Pub Date : 2025-04-23DOI: 10.1016/j.ins.2025.122224
Chengxi Jian , Junsheng Qiao , Shan He
As one of the current hot topics, feature extraction techniques have been widely studied, with the aim of selecting important and distinctive feature subsets from the original data to realize data dimensionality reduction. However, current feature extraction techniques lack the consideration of complex manifold structures in high-dimensional data, thus failing to fully exploit the information value of the data. To solve this problem, we introduce overlap functions (an emerging class of commonly used information aggregation functions with a wide range of applications) into the geodesic fuzzy rough set model and propose a new model named OKGFRS, which can effectively capture the potential manifold structures in high-dimensional data and deal with the imbalanced data. On this basis, we design a new discriminative feature extraction algorithm to improve the discriminative performance of feature extraction and to solve the problems such as poor distinguishing ability of features. After experimental verification, the algorithm demonstrates good classification performance.
{"title":"Geodesic fuzzy rough sets based on overlap functions and its applications in feature extraction","authors":"Chengxi Jian , Junsheng Qiao , Shan He","doi":"10.1016/j.ins.2025.122224","DOIUrl":"10.1016/j.ins.2025.122224","url":null,"abstract":"<div><div>As one of the current hot topics, feature extraction techniques have been widely studied, with the aim of selecting important and distinctive feature subsets from the original data to realize data dimensionality reduction. However, current feature extraction techniques lack the consideration of complex manifold structures in high-dimensional data, thus failing to fully exploit the information value of the data. To solve this problem, we introduce overlap functions (an emerging class of commonly used information aggregation functions with a wide range of applications) into the geodesic fuzzy rough set model and propose a new model named OKGFRS, which can effectively capture the potential manifold structures in high-dimensional data and deal with the imbalanced data. On this basis, we design a new discriminative feature extraction algorithm to improve the discriminative performance of feature extraction and to solve the problems such as poor distinguishing ability of features. After experimental verification, the algorithm demonstrates good classification performance.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"715 ","pages":"Article 122224"},"PeriodicalIF":8.1,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873893","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-23DOI: 10.1016/j.ins.2025.122223
Ying Zhou , Lingjing Kong , Hui Wang
The multiobjective vehicle routing problem with time windows has attracted much attention in recent decades. Until now, various metaheuristic methods have been proposed to solve the problem. However, designing effective methods is not trivial and heavily depends on experts' knowledge. As a research hotspot in recent years, a few deep reinforcement learning methods have been tried to solve the multiobjective vehicle routing problem with symmetric distance and time matrices. However, due to the complex traffic conditions, the travel distance and time between two nodes are probably asymmetric in real-world scenarios. This article introduces a multiobjective edge-based learning algorithm (MOEL) to tackle this issue. In this method, a single neural network model is established and trained to approximate the whole Pareto front of the problem. The edge features, including travel distance and time matrices, are fully learned and used to construct high-quality solutions. MOEL is compared against three state-of-the-art deep reinforcement learning methods (MODRL/D-EL, PMOCO, EMNH) and five metaheuristic methods (NSGA-II, MOEA/D, NSGA-III, MOEA/D-D, MOIA). Experimental results on the real-world instances indicate that MOEL significantly outperforms all competitors, improving IGD by up to 99.80% and HV by up to 62.84%. In addition, MOEL achieves a maximum runtime reduction of 88.65% compared to the deep reinforcement learning methods, highlighting its efficiency and effectiveness for solving the problem.
