation, cost efficiency, energy consumption, and complex dependencies between jobs and system faults. In the end, AI-driven job scheduling seems to be a promising avenue toward effectively responding to the booming demands of cloud infrastructures. Future research should concentrate on three major outlooks: scalability, better integration of AI with traditional scheduling methods, and the use of other emerging technologies like edge computing and blockchain for better optimization of cloud-based job scheduling. The paper underscores the need for more adaptive, secure, and energy-efficient scheduling frameworks to meet the evolving challenges of cloud environments.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 7","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11208-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821957","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-criteria decision support system for the evaluation of UAV intelligent agricultural sensors","authors":"Bartłomiej Kizielewicz,&nbsp;Jarosław Wątróbski,&nbsp;Wojciech Sałabun","doi":"10.1007/s10462-025-11201-1","DOIUrl":"10.1007/s10462-025-11201-1","url":null,"abstract":"<div><p>Precision agriculture is an emerging approach aimed at enhancing agricultural productivity through advanced technological solutions. One of the key technologies integrated into modern agriculture is Unmanned Aerial Vehicles (UAVs), which rely on various sensors to provide critical information about crop fields. However, selecting the most suitable UAV sensors remains a significant challenge due to multiple evaluation criteria and compromises. This paper proposes a novel decision-support framework based on multi-criteria decision-making/analysis (MCDA/MCDM) methods to facilitate UAV sensor selection in precision agriculture. The framework incorporates objective weight selection techniques-Standard Deviation, Entropy, CRITIC, and MEREC-eliminating the need for subjective expert involvement. Furthermore, four MCDA/MCDM methods, including the newly proposed COmbined COmpromise solution with Characteristic Objects METhod (COCOCOMET), are applied to evaluate sensor alternatives. To validate the framework, a case study is conducted using a dataset of UAV sensors, where multiple evaluation criteria are analyzed to determine the most suitable sensor. The results confirm the framework’s effectiveness, demonstrating its robustness and stability in decision-making. Sensitivity analysis and comparative studies further highlight its reliability, particularly in addressing rank reversal issues commonly found in existing MCDA methods such as TOPSIS and AHP. The proposed framework not only provides a structured and adaptable evaluation process for UAV sensors but also offers broader applicability in agricultural decision-making.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 7","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11201-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143822038","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":"Power aggregation operators based on Aczel-Alsina T-norm and T-conorm for intuitionistic hesitant fuzzy information and their application to logistics service provider selection","authors":"Peng Wang,&nbsp;Baoying Zhu,&nbsp;Keyan Yan,&nbsp;Ziyu Zhang,&nbsp;Zeeshan Ali,&nbsp;Dragan Pamucar","doi":"10.1007/s10462-025-11155-4","DOIUrl":"10.1007/s10462-025-11155-4","url":null,"abstract":"<div><p>Logistics service provider selection is a skilled and effective technique used to evaluate and identify third-party enterprises or organizations capable of managing and performing logistics tasks on behalf of a business. In this study, we aim to propose a logistics service provider selection method based on aggregation operators and fuzzy sets. First, we analyze the Aczel-Alsina operational laws building on the intuitionistic hesitant fuzzy sets technique, which combines hesitant fuzzy and intuitionistic fuzzy models. Subsequently, we derive power averaging/geometric aggregation operators for the intuitionistic hesitant fuzzy model based on Aczel-Alsina operational laws, called the intuitionistic hesitant fuzzy Aczel-Alsina power averaging operator, intuitionistic hesitant fuzzy Aczel-Alsina weighted power averaging operator, intuitionistic hesitant fuzzy Aczel-Alsina power geometric operator, and intuitionistic hesitant fuzzy Aczel-Alsina weighted power geometric operator, with their flexible and basic properties such as idempotency, monotonicity, and boundedness. The existing model of drastic aggregation operators, max–min aggregation operators, and algebraic aggregation operators are the special cases of the proposed theory. To address the selection of logistics service providers using the proposed operators, we explore the multi-attribute decision-making methods to evaluate the required problems. Finally, we compare the ranking results of the invented model with those of existing technologies to describe the effectiveness and stability of the proposed methods.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 7","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11155-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143822039","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 of machine learning applications for internet of nano things: challenges and future directions","authors":"Aryan Rana,&nbsp;Deepika Gautam,&nbsp;Pankaj Kumar,&nbsp;Kranti Kumar,&nbsp;Athanasios V. Vasilakos,&nbsp;Ashok Kumar Das,&nbsp;Vivekananda Bhat K","doi":"10.1007/s10462-025-11211-z","DOIUrl":"10.1007/s10462-025-11211-z","url":null,"abstract":"<div><p>In recent years, advances in nanotechnology and the Internet of Things (IoT) have led to the development of the revolutionary Internet of Nano Things (IoNT). IoNT, has found very similar real-life applications in agriculture, military, multimedia, and healthcare. However, despite the rapid advancements in both IoNT and machine learning (ML), there has been no comprehensive review explicitly focused on the integration of these two fields. Existing surveys and reviews on IoNT primarily address its architecture, communication methods, and domain-specific applications, yet overlook the critical role ML could play in enhancing IoNT’s capabilities–particularly in data processing, anomaly detection, and security. This survey addresses this gap by providing an in-depth analysis of IoNT-ML integration, reviewing state-of-the-art ML applications within IoNT, and systematically discussing the challenges that persist in this integration. Additionally, we propose future research directions, establishing a framework to guide advancements in IoNT through ML-driven solutions.