Artificial Intelligence Review最新文献

{"title":"Advances and challenges in learning from experience replay","authors":"Daniel Eugênio Neves,&nbsp;Lucila Ishitani,&nbsp;Zenilton Kleber Gonçalves do Patrocínio Júnior","doi":"10.1007/s10462-024-11062-0","DOIUrl":"10.1007/s10462-024-11062-0","url":null,"abstract":"<div><p>From the first theoretical propositions in the 1950s to its application in real-world problems, Reinforcement Learning (RL) is still a fascinating and complex class of machine learning algorithms with overgrowing literature in recent years. In this work, we present an extensive and structured literature review and discuss how the Experience Replay (ER) technique has been fundamental in making various RL methods in most relevant problems and different domains more data efficient. ER is the central focus of this review. One of its main contributions is a taxonomy that organizes the many research works and the different RL methods that use ER. Here, the focus is on how RL methods improve and apply ER strategies, demonstrating their specificities and contributions while having ER as a prominent component. Another relevant contribution is the organization in a facet-oriented way, allowing different perspectives of reading, whether based on the fundamental problems of RL, focusing on algorithmic strategies and architectural decisions, or with a view to different applications of RL with ER. Moreover, we start by presenting a detailed formal theoretical foundation of RL and some of the most relevant algorithms and bring from the recent literature some of the main trends, challenges, and advances focused on ER formal basement and how to improve its propositions to make it even more efficient in different methods and domains. Lastly, we discuss challenges and open problems and present relevant paths to feature works.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11062-0.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859661","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 survey on deep learning-based automated essay scoring and feedback generation","authors":"Haile Misgna,&nbsp;Byung-Won On,&nbsp;Ingyu Lee,&nbsp;Gyu Sang Choi","doi":"10.1007/s10462-024-11017-5","DOIUrl":"10.1007/s10462-024-11017-5","url":null,"abstract":"<div><p>Deep learning-based automated essay scoring (AES) models exhibit a remarkable ability to identify complex patterns within essays and then generate accurate score predictions in an end-to-end training fashion. However, these models face a critical limitation in explaining the specific patterns and features utilized for scoring, which are essential for interpreting the scores and offering constructive feedback to essay authors. Numerous studies have focused on essay scoring, with the aim of modeling prompt-specific, domain-adaptable, or trait-specific AES. While existing surveys on AES cover topics ranging from representation to scoring models, they primarily emphasize scoring models. This study addresses a crucial gap by encompassing research on feedback generation for essay assessment tasks. By delving into essay scoring and feedback generation, we synthesize several existing literature to provide readers with a comprehensive understanding of ongoing research in both deep learning-based essay scoring and automated feedback generation. We categorized the existing essay scoring studies into prompt-specific and cross-prompt AES models, noting that prompt-specific AES is extensively researched category. However, we have only come across a few studies concerning automated feedback generation, likely because of the limited availability of suitable datasets for researching such types of tasks. Moreover, this survey provides insights into approaches for essay representation, prevalent datasets, evaluation metrics, and challenges in automated essay scoring tasks. By shedding light on these aspects, our goal is to delineate the current landscape, identify key research directions, and pave the way for further advancements in automated essay assessment.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11017-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142858659","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":"Revisiting non-learned operators based deep learning for image classification: a lightweight directional-aware network","authors":"Yuwei Guo,&nbsp;Wenhao Zhang,&nbsp;Yupeng Gao,&nbsp;Licheng Jiao,&nbsp;Shuo Wang,&nbsp;Jiabo Du,&nbsp;Fang Liu","doi":"10.1007/s10462-024-11038-0","DOIUrl":"10.1007/s10462-024-11038-0","url":null,"abstract":"<div><p>Due to the stable feature representation capability provided by non-learned operators, the integration with deep learning models, i.e., non-learned operator based deep learning models, has become a paradigm, however, performance-wise, it is still not promising. In this paper, by revisiting non-learned operator based deep learning models, we reveal the reasons for their underperformance: lack of geometric invariance, insufficient sparsity, and neglect of directional importance. In response, we present a Lightweight Directional-Aware Network (LDAN) for image classification. Specifically, to generate sparse geometric-invariant features, we propose a ShearletNet to capture multi-directional features in three different levels. Then, a Directional-Aware module is designed to highlight the discriminative multi-directional features and generate multi-scale features. Finally, a Pointwise Convolution module is used to integrate the multi-directional features with the multi-scale ones for reducing the computational resources. Experiments on the commonly used CIFAR10, CIFAR100, Self-Taught Learning 10 (STL10), and Tiny ImageNet datasets demonstrate the efficiency and effectiveness of the proposed LDAN. Compared to the existing non-learned operator based models, LDAN reduces the parameter count by 80.83% while achieving a 6.32% increase in accuracy.