Article
Engineering, Electrical & Electronic
Yingbo Sun, Tao Xu, Jingyuan Li, Yuan Chu, Xuewu Ji
Summary: This article presents a macro-micro-hierarchical spatio-temporal attention architecture for predicting the future trajectories of multiple heterogeneous agents in an unsignalized roundabout traffic scenario. By considering the heterogeneity of agents, designing a heterogeneous graph, and utilizing a heterogeneous graph attention network, accurate trajectory prediction for multiple agents is achieved.
IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY
(2023)
Article
Physics, Multidisciplinary
Zhuo-Lin Li, Jie Yu, Xiao-Lin Zhang, Ling-Yu Xu, Bao-Gang Jin
Summary: This paper proposes a multi-hierarchical attention network for multi-scale prediction of multivariate time series in Earth system science, which effectively captures correlations between different spacetime variables and demonstrates its effectiveness and robustness on real-world datasets.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2022)
Article
Mathematics
Haokun Su, Xiangang Peng, Hanyu Liu, Huan Quan, Kaitong Wu, Zhiwen Chen
Summary: This paper proposes a novel spatio-temporal prediction model by combining graph convolutional network (GCN) and temporal convolutional network (TCN) to improve the accuracy of electricity price forecasting. The experimental results show that the model outperforms other models in electricity price forecasting.
Article
Thermodynamics
Yun Wang, Tuo Chen, Shengchao Zhou, Fan Zhang, Ruming Zou, Qinghua Hu
Summary: In this study, a novel forecasting model called ED-Wavenet-TF is proposed, which uses two Wavenet networks as Encoder and Decoder connected by the multi-head self-attention mechanism. The model achieves improved forecast accuracy by correcting errors at intermediate forecasting steps and using teacher forcing as the multi-step-ahead output strategy. Experimental results demonstrate that ED-Wavenet-TF outperforms comparable models in wind speed and wind power forecasting, with a lower symmetric mean absolute percentage error. The effectiveness of the model is also demonstrated in making multi-step-ahead forecasts with multivariate inputs.
ENERGY CONVERSION AND MANAGEMENT
(2023)
Article
Environmental Sciences
Kangling Lin, Hua Chen, Yanlai Zhou, Sheng Sheng, Yuxuan Luo, Shenglian Guo, Chong -Yu Xu
Summary: This study proposes a similarity search-based data-driven framework and applies the advanced temporal convolutional network based encoder-decoder model (S-TCNED) for multi-step ahead flood forecasting. The results show that the S-TCNED model can effectively mimic the long-term rainfall-runoff relationship and provide more reliable and accurate forecasts of large floods compared to the TCNED model, even in extreme weather conditions.
SCIENCE OF THE TOTAL ENVIRONMENT
(2023)
Article
Geography, Physical
Zhu Han, Ce Zhang, Lianru Gao, Zhiqiang Zeng, Bing Zhang, Peter M. Atkinson
Summary: Accurate crop mapping is crucial for crop yield forecasting, agricultural productivity development, and agricultural management. This paper proposes a novel method called STMA for crop mapping using time-series SAR imagery. Experimental results demonstrate that STMA achieves state-of-the-art performance and excels in spatio-temporal modeling on different crops and scenarios.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Computer Science, Information Systems
Nana Huang, Ruimin Hu, Xiaochen Wang, Hongwei Ding
Summary: This study explores how to model users' historical behavior sequences at multiple scales to predict their click-through rates on micro-videos and determine whether to recommend them to users. They propose a novel Multi-scale Modeling Temporal Hierarchical Attention (MMTHA) method that captures users' short-term dynamic interests, describes their coarse-grained interests, and captures their fine-grained interests. They employ a forward multi-headed self-attention mechanism to identify and integrate long-term correlations between previously segmented temporal windows. Experimental results show that the proposed MMTHA model achieves state-of-the-art performance in all tests.
INFORMATION SCIENCES
(2023)
Article
Mathematics
Snezhana Gocheva-Ilieva, Atanas Ivanov, Hristina Kulina, Maya Stoimenova-Minova
Summary: A novel multi-step ahead strategy is developed for forecasting air pollutants, using independent ex-ante data gathered from official weather forecast sites. Separate single models are built using new forecasted values each day, and the sought forecasts are obtained by averaging the predictions of these models. Random forest (RF) and arcing (Arc-x4) machine learning algorithms are used for modeling. The proposed averaging strategy shows promising results compared to single models, with lower root mean squared errors (RMSE) in most cases.
