Search

Global Grand Challenges (GCs) have received only a limited attention within the tourism literature and this Progress paper examines one such GC – dementia and the contribution that tourism research can make to helping people live well with dementia. The paper critically debates the concept of GCs, the challenges of addressing them and how dementia presents both a challenge and opportunity for the academy. Using the concept of the visitor journey and the social model of disability as a paradigm, the paper reviews progress towards creating a more dementia-inclusive society through the auspices of tourism research, reviewing the contributions to date and where the interconnections with dementia and tourism exist. The review draws from a wide interdisciplinary framework including health, medicine, travel medicine and leisure and argues for a greater use of transdisciplinary approaches to address GCs. [ABSTRACT FROM AUTHOR]

This paper examines the characteristics that drive conflicting outcomes on the impact of Twitter data on firm returns using financial micro data. Using 314 European tourism firms as a case study and a sample of 63 million Tweets, we build sentiment and emotion (anger, fear, joy) data series and use them to compute impulse response functions for firm returns. Our results indicate that firm size and popularity are the most important firm features that explain the asymmetric impact of Twitter sentiment and of the anger emotion, while debt explains the variations in the impact of the fear emotion. We also find that the impact of the joy emotion is more evident before the COVID-19 pandemic and more muted after the outbreak. Our findings reconcile varied research on Twitter's impact on tourism industry returns and provide insights to practitioners on using Twitter to gauge online users' collective knowledge of real outcomes. [ABSTRACT FROM AUTHOR]

Keyphrase Extraction is an automatic task that involves identifying the key words or phrases that capture the main content of an article. It is useful for various downstream tasks, including text search, text clustering, and text classification. Embedding-based methods for keyphrase extraction have shown promising results by utilizing pre-trained language models to represent candidate phrases and documents separately. These methods then rank the candidate phrases based on the cosine similarity between the document and the candidate phrases embeddings. However, there are mainly two shortcomings in such methods: I) Redundancy errors, when there are partial repetitions of candidate keyphrases, the methods tend to use redundant long phrases as keyphrases; II) Low keyphrase coverage, such as some keyphrases used to describe locally important information are ignored. In this paper, we propose an unsupervised keyphrase extraction method called "MICRank", which evaluates the importance of candidate keyphrases from three perspectives: global information, local information, and attribute information, and solved the aforementioned issues. The experimental results on six benchmarks demonstrate that the proposed MICRank method outperforms the state-of-the-art unsupervised keyphrase extraction methods. In addition, this paper improves the judgment criterion of correct keyphrase extraction and introduces a new evaluation metric called S1@M (M ∈ {5,10,15}) to address the issue of synonyms being considered incorrect predictions. [ABSTRACT FROM AUTHOR]

In allusion to the privacy and security problems, a secure multi-channel information protection scheme based on an integrated optical device is proposed in this paper. The information to be encrypted and shared is converted into a QR code. For supporting the multi-channel encryption, a 1- n w encryption algorithm is employed, which possesses the capability to map the original image information into n w seemingly random cipher images. In terms of hardware, an ultra-compact optical encryption device is designed, which consists of two rows of ring resonators coupled to the waveguide. This novel encryption device can output ten cipher images, each transmitted on a separate channel. Meanwhile, the vertical-cavity surface-emitting laser subject to optical injection and optical feedback is researched to generate optical chaos sequences, which are introduced for key generation and QR code encryption. Ultimately, combining the 1- n w encryption algorithm and optical chaos, information protection is implemented based on integrated optics. The simulation results demonstrate the practical feasibility and remarkable effectiveness of the proposed scheme. Furthermore, various security and robustness analyses are performed, and the results reveal that the proposed information protection scheme has a high level of security and robustness. This work is of great significance for the development of optical encryption chips and provides potential applications for secure data sharing. [ABSTRACT FROM AUTHOR]

DNA sequence motif discovery is an important issue in gene research, which helps identify transcription factor binding sites in DNA sequences to reveal the mechanisms that regulate gene expression. However, the growing awareness of privacy protection and increasingly stringent regulations pose challenges to data collection and usage. This paper proposes DP-FLMD, a privacy-preserving federated framework for discovering DNA sequence motifs. We employ federated learning, allowing participants to store their raw data locally and upload only selected parameters to protect data privacy. Nevertheless, privacy concerns still persist, as these participant parameters may potentially compromise their privacy to some extent. To enhance privacy protection, we incorporate a differential privacy algorithm to add noise to these parameters. Additionally, we verify that the method satisfies ϵ -differential privacy through theoretical analysis. Extensive experiments are performed on six datasets, demonstrating that DP-FLMD achieves a good trade-off between privacy and utility. Besides, we investigate the effect of parameters on DP-FLMD. In the future, our research will focus on exploring enhanced privacy protection mechanisms, expanding the scope of framework applications, and further optimizing the trade-off between privacy and utility. • Federated learning for DNA sequence motifs discovery. • A strategy for reducing communication costs. • Alleviating privacy concerns in distributed DNA sequence motif discovery. • The new algorithm achieves a better trade-off between privacy and utility. [ABSTRACT FROM AUTHOR]

The standard NMF algorithm is not suitable for sampling data from low-dimensional manifolds embedded in high-dimensional environmental spaces, as the geometric information hidden in feature manifolds and sample manifolds is rarely learned. In order to obtain better clustering performance based on NMF, manifold and orthogonal constraint, a new type of model named Orthogonal Graph regularized Non-negative Matrix Factorization model under Sparse Constraints (OGNMFSC) is proposed. Firstly, this type of model constructs a nearest neighbor graph to encode the geometric information of the data space, in order to obtain more discriminative ability by preserving the structure of the graph. Secondly, this type of model adds orthogonal constraints to achieve better local representation and significantly reduce the inconsistency between the original matrix and the basis vectors. Thirdly, by adding sparse constraints to obtain a sparser representation matrix, the clustering performance of the model can be improved. The main conclusion of this paper is that two effective algorithms have been generated to solve the model, which not only provides theoretical convergence proof for these two algorithms, but also demonstrates significant clustering performance in experiments compared to classical models such as K-means, PCA, NMF, Semi-NMF, NMFSC, ONMF, GNMF, NeNMF. • Effectively combining graph regularization, orthogonality and sparsity. • The convergence proof of the new model's algorithm has been obtained. • Compared with NMF, ONMF and etc, this model has shown good results in experiments. • Explain the effective combination of orthogonality, graph, sparsity via experiments. [ABSTRACT FROM AUTHOR]

This research introduces Automated Cost-Duration Variance Prediction (CDVP) system to empower project managers to proactively predict project budget, duration, cost variance and duration variance. The research involved the development and comparison of a total of 48 machine learning (ML) algorithms, with 12 algorithms dedicated to each predictive model. To conduct the study, a dataset comprising 1277 historical cases of Field Canals Improvement Projects (FCIPs) was utilized. Factor analysis and stepwise regression analysis were employed for feature selection where five key drivers were identified from the initial 25 parameters. The paper optimized the performance of each algorithm based on Bayesian optimization. The proposed system produces four outputs: project cost, project cost variance, project duration, and duration variance. The developed ML models were able to predict all of the system's outputs with an adjusted determination coefficient R2 score greater than 0.93 and Mean Absolute Percentage Error (MAPE) less than 8.97. Environmentally, this research recommends to assess CO 2 emissions as a performance criterion to select the low environmental impact algorithms. As a result, Light Gradient Boosting Machine (LightGBM) and AdaBoost Regressor were green sustainable algorithms that emitted on average 58 and 69 CO 2e grams per computational hours. To enhance interpretability, the study employed the SHapley Additive exPlanations (SHAP) technique to provide explanations for the key drivers. [ABSTRACT FROM AUTHOR]

In artificial intelligence applications, advanced computational models, such as deep learning, are employed to achieve high accuracy, often requiring the execution of numerous operations. Conversely, lightweight computational models are typically more resource-efficient, making them suitable for various devices, including smartphones, tablets, and wearable technology. This paper presents an ultra-low-computation solution for interpreting sign languages to assist deaf and hard-of-hearing individuals without needing specialized hardware or significant computational resources. The proposed approach initially performs data abstraction on the input data. During this process, the image is systematically scanned from various perspectives, and the collected information is then encoded into a one-dimensional vector. Subsequently, the abstracted information undergoes processing through a Fully Connected Neural Network (FCN), resulting in highly accurate output. We also introduced two abstraction methods, namely Opaque and Glass, inspired by the interaction of light with different types of objects. The proposed abstractions facilitate the comprehension of the hand gesture's outer boundary as well as its row-wise and column-wise density of pixels. Our experiments on three datasets confirm the efficiency of the proposed method, achieving an accuracy of 99.4% in recognizing American Sign Language, 99.96% accuracy in recognizing Indian Sign Language, and 99.95% accuracy in recognizing Bangla Sign Language. Notably, the model size and the number of MAC operations are significantly smaller than state-of-the-art computational models trained on the same datasets. • We introduced an ultra-low-computation method for interpreting sign languages. • We developed a new encoding for the abstraction of hand gesture images. • Our approach significantly reduces both model size and MAC operations. • We achieved high accuracy in understanding American, Indian, and Bangla SLs. [ABSTRACT FROM AUTHOR]

