Artificial Intelligence Review最新文献

{"title":"A review on persian question answering systems: from traditional to modern approaches","authors":"Safoura Aghadavoud Jolfaei,&nbsp;Azadeh Mohebi","doi":"10.1007/s10462-025-11122-z","DOIUrl":"10.1007/s10462-025-11122-z","url":null,"abstract":"<div><p>Question answering systems (QAS) are designed to answer questions in natural language. The objective of these types of systems is to reduce the user’s effort to manually check the retrieved documents to find the answer to the query in natural language and to create an accurate answer to the user’s query. In recent years, with the emergence of Large Language Models (LLMs), these systems have evolved significantly across different languages. However, the development of QAS in low resource languages such as Persian, while progressing, still faces unique challenges. Development of these systems has become problematic in Persian language due to the lack of comprehensive processing tools, limited question answering datasets, and specific challenges of this language. The current study provides a brief explanation of these systems’ evolution from traditional architectures to LLM-based approaches, their classification, the challenges specific to Persian language, existing question-answering datasets and language models, and studies conducted concerning Persian QAS.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 5","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11122-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396602","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":"Scalogram based performance comparison of deep learning architectures for dysarthric speech detection","authors":"Shaik Mulla Shabber,&nbsp;E. P. Sumesh,&nbsp;Vidhya Lavanya Ramachandran","doi":"10.1007/s10462-024-11085-7","DOIUrl":"10.1007/s10462-024-11085-7","url":null,"abstract":"<div><p>Dysarthria, a speech disorder commonly associated with neurological conditions, poses challenges in early detection and accurate diagnosis. This study addresses these challenges by implementing preprocessing steps, such as noise reduction and normalization, to enhance the quality of raw speech signals and extract relevant features. Scalogram images generated through wavelet transform effectively capture the time-frequency characteristics of the speech signal, offering a visual representation of the spectral content over time and providing valuable insights into speech abnormalities related to dysarthria. Fine-tuned deep learning models, including pre-trained convolutional neural network (CNN) architectures like VGG19, DenseNet-121, Xception, and a modified InceptionV3, were optimized with specific hyperparameters using training and validation sets. Transfer learning enables these models to adapt features from general image classification tasks to classify dysarthric speech signals better. The study evaluates the models using two public datasets TORGO and UA-Speech and a third dataset collected by the authors and verified by medical practitioners. The results reveal that the CNN models achieve an accuracy (acc) range of 90% to 99%, an F1-score range of 0.95 to 0.99, and a recall range of 0.96 to 0.99, outperforming traditional methods in dysarthria detection. These findings highlight the effectiveness of the proposed approach, leveraging deep learning and scalogram images to advance early diagnosis and healthcare outcomes for individuals with dysarthria.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 5","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11085-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396601","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":"Dmixnet: a dendritic multi-layered perceptron architecture for image recognition","authors":"Weixiang Xu,&nbsp;Yaotong Song,&nbsp;Shubham Gupta,&nbsp;Dongbao Jia,&nbsp;Jun Tang,&nbsp;Zhenyu Lei,&nbsp;Shangce Gao","doi":"10.1007/s10462-025-11123-y","DOIUrl":"10.1007/s10462-025-11123-y","url":null,"abstract":"<div><p>In the field of image recognition, the all-MLP architecture (MLP-Mixer) shows superior performance. However, the current MLP-Mixer is solely based on fully connected layers. The nonlinear capability of fully connected layers is relatively weak, and their simple stacked structure has limitations under complex conditions. Therefore, inspired by the diversity of neurons in the human brain, we propose an innovative DMixNet, a dendritic multi-layered perceptron architecture. Rooted in the theory of dendritic neurons from neuroscience, we propose a dendritic neural unit (DNU) that enhances DMixNet with stronger biological interpretability and more robust nonlinear processing capabilities. The flexibility of dendritic structures allows the DNU to adjust its architecture to achieve different functionalities. Based on the DNU, we propose a novel channel fusion network <span>(text {DNU}_text {E})</span> and a dendritic classifier <span>(text {DNU}_text {C})</span>. The <span>(text {DNU}_text {E})</span> substitutes the traditional two fully connected layers as the channel mixer, constructing a dendritic mixer layer to enhance the fusion capability of channel information within the entire framework. Meanwhile, the <span>(text {DNU}_text {C})</span> replaces the traditional linear classifier, effectively improving the model’s classification performance. Experimental results demonstrate that DMixNet achieves improvements of 2.13%, 4.79%, 4.71%, 23.14% on the CIFAR-10, CIFAR-100, Tiny-ImageNet and COIL-100 benchmark image recognition datasets, respectively, as well as a 14.78% enhancement on the medical image classification dataset PathMNIST, outperforming other state-of-the-art architectures. Code is available at https://github.com/KarilynXu/DMixNet.