Article
Computer Science, Artificial Intelligence
Pu-Yun Kow, Meng-Hsin Lee, Wei Sun, Ming-Hwi Yao, Fi-John Chang
Summary: Precision agriculture control systems rely on reliable and accurate microclimate forecasts to maintain environmental suitability for crop growth. This study proposes a hybrid deep learning model that can produce accurate multi-horizon and multi-factor forecasts without using IoT data. Experimental results show that the proposed model performs similarly to the benchmark model and has excellent noise removal and feature extraction capabilities. This deep learning approach is of great significance in microclimate forecasting.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Thermodynamics
Zhijian Qu, Jian Li, Xinxing Hou, Jianglin Gui
Summary: In this study, a spatio-temporal graph deep neural network is used to learn the spatial characteristics and explore the strong correlation characteristics of wind-farm clusters. By analyzing wind power sequence characteristics, two Stacking models are used for wind power prediction, and a dual stacking model fusion strategy is formed. The proposed model achieves accurate wind power prediction results, outperforming 14 comparative algorithms proposed by other researchers.
Article
Energy & Fuels
Muhammad Aslam, Jun-Sung Kim, Jaesung Jung
Summary: This study proposes a deep learning model based on a dual-attention mechanism for multi-step ahead wind power forecasting. The model achieves the highest forecasting skill score and outperforms other traditional methods in terms of MAE and RMSE. The effectiveness of the proposed model is demonstrated through comparison with other state-of-the-art models.
Article
Computer Science, Information Systems
Rongtao Zhang, Xueling Ma, Weiping Ding, Jianming Zhan
Summary: Currently, prediction is a significant area of research, with the challenge of improving accuracy and generalization capabilities of models. To address the issue of error accumulation in existing prediction models, we propose a multi-step time series prediction model that incorporates prediction correction. Our model effectively rectifies initial predictions and safeguards against deviations, demonstrating its effectiveness through comparative experimental analysis.
INFORMATION SCIENCES
(2023)
Article
Engineering, Civil
Shen Fang, Veronique Prinet, Jianlong Chang, Michael Werman, Chunxia Zhang, Shiming Xiang, Chunhong Pan
Summary: The article introduces a new model for predicting urban traffic flow, which can consider the complex spatio-temporal dependencies in traffic networks and utilizes multi-source data for prediction. In various experiments, the model performs well on different types of traffic networks, particularly showing significant results in handling large-scale traffic networks.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Tinghuai Ma, Kexing Peng, Huan Rong, Yurong Qian, Najla Al-Nabhan
Summary: This article studies hierarchical deep multi-agent reinforcement learning and proposes a novel model called Hierarchical Spatio-Temporal Communication Network (HSTCN). HSTCN addresses the issues of reward sparsity and lack of decision information caused by long-trajectory training and partial observability through the design of high-level and low-level policies. Experimental results demonstrate the superior performance and rationality of HSTCN.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Yanjun Qin, Haiyong Luo, Fang Zhao, Yuchen Fang, Xiaoming Tao, Chenxing Wang
Summary: We propose a novel spatio-temporal hierarchical MLP network (STHMLP) for traffic forecasting, which can capture the trend-cyclical and seasonal features of traffic time series. By using a decomposition architecture and designing fine and coarse modules, the STHMLP can extract spatio-temporal information from roads and regions and effectively capture both fine-grained and coarse-grained spatial dependencies. Experimental results on real-world traffic datasets demonstrate that the STHMLP outperforms state-of-the-art baselines.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Software Engineering
Juanjuan Zhao, Jiexia Ye, Minxian Xu, Chengzhong Xu
Summary: Real-time individuals' destination prediction is important for applications such as user tracking and service recommendation. This paper proposes a practical model based on public transportation metro systems to predict passengers' destinations, combining individual and crowd behavior, as well as discrete choice model and neural network model. Experimental results show better prediction accuracy compared to baselines.
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE
(2023)
Article
Computer Science, Information Systems
Akshi Kumar
Summary: This research proposes a novel crowd documentation model based on social media, which involves identifying duplicate questions, semantic matching, and expert ranking to provide enriched and curated documentation.