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 7","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11211-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143822040","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":"Speech emotion recognition in conversations using artificial intelligence: a systematic review and meta-analysis","authors":"Ghada Alhussein,&nbsp;Ioannis Ziogas,&nbsp;Shiza Saleem,&nbsp;Leontios J. Hadjileontiadis","doi":"10.1007/s10462-025-11197-8","DOIUrl":"10.1007/s10462-025-11197-8","url":null,"abstract":"<div><p> Manifestations of emotion in social conversational interactions stand at a focal point in the rapidly growing affective computing area, with applications in healthcare, education and human-computer interaction. Artificial intelligence (AI) holds great potential in modeling the challenging dynamic nature of affect in speech conversation. In this paper, we analyze and criticize latest trends and open problems through a systematic review and multi-subgroup meta-analysis of AI approaches for emotion recognition in conversation (ERC). We adopt the PRISMA-DTA guidelines toward analysis of AI-driven speech ERC. A comprehensive database search through predefined query strings and selection criteria allowed for data extraction of essential diagnostic performance parameters. We analyze salient patterns related to methodological quality and risk of bias. Univariate random-effects models are then designed with a multi-subgroup perspective, centered around affective annotations models, while encompassing the ERC parameters of modalities, feature extraction and conversation style. 51 studies were systematically reviewed for qualitative analysis, whereas 27 articles were included in the meta-analysis. Diagnostic test performance manifested with high heterogeneity, with intriguing insights regarding affective state annotation, input modality, feature extraction methods, and dataset conversation style. Our analysis raised concerns regarding bias, reporting quality and inter-rater reliability in annotations. Our research contributes fine-grained insights as recommendations that tackle open-problems in ERC. While providing valuable information on diagnostic performance of AI in speech ERC, we underscore the imperative need for further advancements in annotations and models capable of handling diverse emotional expressions. </p><p>Trial Registration: PROSPERO identifier - CRD42023416879.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 7","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11197-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143822042","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":"Dual representation learning for one-step clustering of multi-view data","authors":"Wei Zhang,&nbsp;Zhaohong Deng,&nbsp;Kup-Sze Choi,&nbsp;Jun Wang,&nbsp;Shitong Wang","doi":"10.1007/s10462-025-11183-0","DOIUrl":"10.1007/s10462-025-11183-0","url":null,"abstract":"<div><p>In real-world applications, multi-view data is widely available and multi-view learning is an effective method for mining multi-view data. In recent years, multi-view clustering, as an important part of multi-view learning, has been receiving more and more attention, while how to design an effective multi-view data mining method and make it more pertinent for clustering is still a challenging mission. For this purpose, a new one-step multi-view clustering method with dual representation learning is proposed in this paper. First, based on the fact that multi-view data contain both consistent knowledge between views and unique knowledge of each view, we propose a new dual representation learning method by improving the matrix factorization to explore them and to form common and specific representations. Then, we design a novel one-step multi-view clustering framework, which unifies the dual representation learning and multi-view clustering partition into one process. In this way, a mutual self-taught mechanism is developed in this framework and leads to more promising clustering performance. Finally, we also introduce the maximum entropy and orthogonal constraint to achieve optimal clustering results. Extensive experiments on seven real world multi-view datasets demonstrate the effectiveness of the proposed method.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 7","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11183-0.