Common subspace learning methods only utilize local or global structure in feature extraction, and cannot obtain the global optimal discriminative projection matrix. For this reason, this paper proposes a discriminative sparse subspace learning method based on the manifold regularization framework (DSSL-MR), which introduces the graph Laplacian matrix that reflects the intrinsic geometric structure of the sample as a penalty term. DSSL-MR simultaneously uses both sub-manifold and multi-manifold information of samples for obtaining optimal projection to enhance the discriminability of different classes in subspace. DSSL-MR uses the sparse property of the L 2 , 1 -norm to constrain the projection matrix, which can eliminate redundant features and select features that are significant for classification. It is a linear supervised method, which belongs to the Fisher discriminant analysis framework. Experimental results on multiple real-world datasets show that the algorithm is very effective in classification and has high recognition rates. • Subspace learning and spectral graphs are combined into a unified learning framework. • The learned projection maps samples to the subspace with optimal class separation. • The manifold regularization framework fusing decision boundary and subspace learning. • The sample structure is analyzed by fusing multimanifold and submanifold information. [ABSTRACT FROM AUTHOR]

Emergency Medical Services are essential for health systems as their effective management can improve patient prognosis. Nevertheless, designing an optimized distribution of resources is a difficult task due to the complex nature of these systems. Moreover, locating the resources is particularly challenging in heterogeneous density territories where, in addition to their efficient management, the equity principle in the medical access of inhabitants of rural areas is also desirable. This paper approaches the ambulance (re)location–allocation problem in the geographical area of the Basque Country. The area has three major cities, which account for a third of the emergencies, while there are few emergencies in rural areas, with a sparse population. To that end, a two-stage stochastic 0-1 integer linear programming model that balances the response time between densely populated and isolated areas is proposed. Specifically, the model incorporates two relevant principles: (1) optimizing emergency attendance through the option of allocating ambulances via a multi-interval response time and (2) equitably responding to emergencies so remote areas are not neglected. Conducted experiments have been validated and indicate that the proposed model can improve the success rate in rural areas by 23 percentage points, while reducing the overall success rate by less than 9 percentage points. • We develop two data-driven ambulance (re)location–allocation stochastic models. • Two objectives: Concern for regional equity while maintaining efficiency. • Multi-interval response times are optimized in the Basque Emergency Medical Service. • Robust and global validated results show an improvement in success rates. [ABSTRACT FROM AUTHOR]

The recent multi-objective optimization problems are very complex which require an effective and robust evolutionary method for obtaining a global Pareto-optimal solutions. Most of these multi-objective evolutionary techniques are population-based methods which usually work on random search approaches and often trapped into local minima during their execution. A new fuzzy tuned mayfly algorithm (FTMA) is presented in this paper, to obtain the Pareto-optimal solutions for any complex multi-objective power system problem. It has a self-adapting global exploration capability to get the best Pareto-optimal solutions by varying two crucial parameters i.e., crossover (P c) and mutation P m probabilities. Moreover, a model of Pareto-dominance is employed to rank non-dominating solutions to maintain the diversity within the populations. The proposed evolutionary approach is examined on a standard ZDT benchmark test suite and its algorithmic performance is statistically compared with a few of the recent optimization techniques such as NSGA-II, NSGA-III, MMODE, MOAVOA and MMA algorithm. It is further applied to minimize two distinct objective functions (i.e., total transmission loss and voltage stability index) for IEEE-118 bus system and IEEE-24 bus RTS. A comprehensive statistical analysis is presented for the obtained numerical results, which proves that proposed FTMA has better convergence with better solution diversity in comparison to the other competing algorithms. A comprehensive analysis based on statistical results reveal that the proposed FTMA is superior for obtaining better non-dominating pareto-fronts as compared to other competing methods. [ABSTRACT FROM AUTHOR]

In traditional machine learning, there is often a discrepancy in data distribution between the source and target domains. Domain adaptation (DA) was proposed to learn the robust classifier for target domain by using knowledge from different source domains. Most DA methods focus on only the geometric structure of the data or statistical properties to reduce the differences between domains. The complementarity of these two aspects is ignored, which causes the problem of domain underadaptation to some degree. In this paper, we propose latent space search (LSS) approach for domain adaptation that consider both geometric and statistical properties. LSS consists of two parts: latent subspace learning and space search subspace learning. In the latent subspace, the geometric and statistical properties of the data are preserved by a low-rank coupled projection as well as joint distribution of the data. For the solution, an iterative feedback approach is used to obtain a robust subspace. Furthermore, to improve the discriminability of the subspace, the space search optimization algorithm is used to reconstruct the latent subspace, so that the source domain and the target domain can interleave well in the subspace. A comparative study illustrates that the proposed LSS has better performance than other state-of-art models reported in the literature. • Differences in data distribution exist in traditional machine learning. • Geometric and statistical properties are preserved in the latent subspace. • The discriminability of the subspace is raised in space search subspace. [ABSTRACT FROM AUTHOR]

The task of multimodal image fusion aims to preserve the respective advantages of each modality, such as the detailed texture information from visible light images and the salient target information from infrared images. However, images in real environments are often not perfect high-quality images but are affected by various degradation factors such as noise, blur, and glare. Existing unsupervised multimodal image fusion algorithms use undegraded source images as input to fusion networks and generate high-quality fusion images through complex fusion strategies. This single linear learning approach from high-quality to high-quality cannot be applied to the fusion task of degraded multimodal images in real environments. To address this issue, this paper proposes an end-to-end multimodal image fusion network based on a high-order local degradation model (MFHOD). Firstly, inspired by the idea of probabilistic degradation, we propose a high-order local random degradation model (HODM), which inputs the source multimodal images into the degradation model to obtain degraded images before feeding them into the network. Secondly, we design a simple and efficient dual-branch feature extraction encoder to extract deep features from images. Then, from the perspectives of image pixels, brightness, and gradients, we propose an improved composite loss composed of multiple loss functions to constrain network training. Finally, we propose a L 2 -norm fusion strategy for preserving brightness information in low-light nighttime images. Our MFHOD demonstrates good performance on infrared and visible light image datasets as well as medical image datasets. Experimental results show that MFHOD can effectively suppress the effects of glare, noise, and smoke in adverse environments, and also improve the quality of fusion images in low-light and nighttime environments. • Linear fusion methods cannot fully train the feature extraction of neural networks. • Two-stage high-order degradation models significantly boost network nonlinearity. • End-to-end network simplifies multimodal image fusion process. • MFHOD boosts brightness and contrast, suppressing glare, noise, and smoke effects. [ABSTRACT FROM AUTHOR]

Due to the mobile and random nature of services in cyber–physical–social systems (CPSSs), developing service composition approaches that ensure high availability, minimal energy consumption, and high quality of service (QoS) remains a complex challenge. Over the last two decades, several service composition approaches have been proposed in the literature to deal with this challenge. Nevertheless, the existing approaches have certain limitations, particularly in situations where services may move from one location to another, become unavailable due to intensive battery usage, encounter failures, or undergo a decline in quality. These limitations often arise because these approaches do not simultaneously integrate mobility, energy, and QoS constraints while defining the user's movement in a random manner. In this paper, the learning-based swarm optimization-aware service composition algorithm (LS-SCA) is proposed to overcome the aforementioned shortcoming. This approach surpasses existing ones by accounting simultaneously for the user's mobility, energy, and QoS criteria during the service composition process. First, the Small World in Motion (SWIM) mobility model is employed in this study to determine the user's mobility traces, avoiding the random generation of users' traces. Second, an energy consumption model is proposed to increase the energy efficiency by avoiding the overuse of the devices' batteries that can reduce the availability of services and lead to the composition failure. Third, the two-phase learning-based swarm optimizer (TPLSO) method is used in the composition process to find the sub-optimal composition that satisfies the global QoS constraints with the highest utility in terms of mobility, energy, and QoS. Unlike the most existing metaheuristic-based service composition approaches where the overall composition population is improved over a given number of iterations, the TPLSO method is exploited in this paper to improve only a subset of compositions, which reduces the composition time and increases the QoS utility of the composition. The simulation scenarios using two real datasets demonstrate that the LS-SCA approach outperforms six baselines in terms of energy consumption, QoS utility, and availability of composition. This notable performance makes the proposed approach more suitable for real-world applications where energy efficiency, QoS, and availability are crucial factors to consider. [ABSTRACT FROM AUTHOR]