MULTIMEDIA SYSTEMS
(2023)
Article
Operations Research & Management Science
Kerim U. Kizil, Baris Yildiz
Summary: This paper proposes a new last-mile delivery model that combines several new approaches and technologies to address the limitations of traditional delivery systems. By utilizing public transit, automated service points, crowd-shipping, and zero-emission vehicles, the model provides a low-cost and environmentally friendly express delivery service. The proposed system is formulated as a two-stage stochastic program and solved using a branch-and-price algorithm. Computational studies and simulations with real-world data show the effectiveness of the suggested methodology in providing valuable managerial insights for the proposed system.
TRANSPORTATION SCIENCE
(2023)
Article
Political Science
Tobias Widmann, Maximilian Wich
Summary: Previous research on emotional language relied on sentiment dictionaries that are not suitable for political text. This paper proposes a novel emotional dictionary for political text and compares its performance with word embedding models and transformer-based models. The results show the strengths of the novel transformer-based models and demonstrate that all customized approaches outperform off-the-shelf dictionaries in measuring emotional language in German political discourse.
POLITICAL ANALYSIS
(2023)
Article
Nutrition & Dietetics
Glenys Jones, Elaine Macaninch, Duane D. Mellor, Ayela Spiro, Kathy Martyn, Thomas Butler, Alice Johnson, J. Bernadette Moore
Summary: COVID-19 has worsened health inequalities in the UK, highlighting the need for improved nutrition education in undergraduate medical training.
BRITISH JOURNAL OF NUTRITION
(2023)
Article
Computer Science, Artificial Intelligence
Runhe Zhu, Burcin Becerik-Gerber, Jing Lin, Nan Li
Summary: This study predicts the impact of social and environmental factors on individual wayfinding behavior using machine learning and discrete choice models, and applies them to evaluate crowd evacuation performance under different building design scenarios in evacuation simulations. The results show that different building attributes collectively influence occupant behavior, and the discrete choice model has better interpretability. The study also finds significant differences in agent responses to different building designs and identifies critical factors for the applicability of evacuation models.
ADVANCED ENGINEERING INFORMATICS
(2023)
Article
Robotics
Xiaojun Lu, Hanwool Woo, Angela Faragasso, Atsushi Yamashita, Hajime Asama
Summary: Mobile robots operating in public environments need to navigate among humans and obstacles in a socially compliant and safe manner. This study proposes a novel network structure that encodes the effects of obstacles on the robot and combines it with a human-robot interaction network to consider the combined effects of obstacles and crowds. Experimental results show that this method outperforms existing methods in terms of navigation performance.
Article
Automation & Control Systems
Huiming Li, Hao Chen, Xiangke Wang
Summary: This paper investigates the problem of collision-free leader-follower formation generation and tracking of multiple fixed-wing unmanned aerial vehicles (UAVs). A novel control law based on physicomimetics approach is proposed to integrate formation generation, formation tracking, and collision avoidance. The artificial forces driven by control laws imitating physical forces are used to achieve desired collaborative behaviors and obstacle avoidance naturally. Speed constraints are also considered and modified using a saturation function. Numerical simulations and experiments are provided to verify the effectiveness of the proposed control scheme.
ASIAN JOURNAL OF CONTROL
(2023)
Article
Engineering, Environmental
Long Liang, Chunmin Zhang, Shaolei Zhao, Baozhong Liu, Limin Wang, Fei Liang
Summary: Designing efficient catalysts is a crucial challenge for practical applications of high-capacity hydride in fuel-cell-based hydrogen economy. In this study, platinum-functionalized Ti3C2 material with an accordion-like structure, interlayer, and surface-dispersed nanoparticles was synthesized. The catalyst, Ti3C2@Pt, reduced the initial dehydrogenation temperature of high-density hydride AlH3 by 50% to 62 °C, comparable to commercial AlH3. Moreover, it exhibited high hydrogen supplying performance and retention ratio, achieving 9.3 wt% and 98% respectively, surpassing previously reported catalysts. The exceptional dehydrogenation performance of the material makes it a practical candidate for mobile device applications with the aid of high-efficiency catalysts.
CHEMICAL ENGINEERING JOURNAL
(2023)
Article
Engineering, Environmental
Qiang Liu, Youjin Gong, Boyu Liu, Shunshun Xiong, Hui-Min Wen, Xiaolin Wang
Summary: In this study, an ultra-microporous alkyl Cu-based MOF-11 was developed for highly selective adsorption of xenon. The unique pore system of MOF-11 not only enables strong binding affinity with xenon, but also allows for dense packing of xenon molecules. MOF-11 demonstrates record-high xenon storage density, uptake, and selectivity, making it a promising candidate for xenon separation and capture.