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11038-0.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859647","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":"Faster independent vector analysis with joint pairwise updates of demixing vectors","authors":"Zhongqiang Luo,&nbsp;Ruiming Guo,&nbsp;Ling Wang","doi":"10.1007/s10462-024-11061-1","DOIUrl":"10.1007/s10462-024-11061-1","url":null,"abstract":"<div><p>To achieve more efficient blind separation of multi-channel speech signals, this paper proposes a new algorithm for blind source separation(BSS) of sound sources using auxiliary function-based independent vector analysis (AuxIVA) with joint pairwise updates of demixing vectors. This algorithm is better than AuxIVA using iterative projection with adjustment (AuxIVA-IPA) when separating multiple sources. The IPA method jointly executes iterative projection (IP) and iterative source steering (ISS) to update and updates one row and one column of the separation matrix in each iteration. On this basis, IPA is extended to jointly execute IP2 and ISS2 for updating, which can update two rows and two columns of the separation matrix in each iteration. Accordingly, this proposed method is named by IPA2. Furthermore, it can optimize the same cost function as IPA while maintaining the same time complexity. Finally, the convolutional speech separation experiments are conducted to validate the effectiveness and efficiency of the proposed method. The experimental results corroborate that compared with the state-of-the-art IP, IP2, ISS, ISS2, and IPA used in AuxIVA, the IPA2 method acquires faster convergence speed and better separation performance, enabling the cost function to reach the convergence interval faster and maintaining good separation results.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11061-1.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859719","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":"Computational humor recognition: a systematic literature review","authors":"Antonios Kalloniatis,&nbsp;Panagiotis Adamidis","doi":"10.1007/s10462-024-11043-3","DOIUrl":"10.1007/s10462-024-11043-3","url":null,"abstract":"<div><p>Computational humor recognition is considered to be one of the hardest tasks in natural language processing (NLP) since humor is such a particularly complex emotion. There are very few recent studies that offer an analysis of certain aspects of computational humor. However, there has been no attempt to study the empirical evidence on computational humor recognition in a systematic way. The aim of this research is to examine computational humor detection from three aspects: datasets, features and algorithms. Therefore, a Systematic Literature Review (SLR) was carried out to present in detail the computational techniques for humor identification under these aspects. After posing some research questions, a total of 106 primary papers were identified as relevant to the objectives of these questions and further detailed analysis was conducted. The study revealed that there are a great number of publicly available annotated humor datasets with many different types of humor instances. Twenty-one (21) humor features have been carefully studied, and research evidence of their use in humor computational detection is presented. Additionally, a classification of the humor detection approaches was performed, and the results are presented. Finally, the challenges of applying these techniques to humor recognition as well as promising future research directions are discussed.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11043-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859523","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":"Enhancement of satellite images based on CLAHE and augmented elk herd optimizer","authors":"Malik Braik,&nbsp;Mohammed Azmi Al-Betar,&nbsp;Mohammed A. Mahdi,&nbsp;Mohammed Al-Shalabi,&nbsp;Shahanawaj Ahamad,&nbsp;Sawsan A. Saad","doi":"10.1007/s10462-024-11022-8","DOIUrl":"10.1007/s10462-024-11022-8","url":null,"abstract":"&lt;div&gt;&lt;p&gt;Satellite images often have very narrow brightness value ranges, so it is necessary to enhance the contrast and brightness, maintain the quality of visual information, and preserve pertinent details in the images before conducting additional analysis. This is because improving the brightness and contrast of images is crucial to image processing and analysis as it makes it easier for people to identify and comprehend the images. The Incomplete Beta Function (IBF) is a popular transformation function for Image Contrast Enhancement (ICE). Nevertheless, IBF has modest efficiency in parameter selection, a small set of adjustable parameters for stretching regions with high or low gray levels, and image enhancement is almost ineffective with stretching at either end. Meta-heuristic algorithms have been utilized efficiently and effectively over the past few decades to solve complicated image processing problems. This paper presents an Augmented version of the Elk Herd Optimizer (AEHO) combined with other traditional ICE techniques to improve edge details, entropy, local contrast, and local brightness of low-contrast natural and satellite images. The AEHO method employs a multi-stage strategic procedure, where its mathematical model undergoes several enhancements before being applied to ICE to allow for further exploration and exploitation of its features. This method uses a pre-established fitness criterion for the purpose of optimizing a set of parameters