Shoeprint reconstruction is essential in forensic science, but it is also challenging due to various inconsistencies in the patterns, textures, sizes, abrasions, etc. of shoeprints. The computational reconstruction of sharper and more complete shoeprints is conventionally conducted using handcrafted features, and it often requires human intervention. Prior studies using end-to-end machine learning approaches are limited in number and have not achieved a high level of performance. In this paper, we propose a model named ShoeRec, which employs variational autoencoder (VAE) as a component in a U-Net-like architecture to reconstruct missing regions and borders in shoeprint images. ShoeRec incorporates skip connections to preserve key patterns and employs VAE in the bottleneck to facilitate the reconstruction of desired shoeprints with the restoration of detail as perceived by humans. As a U-Net, the model skips the contextual information from the encoder to the decoder, and the compressed features in the latent space via VAE optimize the probabilistic distribution for reconstructing complete shoeprints. The reconstruction operation is automatically tuned according to the objective function, so as to reduce the structural correlation between the original and projected shoeprint and restore the absent information in the desired shoeprints. To the best of our knowledge, ShoeRec is the first deep learning infusion model that specializes in shoeprint reconstruction. The shoeprints reconstructed by ShoeRec have a close match with the originals in terms of both structures and patterns, and ShoeRec outperforms state-of-the-art generative models in human evaluation. • An end-to-end deep learning model focuses on shoeprint reconstruction for forensics. • Reconstruction of existing patterns together with the generation of missing regions. • Missing regions reconstruction is specifically designed based on the loss function. • The proposed model performs well for noisy, blurry, and partial input images. • Model outperforms state-of-the-art generative models in terms of human evaluation. [ABSTRACT FROM AUTHOR]

Drug-target interaction (DTI) prediction refers to the use of computational methods and models to predict the interaction between drugs and biological targets. DTI can help researchers understand the mechanism of action of drugs, discover new drug targets, and screen drug candidates. Recently, a large number of DTI models integrating deep drug-target interaction features have emerged to make up for the dilemma of incomplete information on shallow drug and target features. However, these models ignore the challenge of overlapping interaction information by simply integrating deep interaction information. This paper proposes a flexible DTI prediction with interactive information extraction and trade-off (FDTIIT) to address the above challenges. The main idea of FDTIIT is to use flexible mutual attention to extract interaction information about drugs and targets, and then limit the dependence between them to avoid redundant information. Specifically, FDTIIT mainly includes three parts: drug and target representation, drug-target interactive information extraction, and drug-target interactive information trade-off. Among them, the drug and target representation module mainly uses the graph convolutional network and convolutional neural network to learn the representation of drugs and targets. Then, the drug-target interactive information extraction module extracts the drug information hidden in the target and the target information hidden in the drug based on mutual attention. To avoid possible information overlap between drug representation and target representation after the fusion of interaction information, FDTIIT designs an interactive information trade-off module. This module limits the dependence between drug and target representation, providing more comprehensive information to support high-performance drug-target interaction prediction. Multiple experiments designed on three publicly available datasets validated FDTIIT's effectiveness. • We propose a flexible interactive information extraction module. • We consider the interaction information overlap problem for the first time. • A novel interactive information trade-off strategy is proposed. • We integrate shallow and deep interaction features to predict DTI. • Experimental results on multiple datasets validate the effectiveness. [ABSTRACT FROM AUTHOR]

Extensive research has been conducted on image steganography and watermarking algorithms, owing to their crucial rules in secret data transmission, copyright protection, and traceability. Despite promising results and numerous surveys proposed in the literature, there is still a lack of comprehensive analysis dedicated to deep learning-based image steganography and watermarking algorithms. In this paper, we focus on investigating three important aspects: neural networks, structure models, and training strategies. Our review covers the vast literature in this field. Furthermore, we provide a comprehensive statistical analysis from diverse perspectives, including models, loss functions, platforms, datasets, and attacks. Moreover, we conduct in a thorough comparative analysis and evaluation of existing representative algorithms, assessing their effectiveness within the context of deep learning. Finally, the challenges and potential research directions in the domain of deep-learning image steganography and watermarking algorithms are discussed to facilitate future research. • Comprehensive review of deep learning-based image steganography and watermarking. • In-depth analysis of neural networks, structure models, and training strategies. • Extensive statistical analysis of current models, datasets, attack, and platform. [ABSTRACT FROM AUTHOR]

Mobile charging is feasible to deal with the energy-constrained problem in wireless rechargeable sensor networks (WRSNs). The mobile chargers (MCs) are usually employed to charge the sensors sequentially according to the charging schemes. Existing studies assume that each sensor should be charged to its maximum energy capacity or to a fixed upper threshold before the next one can be charged. However, they neglect to control the charging time adaptively of each sensor according to the charging demand. Hence, in this paper, we assume that the charging time of each sensor can be controlled, and we study the joint optimization of charging sequence and charging time problem (JCSCT). Correspondingly, we propose a novel deep reinforcement learning with hybrid action space approach for JCSCT (DRLH-JCSCT), which utilizes deep q-network (DQN) to generate the charging sequence, and adopts deep deterministic policy gradient (DDPG) to determine the charging time. An attention-based encoder–decoder model is integrated in the actor network of DDPG, and a modified bi-directional gate recurrent unit network (MBGRU) is utilized as the decoder. We also design a novel reward function to evaluate the quality of the charging actions. Simulations demonstrate the improved charging performance of the proposed approach, with a longer network lifetime and fewer failed sensors compared with the existing mobile charging scheduling approaches. [ABSTRACT FROM AUTHOR]

• Hub capacities are considered for the single-allocation Hub Maximal Covering problem. • A new Mixed-Integer Linear-Programming model is proposed and solved for the problem. • A Comprehensive fitness landscape analysis shows the problem space is rugged plain. • A new GARVND algorithm with original Crossover and Mutation operators is proposed. • Experiments showed that GARVND outdid GA, SA, VNS, TS, & GWO+SA methods in objective value. In this paper, a new variation to the Hub Maximal Covering problem called the Capacitated Single-Allocation Hub Maximal Covering Problem (CSAHMCP) is addressed for the first time and a novel binary-integer mathematical model has been proposed. To analyze the features of the problem's solution space and to determine the most effective heuristic approach to solve it, a comprehensive Fitness Landscape Analysis is performed which reveals that the problem is of 'rugged plain' type and can best be solved using an algorithm with good exploration and exploitation capabilities at the same time. Accordingly, we proposed a new algorithm named GARVND, which is a hybrid of the Genetic Algorithm acting as the main framework for diversification, and a variant of the Variable Neighborhood Descent algorithm called 'Reduced VND' acting as a means to improve individual solutions (intensification) by searching in the spaces generated by four types of neighborhood operators. We also introduced two original crossover and mutation operators that can be used for any hub location solution method. The parameters of GARVND are tuned toward less computational times and higher solution quality by using the Taguchi experiment design technique. We solved 114 CSAHMCP instances (several times each) from the Australian Post (AP) dataset using the proposed GARVND, as well as by the Genetic Algorithm, Variable Neighborhood Search, Simulated Annealing, Tabu Search algorithms, and a hybrid algorithm consisting of Grey Wolf Optimizer and Simulated Annealing, in addition to the Gurobi exact solver. Extensive computational results and statistical analyses showed that GARVND yields favorable solutions, outperforming all the compared algorithms by producing smaller gaps with global optimal objective values. In addition, to illustrate the practical value of the CSAHMCP and the GARVND method, we performed a cost-flow sensitivity analysis for the AP Central Sydney network and came up with the best expansion scenarios to be adopted by the management. [ABSTRACT FROM AUTHOR]