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 5","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11123-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143396603","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":"Text analytics for co-creation in public sector organizations: a literature review-based research framework","authors":"Nina Rizun,&nbsp;Aleksandra Revina,&nbsp;Noella Edelmann","doi":"10.1007/s10462-025-11112-1","DOIUrl":"10.1007/s10462-025-11112-1","url":null,"abstract":"<div><p>The public sector faces considerable challenges that stem from increasing external and internal demands, the need for diverse and complex services, and citizens’ lack of satisfaction and trust in public sector organisations (PSOs). An alternative to traditional public service delivery is the co-creation of public services. Data analytics has been fueled by the availability of immense amounts of data, including textual data, and techniques to analyze data, so it has immense potential to foster data-driven solutions for the public sector. In the paper, we systematically review the existing literature on the application of Text Analytics (TA) techniques on textual data that can support public service co-creation. In this review, we identify the TA techniques, the public services and the co-creation phase they support, as well as envisioned public values for the stakeholder groups. On the basis of the analysis, we develop a Research Framework that helps to structure the TA-enabled co-creation process in PSOs, increases awareness among public sector organizations and stakeholders on the significant potential of TA in creating value, and provides scholars with some avenues for further research.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11112-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143184714","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":"Transformer-based image and video inpainting: current challenges and future directions","authors":"Omar Elharrouss,&nbsp;Rafat Damseh,&nbsp;Abdelkader Nasreddine Belkacem,&nbsp;Elarbi Badidi,&nbsp;Abderrahmane Lakas","doi":"10.1007/s10462-024-11075-9","DOIUrl":"10.1007/s10462-024-11075-9","url":null,"abstract":"<div><p>Image inpainting is currently a hot topic within the field of computer vision. It offers a viable solution for various applications, including photographic restoration, video editing, and medical imaging. Deep learning advancements, notably convolutional neural networks (CNNs) and generative adversarial networks (GANs), have significantly enhanced the inpainting task with an improved capability to fill missing or damaged regions in an image or a video through the incorporation of contextually appropriate details. These advancements have improved other aspects, including efficiency, information preservation, and achieving both realistic textures and structures. Recently, Vision Transformers (ViTs) have been exploited and offer some improvements to image or video inpainting. The advent of transformer-based architectures, which were initially designed for natural language processing, has also been integrated into computer vision tasks. These methods utilize self-attention mechanisms that excel in capturing long-range dependencies within data; therefore, they are particularly effective for tasks requiring a comprehensive understanding of the global context of an image or video. In this paper, we provide a comprehensive review of the current image/video inpainting approaches, with a specific focus on Vision Transformer (ViT) techniques, with the goal to highlight the significant improvements and provide a guideline for new researchers in the field of image/video inpainting using vision transformers. We categorized the transformer-based techniques by their architectural configurations, types of damage, and performance metrics. Furthermore, we present an organized synthesis of the current challenges, and suggest directions for future research in the field of image or video inpainting.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11075-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143184831","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":"New fuzzy zeroing neural network with noise suppression capability for time-varying linear equation solving","authors":"Dongsheng Guo,&nbsp;Chan Zhang,&nbsp;Naimeng Cang,&nbsp;Xiyuan Zhang,&nbsp;Lin Xiao,&nbsp;Zhongbo Sun","doi":"10.1007/s10462-024-11026-4","DOIUrl":"10.1007/s10462-024-11026-4","url":null,"abstract":"<div><p>Recently, the zeroing neural network (ZNN) with continuous/discrete-time forms has realized success in solving the time-varying linear equation (TVLE). In this paper, we provide a further investigation by proposing a new fuzzy zeroing neural network (FZNN) model to solve the TVLE in noisy environment. Such a FZNN model, which has the capability of suppressing noise, is developed by using the integration enhancement and fuzzy control strategy. Then, theoretical analysis is presented to show that the proposed FZNN model can effectively solve the TVLE, even with the existence of noise. Comparative simulation results through different examples further verify the effectiveness and robustness of the proposed FZNN model on TVLE solving.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11026-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143184715","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":"Tornado optimizer with Coriolis force: a novel bio-inspired meta-heuristic algorithm for solving engineering problems","authors":"Malik Braik,&nbsp;Heba Al-Hiary,&nbsp;Hussein Alzoubi,&nbsp;Abdelaziz Hammouri,&nbsp;Mohammed Azmi Al-Betar,&nbsp;Mohammed A. Awadallah","doi":"10.1007/s10462-025-11118-9","DOIUrl":"10.1007/s10462-025-11118-9","url":null,"abstract":"<div><p>This paper proposes a new meta-heuristic algorithm named tornado optimizer with Coriolis force (TOC) which is applied to solve global optimization and constrained engineering problems in continuous search spaces. The fundamental concepts and ideas beyond the proposed TOC Optimizer are inspired by nature based on the observation of the cycle process of tornadoes and how thunderstorms and windstorms evolve into tornadoes using Coriolis force. The Coriolis force is applied to windstorms that directly evolve to form tornadoes based on the developed optimization method. The proposed TOC algorithm mathematically models and implements the behavioral steps of tornado formation by windstorms and thunderstorms and then dissipation of tornadoes on the ground. These steps ultimately lead to feasible solutions when applied to solve optimization problems. These behavioral steps are mathematically represented along with the Coriolis force to allow for a proper balance between exploration and exploitation during the optimization process, as well as to allow search agents to explore and exploit every possible area of the search space. The performance of the proposed TOC optimizer was thoroughly examined on a simple benchmark set of 23 test functions, and a set of 29 well-known benchmark functions from the CEC-2017 test for a variety of dimensions. A comparative study of the computational and convergence analysis results was carried out to clarify the efficacy and stability levels of the proposed TOC optimizer compared to other well-known optimizers. The TOC optimizer outperformed other comparative algorithms using the mean ranks of Friedman’s test by 20.75%, 27.248%, and 25.85% on the 10-, 30-, and 50-dimensional CEC 2017 test set, respectively. The reliability and appropriateness of the TOC optimizer were examined by solving real-world problems including eight engineering design problems and one industrial process. The proposed optimizer divulged satisfactory performance over other competing optimizers regarding solution quality and global optimality as per statistical test methods.