CHEMICAL ENGINEERING JOURNAL
(2023)
Article
Engineering, Electrical & Electronic
Liwen Liu, Xiaobo Wang, Yongqi Hu, Chen Wang, Yan Xu
Summary: Automatic modulation recognition (AMR) technology is widely used in commercial and military communication systems. However, with the increasing complexity and crowdedness of the spectrum environment, the AMR problem of compound wireless signals has attracted attention. In this study, we proposed a multi-label deep forest (MLDF) framework trained on raw compound complex time-domain signals and demonstrated its superiority over other models under low-SNR training conditions. We also showed that MLDF trained on small-scale datasets can achieve high performance with reduced training samples.
CIRCUITS SYSTEMS AND SIGNAL PROCESSING
(2023)
Article
Computer Science, Artificial Intelligence
Ming Wu, Qianmu Li, Fei Yang, Jing Zhang, Victor S. Sheng, Jun Hou
Summary: With the rapid development of crowdsourcing learning, it has become fast and cheap to obtain a large number of labels from crowd workers. However, the varied qualities of amateurish crowd workers pose challenges to the quality of crowd labels. To improve label quality, researchers have focused on inferring the ground truth from noisy labels and considering factors like worker reliability and instance difficulty. However, there is still insufficient research on label aggregation for biased crowdsourced labeling scenarios.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Book Review
Regional & Urban Planning
Mehdi Alidadi
AUSTRALIAN PLANNER
(2023)
Article
Operations Research & Management Science
Kerim U. Kizil, Baris Yildiz
Summary: With the development of urbanization and e-commerce, traditional last-mile delivery systems are unable to meet the demand. This paper proposes a new last-mile delivery model that combines various new approaches and technologies to provide low-cost and environmentally friendly express delivery service.
TRANSPORTATION SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Hongxin Wei, Renchunzi Xie, Lei Feng, Bo Han, Bo An
Summary: Crowdsourcing is a popular approach for large-scale data annotations, but existing methods suffer from poor learning consistency and computational inefficiency. In this article, a novel method called UnionNet is proposed to address these issues by leveraging the union of crowdsourced labels for training deep neural networks. The theoretical analysis and experimental results demonstrate the effectiveness and efficiency of the proposed method.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Information Systems
Xia Liang, Jie Guo, Peide Liu
Summary: This paper investigates a novel consensus model based on social networks to manage manipulative and overconfident behaviors in large-scale group decision-making. By proposing a novel clustering model and improved methods, the consensus reaching is effectively facilitated. The feedback mechanism and management approach are employed to handle decision makers' behaviors. Simulation experiments and comparative analysis demonstrate the effectiveness of the model.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Xiang Li, Haiwang Guo, Xinyang Deng, Wen Jiang
Summary: This paper proposes a method based on class gradient networks for generating high-quality adversarial samples. By introducing a high-level class gradient matrix and combining classification loss and perturbation loss, the method demonstrates superiority in the transferability of adversarial samples on targeted attacks.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Lingyun Lu, Bang Wang, Zizhuo Zhang, Shenghao Liu
Summary: Many recommendation algorithms only rely on implicit feedbacks due to privacy concerns. However, the encoding of interaction types is often ignored. This paper proposes a relation-aware neural model that classifies implicit feedbacks by encoding edges, thereby enhancing recommendation performance.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Jaehong Yu, Hyungrok Do
Summary: This study discusses unsupervised anomaly detection using one-class classification, which determines whether a new instance belongs to the target class by constructing a decision boundary. The proposed method uses a proximity-based density description and a regularized reconstruction algorithm to overcome the limitations of existing one-class classification methods. Experimental results demonstrate the superior performance of the proposed algorithm.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Hui Tu, Shifei Ding, Xiao Xu, Haiwei Hou, Chao Li, Ling Ding
Summary: Border-Peeling algorithm is a density-based clustering algorithm, but its complexity and issues on unbalanced datasets restrict its application. This paper proposes a non-iterative border-peeling clustering algorithm, which improves the clustering performance by distinguishing and associating core points and border points.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Long Tang, Pan Zhao, Zhigeng Pan, Xingxing Duan, Panos M. Pardalos
Summary: In this work, a two-stage denoising framework (TSDF) is proposed for zero-shot learning (ZSL) to address the issue of noisy labels. The framework includes a tailored loss function to remove suspected noisy-label instances and a ramp-style loss function to reduce the negative impact of remaining noisy labels. In addition, a dynamic screening strategy (DSS) is developed to efficiently handle the nonconvexity of the ramp-style loss.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Raghunathan Krishankumar, Sundararajan Dhruva, Kattur S. Ravichandran, Samarjit Kar