This paper aims to propose a new hybrid Machine Learning (ML) with Egret Swarm Optimization Algorithm (ESOA) and Weighted Relevance-based Combination Strategy (WERCS) methods for predicting the patch loading resistance (PLR) of both longitudinally unstiffened and stiffened plate girders. A dataset of both stiffened and unstiffened plate girders under patch loading are collected and used for training and testing to generate the proposed models. Firstly, the WERCS method is used to resample the training set to improve the performance of training data. After that, four ML models, including Gradient Boosted Tree (GTB), Extreme Gradient Boosting algorithm (XGB), Artificial neural network (ANN), and CATBoost regression (CAT) are employed. The hyperparameter of all above ML models are optimized by using ESOA algorithm to choose the best performance models. It was found that the WERCS influences considerably the performance of hybrid ML models, and the WERCS + ESOA + XGB model presents the best performance compared to others. The accuracy of the WERCS + ESOA + XGB model is validated by comparing its predictive results with the existing design codes and formulae. Additionally, the Local Interpretable Model-Agnostic Explanations (LIME) method is used to assess the importance and contributions of each input variable in the proposed model. This algorithm can explain the global and local effects of features with each range and create favourable foundations for practical designing. A Graphical User Interface (GUI) tool is developed to conveniently estimate both the patch loading resistance of longitudinally unstiffened and stiffened plate girders. A Python library has been developed and uploaded to PyPi for more convenient use without the need for accessing the source code. [ABSTRACT FROM AUTHOR]

Lameness is one of the major health problems on dairy farms, which seriously affects dairy cow welfare and increases the risk of premature culling. Accuracy lameness detection ensures timely treatment of hooves and improves the level of health management on dairy farms. Most of the existing lameness detection methods detect one dairy cow, and it is difficult to detect lameness in multiple dairy cows at the same time. In this paper, a lameness detection system based on instance segmentation is constructed to extract the lameness features of multiple dairy cows and automatically detect lameness. First, an improved SOLOv2 network is designed to reduce the semantic gap between low-level and high-level features and improve the precision of dairy cow segmentation. Second, individual matching of dairy cows in different video frames is performed using the Hungarian algorithm. Third, the Canny algorithm is used to extract the neck and back contour features and key gait features of dairy cows. Finally, a deep learning model is constructed, and multiple lameness features are fused to detect the lameness of dairy cows. To evaluate the performance of the constructed system, 172 videos were randomly selected from 246 videos as training videos, and the remaining 74 videos were selected as the test videos. The lameness detection accuracy of the constructed system was 98.65%. The experimental results showed that the constructed system can extract the lameness features of multiple dairy cows at the same time and accurately detect the lameness of dairy cows. [ABSTRACT FROM AUTHOR]

The efficiency of urban traffic management and congestion alleviation relies heavily on accurate forecasting of Origin-Destination (O-D) demand matrices. Existing models primarily focus on estimating O-D demand for various travel purposes throughout the day, which is characterised by its pulsating nature. However, these models often compromise the precision of peak-hour forecasts, leading to unreliable dynamic traffic control and challenges in effectively reducing peak-hour congestion. To tackle this challenge, this paper proposes a novel method for predicting commuting O-D demand matrices. Our method employs community detection algorithms on road networks to precisely partition commute O-D regions, incorporating Points of Interest (POIs). We also present a spatio-temporal dynamic weighted hypergraph model that leverages these partitioned regions, time characteristics from observed O-D trips, and meteorological data to improve forecasting. Comparative analyses with contemporary models and ablation studies indicate our method significantly enhances prediction accuracy, by approximately 5%. These findings imply that the proposed method more effectively encompasses the varied characteristics of commuting during peak hours, thereby providing more accurate demand matrices for urban traffic management. • An integrated framework for community detection and hypergraph representation. • Peak-hour urban commute patterns are captured with embedded POI features. • Time segment and weather condition are used to enhance prediction accuracy. • A spatio-temporal dynamic weighted hypergraph model for O-D prediction. [ABSTRACT FROM AUTHOR]

Aspect-based sentiment analysis (ABSA) aims to conduct fine-grained sentiment analysis, necessitating the extraction of three key components: target entity, aspect category and sentiment polarity. These three components collectively form an integrated ABSA task known as TASD (Target-Aspect-Sentiment jointly Detection). Most of existing approaches on ABSA usually employ Recurrent neural networks(RNNs), Convolutional neural networks(CNNs) or pre-training models such as Bidirectional Encoder Representations from Transformers(BERT). However, they have some common weaknesses. First, most of the existing methods focus on one or two sub-tasks instead of triplet detection, thus they do not establish an end-to-end (training a complex learning system represented by a single model) ABSA model and cannot utilize the relevance of multiple ABSA sub-tasks during training. Second, they cannot achieve accuracy and efficiency simultaneously due to the coupling of context and given aspects. Third, they are poor in recognizing implicit targets. To tackle these limitations, this paper proposes a novel method named the Twin Towers End to End model (TTEE) to solve TASD task. It transforms complex TASD task into a simple end-to-end multi-task framework, simultaneously conducting target and aspect-sentiment detection. It builds twin towers system based on BERT or its updated versions to decouple context and given aspects, which can reduce redundant calculation to improve computational efficiency significantly. It offers great advantage to identify implicit target entity and its associated aspect-sentiment in the context without introducing extra model architecture. Experiments on three public datasets in different domains demonstrate that our approach not only achieves better performance on various evaluation metrics, but also has high efficiency in both training and inference phases, over a wide range of sample size and number of aspect categories. • The T-A-S triplets can be decoded from the end-to-end output of TTEE. • BERT-based model is employed to encode context and aspect-sentiment pairs separately. • TTEE efficiently reduces redundant calculations of embedding process. • TTEE excels in detecting implicit targets within the context. [ABSTRACT FROM AUTHOR]

Electromechanical systems is widely employed in the manufacturing industry, with fault diagnosis being critical for ensuring the reliable operation of them. Vibration signals exhibit distinct fault features, but their acquisition is subject to various limitations. Conversely, current signals, while easily measurable, typically manifest weak fault features. Therefore, selecting a signal for fault diagnosis necessitates a trade-off among cost, performance, and feasibility. To overpass these obstacles and enable flexible fault diagnosis in complex environments, this paper presents a novel contrastive vibration-current (CVC) framework that leverages synchronization information from multiple modalities to enhance the performance of single-modality models, primarily the current model. This allows the choice of any signal for monitoring during the deployment phase. Specifically, we first preprocess current signals using Clarke transformation to highlight their fault information. Subsequently, the vibration model employs semi-supervised learning to make full use of labeled and unlabeled samples. Additionally, a noise-resistant augment Mean Teacher is incorporated to enhance the robustness of the vibration model. Then, using synchronization-induced cross-modal contrastive learning (SICMCL), vibration and current features are aligned. And at the decision level, the current model leverages pseudo-labels derived from vibration. The results of experiments demonstrate that CVC excels when relying solely on single-modal signals, owing to the effectiveness of SICMCL. Moreover, for the current model, SICMCL is more beneficial in improving performance than true labels. [ABSTRACT FROM AUTHOR]

Aiming at addressing the economic dispatching challenges within the park-level integrated energy system (PIES), this study focuses on three PIES with distinct priorities. The proposed approach introduces a two-stage intelligent dispatching strategy for PIES based on a sharing mechanism. This strategy comprises a day-ahead intelligent dispatching method for PIES, grounded in a scenario threshold, and a real-time dispatching strategy for PIES based on the sharing mechanism. In the day-ahead phase, a scenario threshold concept is presented for an individual PIES and resolved through a novel modeling method and an intelligent algorithm to derive the day-ahead dispatching plan for the PIES. Subsequently, in the real-time phase, the synergy among the three PIESs is explored, and the sharing mechanism is leveraged to offset the uncertainty deviation observed in the high-priority PIESs. To meet the energy requirements of each PIES, the energy storage system is configured within the lowest-priority PIES. Simulation results demonstrate that the dispatch model developed in this paper exhibits faster convergence and possesses the capability to compensate for uncertainty deviations, thereby fostering a balance between source-load supply and demand. Additionally, this model establishes a research foundation for the cooperative optimal dispatch of multiple PIESs. [ABSTRACT FROM AUTHOR]

The primary objective of this paper is to aid game developers in finding the most suitable pathfinding algorithm for their games. Despite recent advancements in this field, there are few available studies that can be compared due to the absence of a standard benchmark set for weighted environments. This paper presents a new dataset for pathfinding in weighted environments. Furthermore, an investigation was conducted into the impact of node weights on pathfinding speed, and a correlation between them was identified. The complexity added to the maps due to node weights was defined as weight complexity, and two metrics were introduced to estimate it. The weight correlation factor has been identified as the most effective metric for estimating the weight complexity of the map. Another contribution of this paper pertains to the development of a model for estimating the pathfinding speed of algorithms based on weight complexity. This was accomplished through the utilization of the non-linear least squares method, which was applied to create a model for each algorithm, considering both its average search time and weight correlation factor values associated with the map. Finally an overall score metric was developed by using the integral of the models, enabling the evaluation of different algorithms in various scenarios. [Display omitted] • Testing JPSW (weighted jump point search) algorithm search speed. • Introducing the first standardized benchmark for weighted gridmaps. • Introducing a new metric for analyzing gridmap complexity. • Using edge detection techniques for weighted gridmap analysis. • Introducing a scoring system for pathfinding algorithms in weighted environments. [ABSTRACT FROM AUTHOR]

