Artificial Intelligence Review最新文献

{"title":"Metaheuristic optimization algorithms for multi-area economic dispatch of power systems: Part I—a comprehensive survey","authors":"Yang Wang,&nbsp;Guojiang Xiong","doi":"10.1007/s10462-024-11070-0","DOIUrl":"10.1007/s10462-024-11070-0","url":null,"abstract":"<div><p>Multi-area economic dispatch (MAED) provides an indispensable component for the security and economic operation of contemporary power systems. Over recent years, numerous metaheuristic optimization algorithms (MOAs) have surfaced for addressing the MAED problem. However, none of the literature to date conducted a comprehensive statistical research work on the MAED problem. In part I of this series, we present a comprehensive survey on this problem. (1) We collect all eleven reported MAED cases studied over the years. These cases have different structures, scales, and constraints. We illustrate the structures of all cases and provide their corresponding system parameters. (2) We collect all the MOA solution algorithms. These algorithms are inspired by different ways, and we categorize them in detail and review them comprehensively. (3) We list the detailed applications of MOAs on different cases and count the percentage of studies on each case. (4) Finally, we summarize the current research progress and point out the future research directions in terms of MAED models and solution methods, respectively. This survey provides an extensive overview of the MAED cases and its solution methods. It can provide applicable and reference suggestions for future research on the MAED problem.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11070-0.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109470","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":"Intelligent Cinematography: a review of AI research for cinematographic production","authors":"Adrian Azzarelli,&nbsp;Nantheera Anantrasirichai,&nbsp;David R. Bull","doi":"10.1007/s10462-024-11089-3","DOIUrl":"10.1007/s10462-024-11089-3","url":null,"abstract":"<div><p>This paper offers the first comprehensive review of artificial intelligence (AI) research in the context of real camera content acquisition for entertainment purposes and is aimed at both researchers and cinematographers. Addressing the lack of review papers in the field of <i>intelligent cinematography</i> (IC) and the breadth of related computer vision research, we present a holistic view of the IC landscape while providing technical insight, important for experts across disciplines. We provide technical background on generative AI, object detection, automated camera calibration and 3-D content acquisition, with references to assist non-technical readers. The application sections categorize work in terms of four production types: General Production, Virtual Production, Live Production and Aerial Production. Within each application section, we (1) sub-classify work according to research topic and (2) describe the trends and challenges relevant to each type of production. In the final chapter, we address the greater scope of IC research and summarize the significant potential of this area to influence the creative industries sector. We suggest that work relating to virtual production has the greatest potential to impact other mediums of production, driven by the growing interest in LED volumes/stages for in-camera virtual effects (ICVFX) and automated 3-D capture for virtual modeling of real world scenes and actors. We also address ethical and legal concerns regarding the use of creative AI that impact on artists, actors, technologists and the general public.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 4","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11089-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143109542","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":"Edge deep learning in computer vision and medical diagnostics: a comprehensive survey","authors":"Yiwen Xu,&nbsp;Tariq M. Khan,&nbsp;Yang Song,&nbsp;Erik Meijering","doi":"10.1007/s10462-024-11033-5","DOIUrl":"10.1007/s10462-024-11033-5","url":null,"abstract":"<div><p>Edge deep learning, a paradigm change reconciling edge computing and deep learning, facilitates real-time decision making attuned to environmental factors through the close integration of computational resources and data sources. Here we provide a comprehensive review of the current state of the art in edge deep learning, focusing on computer vision applications, in particular medical diagnostics. An overview of the foundational principles and technical advantages of edge deep learning is presented, emphasising the capacity of this technology to revolutionise a wide range of domains. Furthermore, we present a novel categorisation of edge hardware platforms based on performance and usage scenarios, facilitating platform selection and operational effectiveness. Following this, we dive into approaches to effectively implement deep neural networks on edge devices, encompassing methods such as lightweight design and model compression. Reviewing practical applications in the fields of computer vision in general and medical diagnostics in particular, we demonstrate the profound impact edge-deployed deep learning models can have in real-life situations. Finally, we provide an analysis of potential future directions and obstacles to the adoption of edge deep learning, with the intention to stimulate further investigations and advancements of intelligent edge deep learning solutions. This survey provides researchers and practitioners with a comprehensive reference shedding light on the critical role deep learning plays in the advancement of edge computing applications.