to rework a well-known transformation function and an effective assessment technique as an objective standard for this purpose. In the proposed image enhancement model, contrast limited adaptive histogram equalization was first applied as a prior step to ameliorate the color intensity. Then, the optimal IBF’s parameters for ICE were adaptively determined using AEHO. After that, bilateral gamma correction was used to improve the visual quality of images without sacrificing edge details or natural color quality. The proposed AEHO-based image enhancement model is tested on natural scenes, certain standard images, and publicly available satellite images. In addition to other five techniques built on based on pre-existing meta-heuristics, the performance of the proposed method was compared against other well-known state-of-the-art image enhancement algorithms. The objective evaluation of the enhancement algorithms was achieved utilizing a variety of full-reference, no-reference, and pertinent performance evaluation norms. The experimental findings illustrated that the proposed image enhancement method can successfully outperform several other algorithms that employed the same image enhancement model as AEHO in addition to other conventional image enhancement methods included for comparison. The results on ten natural and satellite color images showed that the presented method performs better than all other comparative methods in the corresponding evaluation criteria in terms of average peak signal-t","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11022-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859489","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":"Graph reduction techniques for instance selection: comparative and empirical study","authors":"Zahiriddin Rustamov,&nbsp;Nazar Zaki,&nbsp;Jaloliddin Rustamov,&nbsp;Ayham Zaitouny,&nbsp;Rafat Damseh","doi":"10.1007/s10462-024-10971-4","DOIUrl":"10.1007/s10462-024-10971-4","url":null,"abstract":"<div><p>The surge in data generation has prompted a shift to big data, challenging the notion that “more data equals better performance” due to processing and time constraints. In this evolving artificial intelligence and machine learning landscape, instance selection (IS) has become crucial for data reduction without compromising model quality. Traditional IS methods, though efficient, often struggle with large, complex datasets in data mining. This study evaluates graph reduction techniques, grounded in graph theory, as a novel approach for instance selection. The objective is to leverage the inherent structures of data represented as graphs to enhance the effectiveness of instance selection. We evaluated 35 graph reduction techniques across 29 classification datasets. These techniques were assessed based on various metrics, including accuracy, F1 score, reduction rate, and computational times. Graph reduction methods showed significant potential in maintaining data integrity while achieving substantial reductions. Top techniques achieved up to 99% reduction while maintaining or improving accuracy. For instance, the Multilevel sampling achieved an accuracy effectiveness score of 0.8555 with 99.16% reduction on large datasets, while Leiden sampling showed high effectiveness on smaller datasets (0.8034 accuracy, 97.87% reduction). Computational efficiency varied widely, with reduction times ranging from milliseconds to minutes. This research advances the theory of graph-based instance selection and offers practical application guidelines. Our findings indicate graph reduction methods effectively preserve data quality and boost processing efficiency in large, complex datasets, with some techniques achieving up to 160-fold speedups in model training at high reduction rates.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-10971-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859720","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 industrial robot selection through a hybrid novel approach: integrating CRITIC-VIKOR method with probabilistic uncertain linguistic q-rung orthopair fuzzy","authors":"Sumera Naz,&nbsp;Muhammad Muneeb ul Hassan,&nbsp;Atif Mehmood,&nbsp;Gabriel Piñeres Espitia,&nbsp;Shariq Aziz Butt","doi":"10.1007/s10462-024-11001-z","DOIUrl":"10.1007/s10462-024-11001-z","url":null,"abstract":"<div><p>The increasing complexity and novel features of industrial robots have made their selection for specific applications a challenging task. Decision-makers are faced with the daunting task of navigating through various attributes and specifications, often under conditions of ambiguity and uncertainty. To assist in this complex decision-making process, this paper introduces a novel decision-making framework based on probabilistic uncertain linguistic <i>q</i>-rung orthopair fuzzy sets (PUL<i>q</i>-ROFS). This framework effectively combines the strengths of probabilistic uncertain linguistic term set (PULTS) and <i>q</i>-rung orthopair fuzzy set (<i>q</i>-ROFS) to provide a more robust approach for handling ambiguity and uncertainty in the robot selection process. The proposed methodology integrates a multi-attribute group decision-making (MAGDM) approach. This approach utilizes the VIseKriterijumska Optimizacija I KOmpromisno Resenje (VIKOR) method in conjunction with the criterion importance via inter-criteria correlation (CRITIC) method. The CRITIC method determines attribute weights by analyzing both the differences and contrast intensity of criteria, thereby accounting for the relative strength and conflict among them. VIKOR is then employed to aggregate individual regret and group utility, resulting in a compromise solution that guides decision-makers toward the optimal