The prevalence of security vulnerabilities in Internet of Things (IoT) applications poses a serious threat to enterprise systems, necessitating sophisticated and reliable defense solutions to counter emerging and evolving threats. For the Industrial Internet of Things (IIoT), stakeholders require trustworthy and sustainable systems that can prevent the loss of human life during critical operations. The impact of multi-variant persistent and sophisticated bot attacks on connected IIoTs is potentially catastrophic, and their detection presents a highly complex and critical challenge. Therefore, there is a pressing need for efficient and timely detection of IIoT botnet attacks. This research paper proposes a robust deep learning model named AttackNet for the detection and classification of different botnet attacks in IIoT based on adaptive based CNN-GRU model. The model is extensively evaluated using the latest dataset and standard performance evaluation metrics, demonstrating its capacity to protect IIoT networks against sophisticated cyber-attacks with a testing accuracy of 99.75%, a loss of 0.0063, precision and recall score of 99.75% and 99.74% respectively. Our proposed model demonstrates superior accuracy, particularly within the N_BaIoT dataset. It achieves an outstanding accuracy of 99.75% across ten classes, surpassing state-of-the-art techniques by a substantial margin ranging from 3.2% to 16.07%. Moreover, the proposed model outperforms state-of-the-art anomaly detection systems in IIoT based on a real-time IoT device dataset in terms of detecting and classifying botnet attacks accurately. [ABSTRACT FROM AUTHOR]

Federated Learning is a collaborative machine learning paradigm that allows training models on decentralized data while preserving data privacy. It has gained significant attention due to its potential applications in various domains. However, the issue of protecting model copyright in the Federated Learning setting has become a critical concern. In this paper, we propose a novel watermarking framework called FedCRMW (Federal Learning Compression-Resistance Model Watermark) to address the challenge of model copyright protection in Federated Learning. FedCRMW embeds unique watermarks into client-contributed models, ensuring ownership, integrity, and authenticity. The framework leverages client-specific identifiers and exclusive logos to construct trigger sets for watermark embedding, enhancing security and traceability. One of the key advantages of FedCRMW is its optimization for the common data compression challenge in the Federated Learning scenario. By utilizing compressed data inputs for copyright verification, we achieve an efficient watermark validation process and reduce communication and storage overheads. Experimental results demonstrate the effectiveness of FedCRMW in terms of watermark success rate, imperceptibility, robustness against attacks, and resistance to model compression and pruning. Compared to existing watermarking methods, FedCRMW exhibits superior performance in the Federated Learning context. • FedCRMW: Robust watermarking for Federated Learning models. • Novel trigger dataset construction scheme for watermarking. • Enhanced robustness with feature-consistent training. [ABSTRACT FROM AUTHOR]

With the increase in the aging population of many countries, the prevalence of neovascular age-related macular degeneration (nAMD) is expected to increase. Morphological parameters such as intraretinal fluid (IRF), subretinal fluid (SRF), subretinal hyperreflective material (SHRM), and pigment epithelium detachment (PED) of spectral-domain optical coherence tomography (SD-OCT) images are vital markers for proper treatment of nAMD, especially to get the information of treatment response to determine the proper treatment interval and switching of anti-vascular endothelial growth factor (VEGF) agents. For the precise evaluation of the change in nAMD lesions and patient-specific treatment, quantitative evaluation of the lesions in the OCT volume scans is necessary. However, manual segmentation requires many resources, and the number of studies of automatic segmentation is increasing rapidly. Improving automated segmentation performance in SD-OCT visual results requires long-range contextual inference of spatial information between retinal lesions and layers. This paper proposes a GAGUNet (graph convolution network (GCN)-assisted attention-guided UNet) model with a novel global reasoning module considering these points. The dataset used in the main experiment of this study underwent rigorous review by a retinal specialist from Konkuk University Hospital in Korea, contributing to both data preprocessing and validation to ensure a qualitative assessment. We conducted experiments on the RETOUCH dataset as well to demonstrate the scalability of the proposed model. Overall, our model demonstrates superior performance over the baseline models in both quantitative and qualitative evaluations. • We demonstrate the limitations of the baseline in retinal segmentation. • We propose a novel model for automated retinal OCT segmentation. • The proposed model performs higher than the baseline in retinal segmentation. [ABSTRACT FROM AUTHOR]

Session-based recommendation leverages the short-term interaction sequence to predict the next item a user is most likely to click on. Generally, in real applications, users often click on different types of items in the same session, which makes the items in the sequence present diversity, and degree of diversity vary with different sequences, that is, there is a phenomenon of personalized diversity. However, the majority of existing session recommendation models primarily focus on improving recommendation accuracy and overlooking the importance of personalized diversity, which leads to the issue of filtering bubbles. In order to alleviate the long-tail effect caused by the filtering bubble, the recommendation system should not only consider the accuracy of recommended items, but also the diversity of users' demands for recommended items. To address this issue, this paper proposes a model, named as LAP-SR, a L ong-tail A lleviation P ost-processing framework based on personalized diversity of S ession R ecommendation. After obtaining the initial recommendation list through the selected initial session model, LAP-SR model generates the long-tail item set and item graph according to session contents. Besides, from the item graph, the model calculates the embedding representation of items, from which the diversity of all sessions is calculated. By combining the obtained diversity and the long-tail item set, the initial recommendation list is soft processed. On this basis, the model yields the final recommendation list, which alleviates the overall long-tail effect. We verified that as a post-processing framework, the proposed model can be applied to various session recommendation models. Extensive experiments on the three datasets of Diginetica, Tmall and Yoochoose1_64 demonstrate that the LAP-SR model has a competitive advantage over the baseline models in mitigating the long-tail effect. • LAP-SR is based on personalized diversity to alleviate the long-tail effect. • The model consists of an initial recommendation model and a post-processing model. • LAP-SR is a universal post-processing model for session-based recommendations. • LAP-SR can significantly improve the exposure of long-tail items. [ABSTRACT FROM AUTHOR]

Exploring the semantic correspondence of image-text pairs is significant as it bridges vision and language. Most prior works focus on global semantic alignment or local semantic alignment, by developing a fine neural network that facilitates the corresponding alignment but neglects the semantic information and relative position information between image regions, or text words, which will lead to a non-meaningful alignment. To this end, a Heterogeneous Graph Fusion Network (HGFN) is proposed to explore the correlation score of vision-language for improving the accuracy of cross-modal image-text retrieval in this paper. Specifically, we first construct an undirected fully-connected graph based on the semantic or relative position information for each image, as well as a textual graph with neighborhood information of the text. Then, we present a graph fusion module to integrate the features of heterogeneous graphs into a unified hybrid representation, in which the graph convolutional network is utilized to gather neighborhood information to alleviate potentially non-meaningful alignment. In addition, we also propose a novel "Dynamic top- K negative" strategy for the selection of negative examples in the training process. Experimental results demonstrate that HGFN achieves comparable performance with state-of-the-art approaches on the Flickr30K and MSCOCO datasets. • Cross-modal image-text retrieval problems can be treated as graph-graph matching. • A graph fusion module is designed to fuse the visual graph and textual graph. • A novel strategy for selecting negative examples is proposed. • Competitive with state-of-the-art performance on two standard benchmarks. [ABSTRACT FROM AUTHOR]

In the realm of decision-based attacks, which aim to mislead target models by manipulating output labels, the advent of the polar coordinates attack has marked a significant evolution. Distinguished by its utilization of sampled affine planes with unique geometrical properties, this method has demonstrated notable enhancements in query efficiency. However, existing implementations of the latest polar coordinates attack encounter two major problems: a contradiction in the update criteria and an overemphasis on allocating query resources to angle binary search. To address these issues, the Efficient Polar Coordinates Attack (EPCA) is introduced, a novel approach that resolves these concerns while significantly augmenting query efficiency. Furthermore, a mechanism termed Adaptive Activation Strategy is proposed to control the deployment of a recently presented plug-and-play module named Frequency Binary Search. The integration of these elements further enables EPCA to effectively circumvent the local optimum. Abundant experiments conducted on the ImageNet dataset demonstrate the exceptional improvements achieved by EPCA, establishing it as a state-of-the-art method in decision-based attacks. The simulation codes of this paper are accessible at https://github.com/RYC-98/EPCA. • Two problems in polar coordinates attacks are revealed. • Efficient Polar Coordinates Attack (EPCA) is used to solve these problems. • An Adaptive Activation Strategy with Frequency Binary Search is proposed. • EPCA attains the SOTA performance. [ABSTRACT FROM AUTHOR]