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 3","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11033-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142994861","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 taxonomy of literature reviews and experimental study of deepreinforcement learning in portfolio management","authors":"Mohadese Rezaei,&nbsp;Hossein Nezamabadi-Pour","doi":"10.1007/s10462-024-11066-w","DOIUrl":"10.1007/s10462-024-11066-w","url":null,"abstract":"<div><p>Portfolio management involves choosing and actively overseeing various investment assets to meet an investor’s long-term financial goals, considering their risk tolerance and desired return potential. Traditional methods, like mean–variance analysis, often lack the flexibility needed to navigate the complexities of today’s financial markets. Recently, Deep Reinforcement Learning (DRL) has emerged as a promising approach, enabling continuous adjustments to investment strategies based on market feedback without explicit price predictions. This paper presents a comprehensive literature review of DRL applications in portfolio management, aimed at finance researchers, data scientists, AI experts, FinTech engineers, and students seeking advanced portfolio optimization methodologies. We also conducted an experimental study to evaluate five DRL algorithms—Advantage Actor-Critic (A2C), Deep Deterministic Policy Gradient (DDPG), Proximal Policy Optimization (PPO), Soft Actor-Critic (SAC), and Twin Delayed DDPG (TD3)—in managing a portfolio of 30 Dow Jones Industrial Average (DJIA) stocks. Their performance is compared with the DJIA index and traditional strategies, demonstrating DRL’s potential to improve portfolio outcomes while effectively managing risk.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 3","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11066-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142994962","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning and computer vision in plant disease detection: a comprehensive review of techniques, models, and trends in precision agriculture","authors":"Abhishek Upadhyay,&nbsp;Narendra Singh Chandel,&nbsp;Krishna Pratap Singh,&nbsp;Subir Kumar Chakraborty,&nbsp;Balaji M. Nandede,&nbsp;Mohit Kumar,&nbsp;A. Subeesh,&nbsp;Konga Upendar,&nbsp;Ali Salem,&nbsp;Ahmed Elbeltagi","doi":"10.1007/s10462-024-11100-x","DOIUrl":"10.1007/s10462-024-11100-x","url":null,"abstract":"<div><p>Plant diseases cause significant damage to agriculture, leading to substantial yield losses and posing a major threat to food security. Detection, identification, quantification, and diagnosis of plant diseases are crucial parts of precision agriculture and crop protection. Modernizing agriculture and improving production efficiency are significantly affected by using computer vision technology for crop disease diagnosis. This technology is notable for its non-destructive nature, speed, real-time responsiveness, and precision. Deep learning (DL), a recent breakthrough in computer vision, has become a focal point in agricultural plant protection that can minimize the biases of manually selecting disease spot features. This study reviews the techniques and tools used for automatic disease identification, state-of-the-art DL models, and recent trends in DL-based image analysis. The techniques, performance, benefits, drawbacks, underlying frameworks, and reference datasets of more than 278 research articles were analyzed and subsequently highlighted in accordance with the architecture of computer vision and deep learning models. Key findings include the effectiveness of imaging techniques and sensors like RGB, multispectral, and hyperspectral cameras for early disease detection. Researchers also evaluated various DL architectures, such as convolutional neural networks, vision transformers, generative adversarial networks, vision language models, and foundation models. Moreover, the study connects academic research with practical agricultural applications, providing guidance on the suitability of these models for production environments. This comprehensive review offers valuable insights into the current state and future directions of deep learning in plant disease detection, making it a significant resource for researchers, academicians, and practitioners in precision agriculture.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 3","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11100-x.