Summary: Health 4.0 is gaining global attention for better healthcare through digital technologies. This study proposes a new decision-making framework for selecting viable blockchain service providers in the Internet of Medical Things (IoMT). The framework addresses the limitations in previous studies and demonstrates its applicability in the Indian healthcare sector. The results show the top ranking BSPs, the importance of various criteria, and the effectiveness of the developed model.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Tao Tan, Hong Xie, Liang Feng
Summary: This paper proposes a heterogeneous update idea and designs HetUp Q-learning algorithm to enlarge the normalized gap by overestimating the Q-value corresponding to the optimal action and underestimating the Q-value corresponding to the other actions. To address the limitation, a softmax strategy is applied to estimate the optimal action, resulting in HetUpSoft Q-learning and HetUpSoft DQN. Extensive experimental results show significant improvements over SOTA baselines.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Chao Yang, Xianzhi Wang, Lina Yao, Guodong Long, Guandong Xu
Summary: This paper proposes a dynamic transformer-based architecture called Dyformer for multivariate time series classification. Dyformer captures multi-scale features through hierarchical pooling and adaptive learning strategies, and improves model performance by introducing feature-map-wise attention mechanisms and a joint loss function.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Xiguang Li, Baolu Feng, Yunhe Sun, Ammar Hawbani, Saeed Hammod Alsamhi, Liang Zhao
Summary: This paper proposes an enhanced scatter search strategy, using opposition-based learning, to solve the problem of automated test case generation based on path coverage (ATCG-PC). The proposed ESSENT algorithm selects the path with the lowest path entropy among the uncovered paths as the target path and generates new test cases to cover the target path by modifying the dimensions of existing test cases. Experimental results show that the ESSENT algorithm outperforms other state-of-the-art algorithms, achieving maximum path coverage with fewer test cases.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Shirin Dabbaghi Varnosfaderani, Piotr Kasprzak, Aytaj Badirova, Ralph Krimmel, Christof Pohl, Ramin Yahyapour
Summary: Linking digital accounts belonging to the same user is crucial for security, user satisfaction, and next-generation service development. However, research on account linkage is mainly focused on social networks, and there is a lack of studies in other domains. To address this, we propose SmartSSO, a framework that automates the account linkage process by analyzing user routines and behavior during login processes. Our experiments on a large dataset show that SmartSSO achieves over 98% accuracy in hit-precision.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Renchao Wu, Jianjun He, Xin Li, Zuguo Chen
Summary: This paper proposes a memetic algorithm with fuzzy-based population control (MA-FPC) to solve the joint order batching and picker routing problem (JOBPRP). The algorithm incorporates batch exchange crossover and a two-level local improvement procedure. Experimental results show that MA-FPC outperforms existing algorithms in terms of solution quality.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Guoxiang Zhong, Fagui Liu, Jun Jiang, Bin Wang, C. L. Philip Chen
Summary: In this study, we propose the AMFormer framework to address the problem of mixed normal and anomaly samples in deep unsupervised time-series anomaly detection. By refining the one-class representation and introducing the masked operation mechanism and cost sensitive learning theory, our approach significantly improves anomaly detection performance.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Jin Zhou, Kang Zhou, Gexiang Zhang, Ferrante Neri, Wangyang Shen, Weiping Jin
Summary: In this paper, the authors focus on the issue of multi-objective optimisation problems with redundant variables and indefinite objective functions (MOPRVIF) in practical problem-solving. They propose a dual data-driven method for solving this problem, which consists of eliminating redundant variables, constructing objective functions, selecting evolution operators, and using a multi-objective evolutionary algorithm. The experiments conducted on two different problem domains demonstrate the effectiveness, practicality, and scalability of the proposed method.
INFORMATION SCIENCES
(2024)
Article
Computer Science, Information Systems
Georgios Charizanos, Haydar Demirhan, Duygu Icen
Summary: This article proposes a new fuzzy logistic regression framework that addresses the problems of separation and imbalance while maintaining the interpretability of classical logistic regression. By fuzzifying binary variables and classifying subjects based on a fuzzy threshold, the framework demonstrates superior performance on imbalanced datasets.
INFORMATION SCIENCES
(2024)