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11118-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143184830","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 Hybridization and Optimization of DNNs for Medical Imaging: A Survey on Disease Detection Techniques","authors":"Maneet Kaur Bohmrah,&nbsp;Harjot Kaur","doi":"10.1007/s10462-024-11049-x","DOIUrl":"10.1007/s10462-024-11049-x","url":null,"abstract":"<div><p>Due to the high classification accuracy and fast computational speed offered by Deep Neural Networks (DNNs), they have been widely used for the design and development of automated Artificial Intelligence (AI) tools for the detection of various diseases. These tools, which are intensive computational learning models, hold tremendous significance in healthcare for identifying various diseases. The primary goal of this review is to understand the applicability and methodology for implementing DNNs, including computational costs, for the classification of distinct diseases from disparate medical imaging datasets. This study presents an extensive survey of DNNs along with their various hybridization forms. To achieve this, the research papers surveyed have been grouped into five categories: pretrained DNNs, hyperparameter-tuned optimized DNNs, hybrid DNNs and ML classifiers, hybrid models with optimization techniques, and meta-heuristics based feature selection DNNs. The major part of this review highlights the significant role of nature-inspired meta-heuristic techniques used for hyperparameter optimization or feature selection algorithms of DNNs. Besides the frameworks and computational costs, descriptions of disparate medical image datasets and image preprocessing techniques have also been discussed under each category. Furthermore, a comparative analysis for each category has been performed on the basis of different parameters, including the type and size of datasets used, image preprocessing, methodology (as per the mentioned category), and performance (in terms of classification accuracy). This study also presents a bibliometric analysis based on the publication count of various articles related to hyperparameter-tuned optimized DNNs and meta-heuristic based feature selection DNNs. This review aims to assist potential AI researchers in choosing the most sound and appropriate DNN-based techniques for disease detection and prediction, all consolidated into a one single research paper.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11049-x.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143107804","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 optic disc segmentation methods in adult and pediatric retinal images: from conventional methods to artificial intelligence (CR-ODSeg-AP-CM2AI)","authors":"Avinash Bansal,&nbsp;Jan Kubíček,&nbsp;Marek Penhaker,&nbsp;Martin Augustynek","doi":"10.1007/s10462-024-11056-y","DOIUrl":"10.1007/s10462-024-11056-y","url":null,"abstract":"<div><p>This review, titled CR-ODSeg-AP-CM2AI (Comprehensive Review of Optic Disc Segmentation in Adult and Pediatric Retinal Images: From Conventional Methods to Artificial Intelligence), explores optic disc segmentation techniques for adult and pediatric retinal images. It emphasizes the clinical implications of these techniques in diagnosing and monitoring retinal diseases across diverse populations. We systematically categorize each segmentation method, comparing traditional approaches with advancements in artificial intelligence (AI) to highlight innovative hybrid techniques that enhance segmentation accuracy and efficiency. This review also discusses evaluation metrics and the use of larger datasets to provide insights into the effectiveness and robustness of these methods.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11056-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143107947","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":"Review of filtering based feature selection for Botnet detection in the Internet of Things","authors":"Mohamed Saied,&nbsp;Shawkat Guirguis,&nbsp;Magda Madbouly","doi":"10.1007/s10462-025-11113-0","DOIUrl":"10.1007/s10462-025-11113-0","url":null,"abstract":"<div><p>Botnets are a major security threat in the Internet of Things (IoT), posing significant risks to user privacy, network availability, and the integrity of IoT devices. With the increasing availability of large datasets that contain hundreds or even thousands of variables, selecting the right set of features can be a challenging task. Feature selection is a critical step in developing effective machine learning-based botnet detection systems, as it enables the selection of a subset of features that are most relevant for detection. This paper provides a comprehensive review of filtering based feature selection techniques for botnet detection in IoT. It examines a range of filtering based techniques and evaluates their effectiveness in addressing the challenges and limitations of botnet detection in IoT. It aims to identify the gaps in the literature and areas for future research, and discuss the broader implications of findings for the field of IoT botnet detection. This review provides valuable insights and guidance for researchers and practitioners working on botnet detection in IoT, and highlights the importance of effective feature selection in developing robust and reliable detection systems.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-025-11113-0.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109863","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}