{"title":"A multiobjective edge-based learning algorithm for the vehicle routing problem with time windows","authors":"Ying Zhou , Lingjing Kong , Hui Wang","doi":"10.1016/j.ins.2025.122223","DOIUrl":"10.1016/j.ins.2025.122223","url":null,"abstract":"<div><div>The multiobjective vehicle routing problem with time windows has attracted much attention in recent decades. Until now, various metaheuristic methods have been proposed to solve the problem. However, designing effective methods is not trivial and heavily depends on experts' knowledge. As a research hotspot in recent years, a few deep reinforcement learning methods have been tried to solve the multiobjective vehicle routing problem with symmetric distance and time matrices. However, due to the complex traffic conditions, the travel distance and time between two nodes are probably asymmetric in real-world scenarios. This article introduces a multiobjective edge-based learning algorithm (MOEL) to tackle this issue. In this method, a single neural network model is established and trained to approximate the whole Pareto front of the problem. The edge features, including travel distance and time matrices, are fully learned and used to construct high-quality solutions. MOEL is compared against three state-of-the-art deep reinforcement learning methods (MODRL/D-EL, PMOCO, EMNH) and five metaheuristic methods (NSGA-II, MOEA/D, NSGA-III, MOEA/D-D, MOIA). Experimental results on the real-world instances indicate that MOEL significantly outperforms all competitors, improving IGD by up to 99.80% and HV by up to 62.84%. In addition, MOEL achieves a maximum runtime reduction of 88.65% compared to the deep reinforcement learning methods, highlighting its efficiency and effectiveness for solving the problem.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"715 ","pages":"Article 122223"},"PeriodicalIF":8.1,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143867881","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-23DOI: 10.1016/j.ins.2025.122216
Hoki Kim , Yunyoung Lee , Woojin Lee , Jaewook Lee
Although deep learning models have shown superior performance for time series classification, prior studies have recently discovered that small perturbations can fool various time series models. This vulnerability poses a serious threat that can cause malfunctions in real-world systems, such as Internet-of-Things (IoT) devices and industrial control systems. To defend these systems against adversarial time series, recent studies have proposed a detection method using time series characteristics. In this paper, however, we reveal that this detection-based defense can be easily circumvented. Through an extensive investigation into existing adversarial attacks and generated adversarial time series examples, we discover that they tend to ignore the trends in local areas and add excessive noise to the original examples. Based on the analyses, we propose a new adaptive attack, called trend-adaptive interval attack (TIA), that generates a hardly detectable adversarial time series by adopting trend-adaptive loss and gradient-based interval selection. Our experiments demonstrate that the proposed method successfully maintains the important features of the original time series and deceives diverse time series models without being detected.
{"title":"Towards undetectable adversarial attack on time series classification","authors":"Hoki Kim , Yunyoung Lee , Woojin Lee , Jaewook Lee","doi":"10.1016/j.ins.2025.122216","DOIUrl":"10.1016/j.ins.2025.122216","url":null,"abstract":"<div><div>Although deep learning models have shown superior performance for time series classification, prior studies have recently discovered that small perturbations can fool various time series models. This vulnerability poses a serious threat that can cause malfunctions in real-world systems, such as Internet-of-Things (IoT) devices and industrial control systems. To defend these systems against adversarial time series, recent studies have proposed a detection method using time series characteristics. In this paper, however, we reveal that this detection-based defense can be easily circumvented. Through an extensive investigation into existing adversarial attacks and generated adversarial time series examples, we discover that they tend to ignore the trends in local areas and add excessive noise to the original examples. Based on the analyses, we propose a new adaptive attack, called trend-adaptive interval attack (TIA), that generates a hardly detectable adversarial time series by adopting trend-adaptive loss and gradient-based interval selection. Our experiments demonstrate that the proposed method successfully maintains the important features of the original time series and deceives diverse time series models without being detected.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"715 ","pages":"Article 122216"},"PeriodicalIF":8.1,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873892","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-23DOI: 10.1016/j.ins.2025.122220
Weihua Xu, Yuzhe Li
In the realm of multi-label feature selection, the intricacy of data structures and semantics has been escalating, rendering traditional single-label feature selection methodologies inadequate for contemporary demands to meet contemporary demands. This manuscript introduces an innovative neighborhood rough set model that integrates δ-neighborhood rough sets with k-nearest neighbor techniques, facilitating a transition from single-label to multi-label learning frameworks. The study delves into the attribute dependency concept within rough set theory and proposes a novel importance function based thereon, which can effectively quantify the significance of features within multi-label decision-making contexts. Building on this theoretical foundation, we have crafted a feature selection algorithm specifically tailored for imbalanced datasets. Extensive experiments conducted on 12 datasets, coupled with comparative analyses with 10 cutting-edge methods, have substantiated the superior performance of our algorithm in managing imbalanced datasets. This research not only offers a fresh theoretical perspective but also has significant practical implications, particularly in scenarios involving imbalanced datasets with multiple labels.