robot selection. Proposed method provide the clarity and confidence to decision makers for choice of attributes and crediting this enhancement to the framework’s transparency and its ability to incorporate a wide range of stakeholder perspectives. This framework facilitates a more inclusive decision-making process that acknowledges differing viewpoints and preferences. This proposed approach not only directs users toward optimal selections but also encourages collaboration among decision-makers, promoting a sense of shared ownership and responsibility in the selection process. This paper select the best robot by utilizing the benefits of both CRITIC and VIKOR method. The effectiveness of this integrated approach is validated through parameter and comparative analysis. The results demonstrate the potential applicability of the proposed methodology in real-world industrial robot selection scenarios, providing decision-makers with a powerful tool to navigate the complexities of modern robotic systems.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11001-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859696","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":"Specification overfitting in artificial intelligence","authors":"Benjamin Roth,&nbsp;Pedro Henrique Luz de Araujo,&nbsp;Yuxi Xia,&nbsp;Saskia Kaltenbrunner,&nbsp;Christoph Korab","doi":"10.1007/s10462-024-11040-6","DOIUrl":"10.1007/s10462-024-11040-6","url":null,"abstract":"<div><p>Machine learning (ML) and artificial intelligence (AI) approaches are often criticized for their inherent bias and for their lack of control, accountability, and transparency. Consequently, regulatory bodies struggle with containing this technology’s potential negative side effects. High-level requirements such as fairness and robustness need to be formalized into concrete specification metrics, imperfect proxies that capture isolated aspects of the underlying requirements. Given possible trade-offs between different metrics and their vulnerability to over-optimization, integrating specification metrics in system development processes is not trivial. This paper defines <i>specification overfitting</i>, a scenario where systems focus excessively on specified metrics to the detriment of high-level requirements and task performance. We present an extensive literature survey to categorize how researchers propose, measure, and optimize specification metrics in several AI fields (e.g., natural language processing, computer vision, reinforcement learning). Using a keyword-based search on papers from major AI conferences and journals between 2018 and mid-2023, we identify and analyze 74 papers that propose or optimize specification metrics. We find that although most papers implicitly address specification overfitting (e.g., by reporting more than one specification metric), they rarely discuss which role specification metrics should play in system development or explicitly define the scope and assumptions behind metric formulations.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11040-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859646","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 end-to-end occluded person re-identification network with smoothing corrupted feature prediction","authors":"Caijie Zhao,&nbsp;Ying Qin,&nbsp;Bob Zhang,&nbsp;Yajie Zhao,&nbsp;Baoyun Wu","doi":"10.1007/s10462-024-11047-z","DOIUrl":"10.1007/s10462-024-11047-z","url":null,"abstract":"<div><p>Occluded person re-identification (ReID) is a challenging task as the images suffer from various obstacles and less discriminative information caused by incomplete body parts. Most current works rely on auxiliary models to infer the visible body parts and partial-level features matching to overcome the contaminated body information, which consumes extra inference time and fails when facing complex occlusions. More recently, some methods utilized masks provided from image occlusion augmentation (OA) for the supervision of mask learning. These works estimated occlusion scores for each part of the image by roughly dividing it in the horizontal direction, but cannot accurately predict the occlusion, as well as failing in vertical occlusions. To address this issue, we proposed a Smoothing Corrupted Feature Prediction (SCFP) network in an end-to-end manner for occluded person ReID. Specifically, aided by OA that simulates occlusions appearing in pedestrians and providing occlusion masks, the proposed Occlusion Decoder and Estimator (ODE) estimates and eliminates corrupted features, which is supervised by mask labels generated via restricting all occlusions into a group of patterns. We also designed an Occlusion Pattern Smoothing (OPS) to improve the performance of ODE when predicting irregular obstacles. Subsequently, a Local-to-Body (L2B) representation is constructed to mitigate the limitation of the partial body information for final matching. To investigate the performance of SCFP, we compared our model to the existing state-of-the-art methods in occluded and holistic person ReID benchmarks and proved that our method achieves superior results over the state-of-the-art methods. We also achieved the highest Rank-1 accuracies of 70.9%, 87.0%, and 93.2% in Occluded-Duke, Occluded-ReID, and P-DukeMTMC, respectively. Furthermore, the proposed SCFP generalizes well in holistic datasets, yielding accuracies of 95.8% in Market-1510 and 90.7% in DukeMTMC-reID.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 2","pages":""},"PeriodicalIF":10.7,"publicationDate":"2024-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11047-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142859657","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}