Adversarial sample generation problem is a hot issue in the security field of deep learning. Evolutionary algorithm has been widely used to solve this problem in recent years because of its good global search ability. However, existing methods still suffer from the "curse of dimensionality" when attacking high-resolution images. In this paper, a two-stage frequency domain generation algorithm of black-box adversarial samples based on differential evolution is proposed. In the first stage, a representative image-guided differential evolution method is proposed to quickly generate a universal adversarial perturbation with a high attack success rate in the frequency-domain. In the second stage, a space reduction strategy based on frequency-domain pixel blocks is designed to reduce the search space and alleviate the problem of "curse of dimensionality". In addition, a new space–frequency interaction sensitivity measure is introduced to evaluate the similarity between the adversarial samples and the original images. The adversarial perturbations obtained by the measure are more in line with the subjective perception of the human eye. Finally, compared with several typical black-box adversarial sample generation algorithms, experimental results show that the proposed algorithm can achieve higher attack success rate with less prediction times. • Proposing a two-stage evolutionary generation framework of adversarial samples. • Developing a space–frequency interaction sensitivity measure. • Presenting a representative image-guided differential evolutionary algorithm. • Designing an individual adversarial perturbation generation algorithm. [ABSTRACT FROM AUTHOR]

The fifth-generation wireless telecommunication technology (5G) with high speed, wide coverage and large amount of data have posed new requirements for the running time, security, and watermark capacity of digital watermark algorithms. For realizing the copyright protection of color image in 5G environment, a spatial digital watermarking scheme for dual color image fusing Hadamard transform is proposed in this paper. In this scheme, the protected color cover image is firstly channeled and then divided into 4 × 4 non-overlapping image blocks. Then, the maximum energy coefficient of image block in the Hadamard domain is acquired directly in the spatial domain, and the watermark information is hidden by quantizing this coefficient. Finally, the changed quantization variation of the special coefficient are evenly distributed to each coefficient of the original spatial image block, so as to realize the complete process of hiding the watermark. In the extraction stage, only the special coefficient of the watermarked image in the Hadamard domain is obtained in the spatial domain, and the coefficient threshold is analyzed to determine the coding of watermark information. This scheme combines the characteristics of spatial domain algorithm and frequency domain algorithm, and designs a set of high security encryption method, at the same time, considers the human vision system to color sensitivity and the correlation of color image channel. Compared with other schemes, it can be found that this scheme has excellent security and real-time performance, strong robustness to common attacks, and high image quality. [ABSTRACT FROM AUTHOR]

The data distribution shifting under varied working conditions impedes the application of deep learning models for fault diagnosis. Transfer learning can reduce the discrepancy in data distribution and improve model generalization from one condition with abundant labeled source data to another condition with unlabeled target data. However, existing transfer learning methods may hardly be applied to scenarios with few source domain data, and the insufficient source data hinder the establishment of the initial model. To address this issue, a deep semi-supervised transfer learning method with sensitivity-aware decision boundary adaptation (Sa-DBA) is proposed in this paper. The proposed triangular consistency in semi-supervised learning module can ensure the construction of the initial decision boundary. In transfer process, the Sa-DBA directly reduces the discrepancy of predicted matrixes between two domains to facilitate the adaptation of decision boundaries to the target data. Given that target data adjacent to decision boundaries may exhibit varying boundary sensitivities across different tasks, the nuclear-norm with task-specific parameter is introduced to further refine decision boundaries. Finally, extensive experiments on three diverse datasets, including CWRU, PHM, our own DPS test bench, and the best average accuracy reached 99.70%, 86.41%, and 88.59% respectively, which performed better than the other methods. • A deep semi-supervised transfer learning framework is proposed. • The triangular regularization is proposed in semi-supervised learning module. • Our method aligns predicted matrixes to reduce the domain discrepancy. • The task-weighted nuclear-norm is adopted to further refine decision boundaries. • Our method outperforms current methods on fault diagnosis with few source data. [ABSTRACT FROM AUTHOR]

Online group streaming feature selection holds significant research value in large-scale streaming data processing scenarios, and the related work based on rough set theory has attracted academic interest. Nevertheless, most relevant algorithms come with parameters, and performing a grid search for the optimal parameters reduces the efficiency. Moreover, the existing online group streaming feature selection frameworks cannot effectively fit the situation in practice. Focusing on these issues, this paper investigates an online group streaming feature selection method based on fuzzy neighborhood granular ball rough sets. First, Canopy clustering is introduced to granular ball computing, and the adaptive neighborhood of samples is generated based on the granular ball distribution. Second, we construct a fuzzy neighborhood granular ball rough set (FNGBRS) model and propose the integrated dependence degree to achieve maximal dependency and minimum classification error. Then, the purity of granular balls is considered as the weight of features, and some uncertainty measures based on FNGBRS are presented. Finally, we define a random factor to control the size of streaming groups and design an online group streaming feature selection algorithm. Comparative experimental results on sixteen public datasets demonstrate that the proposed algorithm exhibits superior and stable classification performance, coupled with increased efficiency from its parameter-free design. • Generate the sample's parameterless fuzzy neighborhood information granule. • We construct a fuzzy neighborhood granular ball rough set model. • Uncertainty measures are developed using the purity of granular balls as weights. • Control the size of feature streaming groups by a random factor. [ABSTRACT FROM AUTHOR]

Mesoscale Eddy (ME) is a widely recognized and significant oceanic phenomenon characterized by extensive energy exchange. Accurate predictions of the future geospatial distribution of ME are crucial for maritime activities. Deep learning (DL) methods for ME prediction have consistently outperformed classical mathematical models and numerical modeling approaches. However, current DL methods based on two-dimensional gridded data suffer from error accumulation issues due to their indirect prediction approach, which involves first predicting the future environmental field and then identifying the ME patterns within it. Additionally, these models ignore the potential influence of prior knowledge related to ME on prediction outcomes. To address these issues, this paper proposes a direct prediction DL network called ES-ConvGRU, which incorporates historical prior statistics. First, we conduct an analysis of ME data from the Northwest Pacific region, extracting prior statistical knowledge such as interannual variations and seasonal characteristics. Second, mining patterns for the spatiotemporal features of both ME sequence and sea surface environmental field sequence are established. Furthermore, a nonlinear approach for mapping prior statistical knowledge and a multi-step autoregressive prediction model are designed, facilitating the integration of prior knowledge with multidimensional spatiotemporal features for ME prediction. Finally, the performance and effectiveness of ES-ConvGRU are evaluated. It achieves an average pixel accuracy of 93.21%, a mean intersection over union of 81.48%, and a frequency-weighted intersection over union of 87.55% for the 7-day distribution prediction. The results show that ES-ConvGRU exhibits notable advantages compared to the existing indirect and direct prediction methods. • A direct predictive approach for mesoscale eddy distribution prediction is developed. • Prior seasonal statistics is considered for prediction. • An end-to-end deep prediction network architecture, ES-ConvGRU, is proposed. • The results show that ES-ConvGRU performs better than other predictive methods. [ABSTRACT FROM AUTHOR]

The extended producer responsibility system and government intervention are among the essential requirements for constructing sustainable economies. In a closed-loop supply chain consisting of an original equipment manufacturer (OEM) and a third-party remanufacturer (TPR), this paper develops four Stackelberg game models for four modes of government intervention under patent licensing: no intervention, tax and subsidy, tax and subsidy plus eco-design tax deduction, and eco-design tax deduction only. The impacts of the different government interventions on the optimal production decision, social welfare, and environment are analyzed and compared. The results indicate that, compared with no government intervention, government tax and subsidy decrease the supply of new products, increase supply of remanufactured products, and raise recycling rates of end-of-life products; on top of tax and subsidy, the government's eco-design tax deduction can further increase OEM and TPR profits; the environmental impact under different government intervention depends on the ratio of the environmental impacts of remanufactured to new products and the level of eco-design effort. This study provides decision makers (i.e., OEM and TPR) and policy makers (i.e., governments) with new managerial insights on the benefits of eco-design and effective mechanisms to promote it. [ABSTRACT FROM AUTHOR]

