Artificial Intelligence Review最新文献

{"title":"An adaptive spatio-temporal neural network for PM2.5 concentration forecasting","authors":"Xiaoxia Zhang,&nbsp;Qixiong Li,&nbsp;Dong Liang","doi":"10.1007/s10462-023-10503-6","DOIUrl":"10.1007/s10462-023-10503-6","url":null,"abstract":"<div><p>Accurate PM_2.5 concentration prediction is essential for environmental control management, therefore numerous air quality monitoring stations have been established, which generate multiple time series with spatio-temporal correlation. However, the statistical distribution of data from different monitoring stations varies widely, which needs to provide higher flexibility in the feature extraction stage. Moreover, the spatio-temporal correlation and mutation characteristics of the time series are difficult to capture. To this end, an adaptive spatio-temporal prediction network (ASTP-NET) is proposed, in which the encoder adaptively extracts the input data features, then captures the spatio-temporal dependencies and dynamic changes of the time series, the decoder part maps the encoded features into a predicted future time series representation, while an objective function is proposed to measure the overall fluctuations of the model’s multi-step prediction. In this paper, ASTP-NET is evaluated based on the Xi'an air quality dataset, and the results show that the model outperforms other baseline methods.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"56 12","pages":"14483 - 14510"},"PeriodicalIF":12.0,"publicationDate":"2023-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"50057828","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An adaptive spatio-temporal neural network for PM2.5 concentration forecasting","authors":"Xiaoxia Zhang, Qixiong Li, Dong Liang","doi":"10.1007/s10462-023-10503-6","DOIUrl":"https://doi.org/10.1007/s10462-023-10503-6","url":null,"abstract":"","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"56 1","pages":"14483 - 14510"},"PeriodicalIF":12.0,"publicationDate":"2023-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"52215411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A brief survey on recent advances in coreference resolution","authors":"Ruicheng Liu,&nbsp;Rui Mao,&nbsp;Anh Tuan Luu,&nbsp;Erik Cambria","doi":"10.1007/s10462-023-10506-3","DOIUrl":"10.1007/s10462-023-10506-3","url":null,"abstract":"<div><p>The task of resolving repeated objects in natural languages is known as coreference resolution, and it is an important part of modern natural language processing. It is classified into two categories depending on the resolved objects, namely entity coreference resolution and event coreference resolution. Predicting coreference connections and identifying mentions/triggers are the major challenges in coreference resolution, because these implicit relationships are particularly difficult in natural language understanding in downstream tasks. Coreference resolution techniques have experienced considerable advances in recent years, encouraging us to review this task in the following aspects: current employed evaluation metrics, datasets, and methods. We investigate 10 widely used metrics, 18 datasets and 4 main technical trends in this survey. We believe that this work is a comprehensive roadmap for understanding the past and the future of coreference resolution.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"56 12","pages":"14439 - 14481"},"PeriodicalIF":12.0,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42942677","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"nmODE: neural memory ordinary differential equation","authors":"Zhang Yi","doi":"10.1007/s10462-023-10496-2","DOIUrl":"10.1007/s10462-023-10496-2","url":null,"abstract":"<div><p>Brain neural networks are regarded as dynamical systems in neural science, in which memories are interpreted as attractors of the systems. Mathematically, ordinary differential equations (ODEs) can be utilized to describe dynamical systems. Any ODE that is employed to describe the dynamics of a neural network can be called a neuralODE. Inspired by rethinking the nonlinear representation ability of existing artificial neural networks together with the functions of columns in the neocortex, this paper proposes a theory of memory-based neuralODE, which is composed of two novel artificial neural network models: nmODE and <span>(epsilon)</span>-net, and two learning algorithms: nmLA and <span>(epsilon)</span>-LA. The nmODE (neural memory Ordinary Differential Equation) is designed with a special structure that separates learning neurons from memory neurons, making its dynamics clear. Given any external input, the nmODE possesses the global attractor property and is thus embedded with a memory mechanism. The nmODE establishes a nonlinear mapping from the external input to its associated attractor and does not have the problem of learning features homeomorphic to the input data space, as occurs frequently in most existing neuralODEs. The nmLA (neural memory Learning Algorithm) is developed by proposing an interesting three-dimensional inverse ODE (invODE) and has advantages in memory and parameter efficiency. The proposed <span>(epsilon)</span>-net is a discrete version of the nmODE, which is particularly feasible for digital computing. The proposed <span>(epsilon)</span>-LA (<span>(epsilon)</span> learning algorithm) requires no prior knowledge of the number of network layers. Both nmLA and <span>(epsilon)</span>-LA have no problem with gradient vanishing. Experimental results show that the proposed theory is comparable to state-of-the-art methods.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"56 12","pages":"14403 - 14438"},"PeriodicalIF":12.0,"publicationDate":"2023-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-023-10496-2.