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821956","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 enhanced framework for real-time dense crowd abnormal behavior detection using YOLOv8","authors":"Rabia Nasir,&nbsp;Zakia Jalil,&nbsp;Muhammad Nasir,&nbsp;Tahani Alsubait,&nbsp;Maria Ashraf,&nbsp;Sadia Saleem","doi":"10.1007/s10462-025-11206-w","DOIUrl":"10.1007/s10462-025-11206-w","url":null,"abstract":"<div><p>Abnormal behavior detection in dense crowd, during the Hajj pilgrimage is vital to public security. Existing approaches face challenges due to factors like occlusions, illumination variations, and uniform attire. This research introduces the Crowd Anomaly Detection Framework (CADF), an improved YOLOv8-based model, integrating Soft-NMS to improve detection accuracy under complex conditions. CADF extensively evaluated on the Hajjv2 dataset, delivering an AUC of 88.27%, a 13.09% improvement over YOLOv2 and 12.19% over YOLOv5, with an Accuracy of 91.6%. To validate its generalizability, the framework is also tested on UCSD and ShanghaiTech datasets. Comparisons with state-of-the-art models, including VGG19 and EfficientDet, demonstrated CADF’s superiority in accuracy, AUC, precision, recall, and mAP metrics. By addressing the unique challenges of Hajj crowd and achieving strong performance across diverse datasets, CADF highlights its potential for real-time crowd anomaly detection, contributing to enhanced safety in large-scale public gatherings and aligning with Sustainable Development Goals 3 and 11.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 7","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11206-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821955","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":"Challenges in detecting security threats in WoT: a systematic literature review","authors":"Ruhma Sardar,&nbsp;Tayyaba Anees,&nbsp;Ahmad Sami Al-Shamayleh,&nbsp;Erum Mehmood,&nbsp;Wajeeha Khalil,&nbsp;Adnan Akhunzada,&nbsp;Fatema Sabeen Shaikh","doi":"10.1007/s10462-025-11176-z","DOIUrl":"10.1007/s10462-025-11176-z","url":null,"abstract":"<div><p>The rapid expansion of the Web of Things (WoT) and the Internet of Things (IoT) has raised security issues, with Denial of Service (DoS) attacks becoming increasingly prevalent. So, the aim of this study is to identify the security concerns in the four architectural layers of the Web of Things, particularly DoS attacks. For this study, existing literature are identified using search queries, and approximately 80 of relevant primary papers published in the recent decade are obtained after a thorough review which helps in addressing our research questions. After finding the relevant primary studies, we applied strict quality evaluation criteria to verify that all studies are evaluated. In addition, a taxonomy of deep learning (DL) techniques is presented on the basis of literature analysis conducted in this research, which is then used to characterize the various security concerns that occur in IoT and WoT systems. The study also examines which DL approaches are used to detect DoS/DDoS attacks in IoT and WoT. Our findings indicate that the optimal form of Intrusion Detection System (IDS) for dealing with DoS attacks is a hybrid IDS, which uses both the signature-based and the anomaly-based IDS. Moreover, DL techniques such as, CNNs and LSTMs, produced excellent results but are still in the development stage in terms of scalability and practical use. This review further highlights the present state of security mechanisms and sets the basis for future research, with an emphasis on refining DL-based techniques and improving the scalability and adaptability of security systems for WoT networks.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 7","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11176-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143822026","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 systematic survey of hybrid ML techniques for predicting peak particle velocity (PPV) in open-cast mine blasting operations","authors":"Gundaveni Shylaja,&nbsp;Ragam Prashanth","doi":"10.1007/s10462-025-11156-3","DOIUrl":"10.1007/s10462-025-11156-3","url":null,"abstract":"<div><p>Blasting operations in open-cast mines, while essential for mineral extraction, can generate significant peak particle velocity (PPV), posing environmental and structural risks. Accurate PPV prediction is critical to mitigate these effects and optimize blasting practices. This review introduces a hybrid ML approach that combines traditional methods, such as decision trees and SVMs, with advanced techniques like ensemble learning and neural networks. The performance of these models is evaluated based on blast parameters, geographical conditions, and monitoring data. The study highlights that hybrid and ensemble methods outperform other techniques in the majority of cases, especially in surface blasting scenarios. The increasing use of these advanced methods underscores their potential to address key challenges in blasting operations. Hybrid machine learning models over traditional methods by combining the strengths of multiple algorithms, effectively reducing bias and variance while enhancing predictive accuracy. Unlike conventional models, which often struggle with nonlinear relationships and high-dimensional data, hybrid approaches leverage advanced feature engineering, ensemble learning, and optimization techniques to improve robustness and generalization. In our study, these models demonstrated superior reliability in predicting PPV, achieving higher accuracy in terms of RMSE and <span>(text {R}^2)</span>. By combining different techniques, they mitigate individual model weaknesses, reduce errors, and improve feature selection. In addition, hybrid models prevent overfitting and optimize predictions through ensemble strategies such as boosting and stacking. This study explores the advantages of hybrid ML models, demonstrating their superior performance compared to conventional approaches. The review also identifies gaps in research on underground blasting and suggests future directions, emphasizing the importance of ongoing technological advancements and industry awareness of ML techniques benefits. Enhanced accuracy and robustness in PPV prediction, driven by hybrid approaches and real-time systems, are crucial to improve safety and efficiency in mining operations.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 7","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11156-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143822043","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}