In visual tracking, Convolutional Neural Network (CNN) is usually used as feature extractor, and can fully explore local dependencies of image blocks, which is help for improving tracking performance. However, CNN ignores global dependencies in image blocks. The global modeling is crucial in visual tracking. Recently, Transformer has gained attention to fully explore global dependencies on sequential data. However, Transformer's unique multi-head self-attention mechanism results in high computational complexity. In this paper, we design a pyramid pooling based Transformer backbone network for visual tracking. Pyramid pooling refers to multiple pooling operations for feature map with different receptive fields and strides. The output data of each pooling layer is concatenated to form the final pooled feature map. On the one hand, after flattening the feature map with pyramid pooling, its sequence length will be greatly reduced. This will effectively reduce the computational complexity of the multi-head self-attention. On the other hand, pyramid pooling can extract multi-scale features, makes the feature maps contain more global context information. Finally, we propose a novel tracker with the designed pyramid pooling based Transformer backbone network and the Transformer based model predictor. We train the proposed tracker by end-to-end, and evaluate it on seven tracking benchmarks including UAV123, NFS, Trackingnet, LaSOT, GOT-10K, VOT2020 and RGBT2019. The proposed tracker achieves 79.8% robustness and 35 FPS on the VOT2020 dataset. The experiment demonstrates that proposed tracker achieves superior tracking performance with state-of-the-art trackers. • A pyramid pooling based Transformer backbone network is designed for visual tracking. • The local and global dependencies are explored on sequential data. • The pyramid pooling effectively extract the spatial context information. • The pyramid pooling reduces the computational complexity of multi-head self-attention. • A tracker is proposed based on the backbone network and Transformer model predictor. [ABSTRACT FROM AUTHOR]

• The interval data effectively aggregated by the induced continuous aggregation operator. • The Shanghai composite index can be accurately predicted by the combination prediction model. • Thorough study of effectiveness theory of the combination prediction model has been completed. This paper proposes an interval combination prediction model for Shanghai composite index, utilizing the Theil coefficient and the induced continuous generalized ordered weighted logarithmic harmonic averaging (ICGOWLHA) operator. The effectiveness of the proposed model under specific weight conditions and the existence of its analytical solution are demonstrated. Shanghai composite index's case analysis demonstrates that, in terms of interval root mean squared error (IRMSE), interval mean absolute error (IMAE), interval mean absolute percentage error (IMAPE), and interval mean squared percentage error (IMSPE), the proposed model's predictive performance improvements over the best-performing single prediction model are 29.33%, 25.72%, 26.10%, and 28.86%, respectively. At the same time, the theoretical properties of the model are verified in the results of the case analysis, and the model's convergence is reflected in sensitivity analysis. Through extensive model comparisons, it is observed that the model proposed in this paper exhibits strong generalization, without specific limitations on data size or feature count. It demonstrates good aggregation prediction performance for interval data. Moreover, it is applicable to various fields, including finance, environment, and others. [ABSTRACT FROM AUTHOR]

In the era of digital markets, the challenge for consumers is discerning quality amidst information asymmetry. While traditional markets use brand mechanisms to address this issue, transferring such systems to internet-based P2P markets – where misleading practices like fake ratings are rampant – remains challenging. Current internet platforms strive to counter this through verification algorithms, but these efforts find themselves in a continuous tug-of-war with counterfeit actions. Exploiting the transparency, immutability, and traceability of blockchain technology, this paper introduces a robust reputation voting system grounded in it. Unlike existing blockchain-based reputation systems, our model harnesses an intrinsically economically incentivized approach to bolster agent integrity. We optimize this model to mirror real-world user behavior, preserving the reputation system's foundational sustainability. Through Monte-Carlo simulations, using both uniform and power-law distributions enabled by an innovative inverse transform method, we traverse a broad parameter landscape, replicating real-world complexity. The findings underscore the promise of a sustainable, transparent, and formidable reputation mechanism. Given its structure, our framework can potentially function as a universal, sustainable oracle for offchain-onchain bridging, aiding entities in perpetually cultivating their reputation. Future integration with technologies like Ring Signature and Zero Knowledge Proof could amplify the system's privacy facets, rendering it particularly influential in the ever-evolving digital domain. • System penalizes divergent ratings, fostering genuine, consensus-aligned feedback. • Adaptive staking aligns with market dynamics, optimizes task assignment efficiency. • Balanced stakeholder equilibrium amplifies credit point value and market impact. • Reputation system offers a unique, sustainable blockchain oracle solution. [ABSTRACT FROM AUTHOR]

During past several decades, extensive interdisciplinary collaboration has greatly promoted the scientific innovation. Especially, collaborations among top scholars have become major driving forces in advancing scientific progress. Therefore, this article focusses on the mechanism and modes underlying collaboration of scholars. The advantage of this paper resides in its capacity to elucidate the hierarchical architecture and dynamics of top scholars' collaboration, discern the hidden collaborative association probabilities, and provide recommendation for fostering intellectual exchange and information dissemination. Results show that: (1) In terms of collaboration features, it exhibits a significant clustering effect. The largest three communities concentrate in Asia, Europe, and North America. Meanwhile, we further introduce dominant flow analysis, an algorithm originated from the geography, to excavate hierarchy architecture of collaborations. We have identified three kinds of scholars who perform distinct roles in the dissemination of knowledge: leading scholars tend to play a guiding role, senior scholars tend to act as a bridge to enlarge the collaboration range, and general scholars predominantly contribute to the collaborations. (2) In terms of formation mechanism of collaborations, we found the collaboration of top scholars are grounded in their publications' quality instead of quantity. Furthermore, the closed rather than opened triangular configuration potently catalyzes knowledge transmission. (3) Meanwhile, hidden collaborative partnerships are extrapolated based on the model proposed in this article. In these hidden patterns, scholars who are geographically close and have similar research background potentially develop further research collaborations, and at least one scholar in collaborations has high citations. [ABSTRACT FROM AUTHOR]

Learning the discriminative features of different faces is an important task in face recognition. By extracting face features with neural networks, it becomes easy to measure the similarity of different face images, which makes face recognition possible. To enhance a neural network's face feature separability, incorporating an angular margin during training is common practice. The state-of-the-art loss functions CosFace and ArcFace apply fixed margins between the weights of classes to enhance the inter-class separation of face features. Since the distribution of samples in the training set is uneven, the similarities between different identities are unequal. Therefore, using an inappropriately fixed angular margin may lead to problems such as that the model has difficulty converging or that the face features are not sufficiently discriminative. It is more intuitive to use adaptive angular margins are angular adaptive, which can increase as the angles between classes increase. In this paper, we propose a new angular margin loss named X2-Softmax. X2-Softmax loss has adaptive angular margins, which increase as the angle between different classes increases. The angular adaptive margin ensures model flexibility and effectively improves the effect of face recognition. We trained a neural network with X2-Softmax loss on the MS1Mv3 dataset and tested it on several evaluation benchmarks to demonstrate the effectiveness and superiority of our loss function. [ABSTRACT FROM AUTHOR]

The human-centric philosophy is considered a crucial development direction for smart Product-Service Systems (smart PSS). Smart PSS has effectively advanced the circular economy by enhancing user experience, incorporating servitization and digital servitization, and extending product lifecycle, among other key aspects of sustainable development. However, there is currently a lack of a reasonable, accurate, and actionable framework and methodology to express the human element, namely user profiling, within this philosophy. To address this gap, this paper utilizes statistical methods such as factor analysis, cluster analysis, and regression analysis, building upon the concept of traditional user profiling. The aim is to integrate the three prominent approaches of goal-oriented, scenario-based, and data-driven user profiling, with the goal of complementing each other and designing a top-level user profiling framework for smart PSS. Furthermore, Industry 4.0 technologies and text mining techniques are employed to collect data on users' product and service usage. As these data contain real-time information about user needs, behaviors and goals at different time periods, they can be used to construct dynamic features of user profiling, and ultimately achieve the construction of dynamic user profiling for smart PSS. To validate the proposed user profiling framework and dynamic user features of smart PSS, this paper presents a case study focusing on the user group of maternal women. This case study promotes the in-depth exploration of smart PSS research in expressing the human element. [ABSTRACT FROM AUTHOR]