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43516259","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 value iteration for discrete-time intelligent critic control: A survey","authors":"Mingming Zhao,&nbsp;Ding Wang,&nbsp;Junfei Qiao,&nbsp;Mingming Ha,&nbsp;Jin Ren","doi":"10.1007/s10462-023-10497-1","DOIUrl":"10.1007/s10462-023-10497-1","url":null,"abstract":"<div><p>Optimal control problems are ubiquitous in practical engineering applications and social life with the idea of cost or resource conservation. Based on the critic learning scheme, adaptive dynamic programming (ADP) is regarded as a significant avenue to address the optimal control problems by combining the advanced design ideas such as adaptive control, reinforcement learning, and intelligent control. This survey introduces the recent development of ADP and related intelligent critic control with an emphasis on advanced value iteration (VI) schemes for discrete-time nonlinear systems. The theoretical results focus on convergence and stability properties for general VI, stabilizing VI, integrated VI, evolving VI, adjustable VI schemes and so on. Several significant applications are also elaborated in aspects of optimal regulation, optimal tracking, and zero-sum games. We aim to break through the bottleneck problems for VI algorithms in realizing evolving control, accelerating learning speed, and reducing the calculation expense. In addition, the prospects of new theoretical and technical fields for advanced VI schemes are looked ahead.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"56 10","pages":"12315 - 12346"},"PeriodicalIF":12.0,"publicationDate":"2023-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46906955","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automatic path planning of unmanned combat aerial vehicle based on double-layer coding method with enhanced grey wolf optimizer","authors":"Yingjuan Jia,&nbsp;Liangdong Qu,&nbsp;Xiaoqin Li","doi":"10.1007/s10462-023-10481-9","DOIUrl":"10.1007/s10462-023-10481-9","url":null,"abstract":"<div><p>The unmanned combat aerial vehicle (UCAV) technology has to deal with a lot of challenges in complex battlefield environments. The UCAV requires a high number of points to build the path to avoid dangers in order to achieve a safe and low-energy flying path, which increases the issue dimension and uses more computer resources while producing unstable results. To address the issue, this paper proposes a double-layer (DLC) model for path planning, which reduces the outputting dimension of path-forming points, reduces the computational cost and enhances the path stability. Meanwhile, this paper improves the grey wolf optimizer (K-FDGWO) by introducing adaptive K-neighbourhood-based learning strategy and differential “hunger-hunting strategy”, and using fitness distance correlation (FDC) to balance the global exploration and local exploitation. Besides, the K-FDGWO and Differential Evolution (DE) algorithm are jointly used for the DLC model (DLC-K-FDGWO). The experimental results indicated that the proposed DLC-K-FDGWO method for path planning always generated the ideal flight path in complicated environments.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"56 10","pages":"12257 - 12314"},"PeriodicalIF":12.0,"publicationDate":"2023-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-023-10481-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41825669","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":"Exploring the knowledge diffusion and research front of OWA operator: a main path analysis","authors":"Dejian Yu,&nbsp;Tianxing Pan,&nbsp;Zeshui Xu,&nbsp;Ronald R. Yager","doi":"10.1007/s10462-023-10462-y","DOIUrl":"10.1007/s10462-023-10462-y","url":null,"abstract":"<div><p>In recent years, more and more attention is paid to the OWA operator in the academy. Growth curve analysis, which is often used in ecosystem studies, also indicates that this growth trend will continue. However, prior literature has not made a big picture to help researchers make clear of the development of this field by identifying the evolution path. The classic main path analysis is an excellent method combining quantitative analysis and qualitative analysis. We conducted the classic main path analysis and its variants on a citation network with 1474 papers to probe the development trajectories and research topics of OWA. We obtained several findings by constructing local and global main path, and multiple main paths. The path results indicate that weight generation and operator generalization run through the overall OWA domain, show that the multiple criteria decision making process assumed in the related research begins to be dynamic and multi-period, and reveal that some theories such as social network theory are introduced into the OWA operator and the applications are also greatly expanded.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"56 10","pages":"12233 - 12255"},"PeriodicalIF":12.0,"publicationDate":"2023-05-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-023-10462-y.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48678657","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":"Gift: granularity over specific-class for feature selection","authors":"Jing Ba,&nbsp;Keyu Liu,&nbsp;Xibei Yang,&nbsp;Yuhua Qian","doi":"10.1007/s10462-023-10499-z","DOIUrl":"10.1007/s10462-023-10499-z","url":null,"abstract":"<div><p>As a fundamental material of Granular Computing, information granulation sheds new light on the topic of feature selection. Although information granulation has been effectively applied to feature selection, existing feature selection methods lack the characterization of feature potential. Such an ability is one of the important factors in evaluating the importance of features, which determines whether candidate features have sufficient ability to distinguish different target variables. In view of this, a novel concept of granularity over specific-class from the perspective of information granulation is proposed. Essentially, such a granularity is a fusion of intra-class and extra-class based granularities, which enables to exploit the discrimination ability of features. Accordingly, an intuitive yet effective framework named G<span>ift</span>, i.e., granularity over specific-class for feature selection, is proposed. Comprehensive experiments on 29 public datasets clearly validate the effectiveness of G<span>ift</span> as compared with other feature selection strategies, especially in noisy data.