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142994860","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 securing federated learning with IDS: a comprehensive review of neural networks and feature engineering techniques for malicious client detection","authors":"Naila Latif,&nbsp;Wenping Ma,&nbsp;Hafiz Bilal Ahmad","doi":"10.1007/s10462-024-11082-w","DOIUrl":"10.1007/s10462-024-11082-w","url":null,"abstract":"<div><p>Federated Learning (FL) is a technique that can learn a global machine-learning model at a central server by aggregating locally trained models. This distributed machine-learning approach preserves the privacy of local models. However, FL systems are inherently vulnerable to significant security challenges such as cyber-attacks, handling non-independent and identically distributed (non-IID) data, and data privacy concerns. This systematic literature review addresses these issues by examining advanced neural network models, feature engineering methods, and privacy-preserving techniques within intrusion detection systems (IDS) for FL environments. These are key elements for improving the security of FL systems. To the best of our knowledge, this review is among the first to comprehensively explore the combined impacts of these technologies. We analyzed 88 studies published between 2021 and October 2024. This study offers valuable insights for future research directions, including scaling FL in a real-world environment.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 3","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11082-w.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142963029","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":"Textual variations in social media text processing applications: challenges, solutions, and trends","authors":"Jebran Khan,&nbsp;Kashif Ahmad,&nbsp;Senthil Kumar Jagatheesaperumal,&nbsp;Kyung-Ah Sohn","doi":"10.1007/s10462-024-11071-z","DOIUrl":"10.1007/s10462-024-11071-z","url":null,"abstract":"<div><p>Being an informal communication source, social media text is susceptible to several intentional and unintentional textual variations. These variations lead to various out-of-vocabulary (OOV) words, making social media text processing more challenging. This work analyses and discusses such challenges by providing a detailed overview of different sources of intentional and unintentional OOV words and associated challenges. We provide a detailed survey of pre-processing techniques, including traditional and application-specific methods proposed in the literature to handle intentional and unintentional textual variations, while highlighting their pros and cons. The paper analyses the implications of text normalization (standardization) in different social media text-processing applications. Moreover, the paper provides an overview of the recent challenges and trends in handling social media textual variations, and it is expected to provide a baseline for future research.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 3","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11071-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142963028","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":"Digital forgetting in large language models: a survey of unlearning methods","authors":"Alberto Blanco-Justicia,&nbsp;Najeeb Jebreel,&nbsp;Benet Manzanares-Salor,&nbsp;David Sánchez,&nbsp;Josep Domingo-Ferrer,&nbsp;Guillem Collell,&nbsp;Kuan Eeik Tan","doi":"10.1007/s10462-024-11078-6","DOIUrl":"10.1007/s10462-024-11078-6","url":null,"abstract":"<div><p>Large language models (LLMs) have become the state of the art in natural language processing. The massive adoption of generative LLMs and the capabilities they have shown have prompted public concerns regarding their impact on the labor market, privacy, the use of copyrighted work, and how these models align with human ethics and the rule of law. As a response, new regulations are being pushed, which require developers and service providers to evaluate, monitor, and forestall or at least mitigate the risks posed by their models. One mitigation strategy is digital forgetting: given a model with undesirable knowledge or behavior, the goal is to obtain a new model where the detected issues are no longer present. Digital forgetting is usually enforced via machine unlearning techniques, which modify trained machine learning models for them to behave as models trained on a subset of the original training data. In this work, we describe the motivations and desirable properties of digital forgetting when applied to LLMs, and we survey recent works on machine unlearning. Specifically, we propose a taxonomy of unlearning methods based on the reach and depth of the modifications done on the models, we discuss and compare the effectiveness of machine unlearning methods for LLMs proposed so far, and we survey their evaluation. Finally, we describe open problems of machine unlearning applied to LLMs and we put forward recommendations for developers and practitioners.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 3","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11078-6.