Images captured under low-light conditions usually co-exist with low light and blur degradation. Most existing cascade and joint enhancement methods may provide undesirable results, suffering from severe artifacts, deteriorating blur, and unclear details. In this paper, we propose a novel network with structural priors, including high-frequency and edge, to enable effective image representation learning for joint low-light enhancement and deblurring. Specifically, we employ a Transformer backbone to explore the global clues of the image. To compensate for the inadequate local detail optimization, we propose a multi-patch perception pyramid block that models the correlation between different size patches and ambiguity, and identifies non-uniform deblurring spatial features, facilitating the reconstruction of potential high-frequency and edge information. Furthermore, a prior-guided reconstruction block based on the parallel attention mechanism is present to adaptively correct global image with statistical features, which helps guide the model to refine sharp texture and structure. Extensive experiments performed on simulated and real-world datasets demonstrate the efficacy of our proposed method in restoring low-light blurry and low-light images with increased visual perception compared to state-of-the-art methods. • Multi-structural priors are employed to joint low-light enhancement and deblurring. • A multi-patch perception pyramid block to remove non-uniform is proposed. • Prior-guided reconstruction block to adaptively correct global features is present. • The proposed method is extensively evaluated and attains superior performance. [ABSTRACT FROM AUTHOR]

Many studies on predicting firm failure have focused on finding new features that improve the accuracy of the models. In this paper, genetic programming (GP) is used for this purpose. The main problem in GP is to specify a function that evaluates the fitness of the feature. Direct optimization of a machine learning (ML) model that uses a generated feature in most cases leads to high computational costs since evolving a population of N programs over G generations while evaluating each model using K -fold cross-validation requires N * G * K model learning cycles. Thus, many researchers use scores that measure the relationship of the generated features to the class label. However, our empirical analysis shows that most such scores correlate poorly with ML model performance. The novelty of our work is that we introduce several ways of combining different scores into a single measure of expected model performance. Experimental results on data from Hungarian firms (7167 observations, class imbalance 9.37) using five ML models (Logistic Regression, Random Forest, Gradient Boosting, Histogram Boosting, and AdaBoost) prove that the proposed way of setting the fitness function increases the ROC AUC of the listed models by 6.6%, 5.2%, 6.8%, 5.5% and 5.2% respectively. Moreover, by applying the found formula to the data from Czech firms (3872 observations, class imbalance of 74.92), which were not used for the feature search, we obtained increases in ROC AUC by 13.1%, 11.8%, 14.9%, 9.9%, and 8.2%, respectively. This indicates that the proposed method allows to find universal features, which opens the way to build effective models in case of insufficient data (small number of observations, extreme imbalance, etc.). [ABSTRACT FROM AUTHOR]

As the data-driven Francis turbine units (FTUs) deterioration assessment method is widely investigated, the data quality in actual industrial scene has become an important prerequisite to restrict the method performance. However, existing data augmentation methods often lie in simply mixing real data and simulated data, ignoring the inherent cross-domain state relationships. In this paper, a data-model-interactive enhancement-based FTU health condition assessment using graph driven health benchmark model (HBM) is proposed. First, the pseudo-signals outputs of the mechanism digital twin (DT) model by calculative fluid dynamics (CFD) calculation are modified by critical working condition parameters, extending the theoretical assessment domain. To achieve data-model-interaction through cross-domain state coordination of digital and real data, a digital-reality hybrid graph is constructed from nodes mixing both actual and pseudo through node similarity. The hybrid spatial condition structure created by capture node status association explicitly enhances the sample representation ability. Avoiding excess noise effects, a knowledge-based unsupervised graph pruning regulation is adopted to refine the original graph considering noise and data reality. Finally, the spatial–temporal dependencies hidden in the refined graphs are mined by the designed hybrid graph neural network-based HBM, and output the assessed degradation label values. Verification experiments show that the proposed assessment method can effectively fit FTU deterioration under low-quality onsite data, and is more robust than the comparison methods. [ABSTRACT FROM AUTHOR]

It is a crucial means for humans to perceive geomorphic features and landscape architectures by classifying ground objects in hyperspectral images (HSIs). Currently, the exponential development of neural networks has provided a powerful support for the accurate HSI classification. However, existing neural network-based methods usually rely solely on the data to drive the classification model, lacking attention to valuable land-cover distribution knowledge in HSIs. In view of this, to utilize hyperspectral data and distribution knowledge simultaneously, a data and knowledge-driven deep multiview fusion network based on diffusion model (DKDMN) is proposed in this paper. DKDMN extracts knowledge from unlabeled data in HSIs through a diffusion model-based knowledge learning framework (DMKLF), and combines raw hyperspectral data with the acquired knowledge through a designed deep multiview network architecture (DMNA) to mine complicated land-cover distribution information and reflect sample relationships. First, the proposed DMKLF utilizes the data distribution reconstructed by the diffusion model as a knowledge source for one view to enhance the network cross-sample awareness ability. On the other hand, the original HSI patches are considered a data source for another view, which co-drives DMNA with the unsupervised diffusion knowledge extracted by DMKLF to perform effective feature extraction. Second, taking into account the characteristics of each view and the feature similarity between these two views, a joint loss function specifically for DMNA is suggested to minimize the difference between the model predictions and the real labels. Finally, a multi-backbone integration classification framework (MBICF) is designed by deeply fusing three vision architectures to capture multi-scale spectral features and local–global features, thereby achieving pixel-wise classification effectively. Experimental results on four publicly available HSI datasets demonstrate that the proposed DKDMN achieves competitive classification accuracy compared with other state-of-the-art methods. For instance, the proposed DKDMN achieves an overall accuracy improvement of 1.62% and 2.18% on the Indian Pines and Salinas Valley datasets, respectively, compared to the multiple vision architecture-based hybrid network (MVAHN). The related code will be released at https://github.com/ZJier/DKDMN. • Designing a data and knowledge-driven HSI classification method. • Learning unsupervised spectral spatial distribution knowledge. • Combining data source and knowledge source by deep multiview network architecture. • Realizing HSI classification by multi-backbone integration classification framework. [ABSTRACT FROM AUTHOR]

• A pioneering framework called Cross-to-Merge Training (C2MT) is proposed for learning with noisy labels (LNL). • We are the first to perform parameter aggregation on LNL after cross-training. • We introduce a novel class-balancing strategy, the Median Balance Strategy (MBS), which proves useful for sample selection. • C2MT demonstrates robustness with respect to hyper-parameters and network architectures. • Comprehensive experimental results substantiate the advantages of C2MT. The collection of large-scale datasets inevitably introduces noisy labels, leading to a substantial degradation in the performance of deep neural networks (DNNs). Although sample selection is a mainstream method in the field of learning with noisy labels, which aims to mitigate the impact of noisy labels during model training, the testing performance of these methods exhibits significant fluctuations across different noise rates and types. In this paper, we propose Cross-to-Merge Training (C2MT), a novel framework that is insensitive to the prior information in sample selection progress, enhancing model robustness. In practical implementation, using cross-divided training data, two different networks are cross-trained with the co-teaching strategy for several local rounds, subsequently merged into a unified model by performing federated averages on the parameters of two models periodically. Additionally, we introduce a new class balance strategy, named Median Balance Strategy (MBS), during the cross-dividing process, which evenly divides the training data into a labeled subset and an unlabeled subset based on the estimated loss distribution characteristics. Extensive experimental results on both synthetic and real-world datasets demonstrate the effectiveness of C2MT. The Code will be available at: https://github.com/LanXiaoPang613/C2MT. [ABSTRACT FROM AUTHOR]

The growing demand for cloud computing and storage necessitates the expansion of Data Centers, thereby increasing energy consumption and environmental footprint. The data-driven dynamic thermal control presents a promising solution to mitigate this issue, as it can offer improved thermal control strategies and address the intricate and stochastic nature of the thermal environment. Nonetheless, existing solutions typically exhibit deficiencies in terms of reliability and practicability. To this end, this paper proposes a feasible uncertainty-aware online learning method with reliable dynamics models and controllers. Specifically, Deep Learning models with the capacity for uncertainty quantification are employed to account for the imperfections in pre-trained models, hence providing reliable predictions. The Cross-Entropy Method and Monte Carlo trajectory sampling solve optimization problems while considering uncertainties. It is mathematically analyzed that the uncertainty quantified by the system dynamics model can be captured with sufficient trajectory samples, rendering a reliable controller. A practical framework is proposed and tested on a widely used Data Center model to mitigate the insufficient capacity of Deep Learning models to quantify uncertainties. The results demonstrate its superior performance during the initial training stage, achieving a 66% reduction in violations and 3.76% power savings compared to the default controller. Continuous active online learning enhances its adaptability, resulting in an 80% reduction in violations and a 6% energy saving. The thermal control strategies are thus improved by the proposed new solutions and insights. [ABSTRACT FROM AUTHOR]