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"56 10","pages":"12201 - 12232"},"PeriodicalIF":12.0,"publicationDate":"2023-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43290432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning implementations in mining applications: a compact critical review","authors":"Faris Azhari,&nbsp;Charlotte C. Sennersten,&nbsp;Craig A. Lindley,&nbsp;Ewan Sellers","doi":"10.1007/s10462-023-10500-9","DOIUrl":"10.1007/s10462-023-10500-9","url":null,"abstract":"<div><p>Deep learning is a sub-field of artificial intelligence that combines feature engineering and classification in one method. It is a data-driven technique that optimises a predictive model via learning from a large dataset. Digitisation in industry has included acquisition and storage of a variety of large datasets for interpretation and decision making. This has led to the adoption of deep learning in different industries, such as transportation, manufacturing, medicine and agriculture. However, in the mining industry, the adoption and development of new technologies, including deep learning methods, has not progressed at the same rate as in other industries. Nevertheless, in the past 5 years, applications of deep learning have been increasing in the mining research space. Deep learning has been implemented to solve a variety of problems related to mine exploration, ore and metal extraction and reclamation processes. The increased automation adoption in mining provides an avenue for wider application of deep learning as an element within a mine automation framework. This work provides a compact, comprehensive review of deep learning implementations in mining-related applications. The trends of these implementations in terms of years, venues, deep learning network types, tasks and general implementation, categorised by the value chain operations of exploration, extraction and reclamation are outlined. The review enables shortcomings regarding progress within the research context to be highlighted such as the proprietary nature of data, small datasets (tens to thousands of data points) limited to single operations with unique geology, mine design and equipment, lack of large scale publicly available mining related datasets and limited sensor types leading to the majority of applications being image-based analysis. Gaps identified for future research and application includes the usage of a wider range of sensor data, improved understanding of the outputs by mining practitioners, adversarial testing of the deep learning models, development of public datasets covering the extensive range of conditions experienced in mines.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"56 12","pages":"14367 - 14402"},"PeriodicalIF":12.0,"publicationDate":"2023-05-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://link.springer.com/content/pdf/10.1007/s10462-023-10500-9.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"47927500","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":"Optimal energy management in EVCS and distribution system considering QoS using hybrid technique","authors":"Uma Dharmalingam,&nbsp;Vijayakumar Arumugam","doi":"10.1007/s10462-023-10458-8","DOIUrl":"10.1007/s10462-023-10458-8","url":null,"abstract":"<div><p>This manuscript proposes a hybrid method to effectively manage the energy on electric vehicle charging station (EVCS) and distribution system. The proposed method is consolidation of shell game optimization (SGO) and recalling-enhanced recurrent neural network (RERNN) named SGO-RERNN technique. The main aim of this work is to offer maximal amount of energy in this system and charging plans for EVCSs. The hybrid SGO-RERNN system is used to obtain the balancing solution. The intention of the distribution system is to maximize the planning charged for EVCSs. The proposed algorithm is related to supply function equilibrium method and it is used to modify and examine the interaction of each electric vehicle charging known as leader and the distributed system is known as follower. The hybrid SGO-RERNN technique is used to acquire the equilibrium solution. The SGO-RERNN system is implemented on MATLAB platform and the performance is compared to existing systems. Furthermore, the EVCS and distribution system efficiency is analyzed with the help of proposed method. The SGO-RERNN method attains electric vehicle charging station 1 attains 600.234, electric vehicle charging station 2 attains 3509.19, electric vehicle charging station 3 attains 4413.09, and distribution system attains 4327.033. The experimental outcomes prove that the integrated energy system costs minimized 3.89% and gains maximized to 7.8%. Finally, the SGO-RERNN method locates the optimum global solutions efficiently and accurately over the existing methods.</p></div>","PeriodicalId":8449,"journal":{"name":"Artificial Intelligence Review","volume":"56 12","pages":"14297 - 14326"},"PeriodicalIF":12.0,"publicationDate":"2023-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42575049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}