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142963030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comprehensive survey on impact of applying various technologies on the internet of medical things","authors":"Shorouk E. El-deep,&nbsp;Amr A. Abohany,&nbsp;Karam M. Sallam,&nbsp;Amr A. Abd El-Mageed","doi":"10.1007/s10462-024-11063-z","DOIUrl":"10.1007/s10462-024-11063-z","url":null,"abstract":"<div><p>This paper explores the transformative impact of the Internet of Medical Things (IoMT) on healthcare. By integrating medical equipment and sensors with the internet, IoMT enables real-time monitoring of patient health, remote patient care, and individualized treatment plans. IoMT significantly improves several healthcare domains, including managing chronic diseases, patient safety, and drug adherence, resulting in better patient outcomes and reduced expenses. Technologies like blockchain, Artificial Intelligence (AI), and cloud computing further boost IoMT’s capabilities in healthcare. Blockchain enhances data security and interoperability, AI analyzes massive volumes of health data to find patterns and make predictions, and cloud computing offers scalable and cost-effective data processing and storage. Therefore, this paper provides a comprehensive review of the Internet of Things (IoT) and IoMT-based edge-intelligent smart healthcare, focusing on publications published between 2018 and 2024. The review addresses numerous studies on IoT, IoMT, AI, edge and cloud computing, security, Deep Learning, and blockchain. The obstacles facing IoMT are also covered in this paper, including interoperability issues, regulatory compliance, and privacy and data security concerns. Finally, recommendations for further studies are provided.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 3","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11063-z.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142938854","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 evaluation of ChatGPT and human generated image/textual contents by bipolar generalized fuzzy hypergraph","authors":"Abbas Amini,&nbsp;Narjes Firouzkouhi,&nbsp;Wael Farag,&nbsp;Omar Ali,&nbsp;Isam Zabalawi,&nbsp;Bijan Davvaz","doi":"10.1007/s10462-024-11015-7","DOIUrl":"10.1007/s10462-024-11015-7","url":null,"abstract":"&lt;div&gt;&lt;p&gt;Artificial Intelligence (AI) tools, i.e., ChatGPT (Chat Generative Pre-Trained Transformer), are positively and negatively revolutionizing the culture of industries, science, and education. The main objectives of this study are to address uncertainty and vagueness in ChatGPT systems, apply bipolarity as two-sided states of data, model generalized graph-based network with derivations, develop bipolar multi-dimensional fuzzy relation, advance entropy metrics for quantifying ambiguity, cluster entities based on level cuts, present pattern recognition in terms of statistical correlation coefficient, analyze speech recognition framework, and schedule online surgeries on the basis of blockchain technology. The outlined innovation pinpoints on the self-evaluation of ChatGPT systems, merging the bipolarity and generalized fuzzy hypergraph approach, developing the interpretation of graph-based patterns, and benchmarking the AI analysis and metrics advancement. To assess the efficiency of AI bipolar generalized fuzzy hypergraph (BGFH) model, the key conceptual benchmarks are clustering technique for detecting patterns and similar groups of data, statistical methods for the analysis of pattern recognition, and entropy metrics for quantifying the fuzziness within a system. This layout furnishes important characteristics such as union, intersection, complement, homomorphism, isomorphism, verifying the overlapping (intersection) and complement of two strong BGFHs as a strong BGFH. In addition, certain specifications of reflexive, symmetric, transitive, overlapping and integration, are defined using bipolar multi-dimensional fuzzy relation. Eleven classes are derived based on different values within &lt;span&gt;(tin [0,1])&lt;/span&gt; and &lt;span&gt;(sin [-1,0],)&lt;/span&gt; classifying analogous data that aids the similarity detection of generated outputs. Through this approach, a new pattern recognition is used as a data evaluation technique to intelligently facilitate the process in terms of correlation coefficient. It is revealed that the highest magnitude of 0.145 is adopted for patterns &lt;span&gt;(C_{1})&lt;/span&gt; and &lt;i&gt;D&lt;/i&gt;, indicating the most positive correlation between patterns, while patterns &lt;span&gt;(C_{4})&lt;/span&gt; and &lt;i&gt;D&lt;/i&gt; with the value of &lt;span&gt;(-0.35)&lt;/span&gt; are negatively correlated. The results verify that the entropy measure of visual data (0.75) is higher than the entropy measure of textual data with the value of 0.68, indicating more vagueness and ambiguity in visual generated systems. The corresponding textual data &lt;span&gt;(E^{P}(U))&lt;/span&gt; and &lt;span&gt;(E^{N}(U))&lt;/span&gt; are, respectively, calculated as 0.62 and 0.45 for human-created contents and ChatGPT-generated contents, whilst for visual data, the entropy measures &lt;span&gt;(E^{P}(U))&lt;/span&gt; and &lt;span&gt;(E^{N}(U))&lt;/span&gt; are, respectively, 0.25 and 0.66, showing the higher values for the entropy measure of ChatGPT-generated visual data compared to the ChatGPT-generated textual data. In relation to the speech re","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"58 3","pages":""},"PeriodicalIF":10.7,"publicationDate":"2025-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-024-11015-7.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142938855","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}