{"title":"Multi-label feature selection for imbalanced data via KNN-based multi-label rough set theory","authors":"Weihua Xu, Yuzhe Li","doi":"10.1016/j.ins.2025.122220","DOIUrl":"10.1016/j.ins.2025.122220","url":null,"abstract":"<div><div>In the realm of multi-label feature selection, the intricacy of data structures and semantics has been escalating, rendering traditional single-label feature selection methodologies inadequate for contemporary demands to meet contemporary demands. This manuscript introduces an innovative neighborhood rough set model that integrates <em>δ</em>-neighborhood rough sets with <em>k</em>-nearest neighbor techniques, facilitating a transition from single-label to multi-label learning frameworks. The study delves into the attribute dependency concept within rough set theory and proposes a novel importance function based thereon, which can effectively quantify the significance of features within multi-label decision-making contexts. Building on this theoretical foundation, we have crafted a feature selection algorithm specifically tailored for imbalanced datasets. Extensive experiments conducted on 12 datasets, coupled with comparative analyses with 10 cutting-edge methods, have substantiated the superior performance of our algorithm in managing imbalanced datasets. This research not only offers a fresh theoretical perspective but also has significant practical implications, particularly in scenarios involving imbalanced datasets with multiple labels.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"715 ","pages":"Article 122220"},"PeriodicalIF":8.1,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143867882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-22DOI: 10.1016/j.ins.2025.122218
Yuzhuo Zhang , Mengzhuo Luo , Jun Cheng , Huaicheng Yan , Kaibo Shi
This paper proposes an intelligent frequency control (IFC) scheme integrating multiple strategies, which aims to address the frequency control challenges of multi-area Markov jumping power systems (MMJPSs) under load fluctuations and external disturbances. Firstly, the Markov superposition technique is employed to conduct refined modeling on the system component matrices, precisely capturing the diversity of load operating states. Secondly, within the framework of the multiplayer Stackelberg-Nash game (MSNG), the load aggregator (LA) is set as the leader and the turbines in each area are regarded as the followers. By constructing the value functions of the leader and the followers, the dynamic process of hierarchical decision-making is elaborately depicted. Meanwhile, an adaptive event-triggered mechanism (AETM) is designed to alleviate the computational and communication burdens. On this basis, by combining the integral reinforcement learning (IRL) algorithm with the neural network (NN), the Hamilton-Jacobi-Bellman (HJB) equation based on the AETM is solved to obtain the approximately optimal control law and achieve the Stackelberg-Nash equilibrium (SNE). Utilizing Lyapunov stability theory, the uniform ultimate boundedness (UUB) of the system states and the NN weight errors is rigorously proved. Finally, comparative simulation results validate the effectiveness and practicality of the proposed method.
{"title":"Learning-boosted intelligent frequency control of multi-area Markov jumping power system via multiplayer Stackelberg-Nash game","authors":"Yuzhuo Zhang , Mengzhuo Luo , Jun Cheng , Huaicheng Yan , Kaibo Shi","doi":"10.1016/j.ins.2025.122218","DOIUrl":"10.1016/j.ins.2025.122218","url":null,"abstract":"<div><div>This paper proposes an intelligent frequency control (IFC) scheme integrating multiple strategies, which aims to address the frequency control challenges of multi-area Markov jumping power systems (MMJPSs) under load fluctuations and external disturbances. Firstly, the Markov superposition technique is employed to conduct refined modeling on the system component matrices, precisely capturing the diversity of load operating states. Secondly, within the framework of the multiplayer Stackelberg-Nash game (MSNG), the load aggregator (LA) is set as the leader and the turbines in each area are regarded as the followers. By constructing the value functions of the leader and the followers, the dynamic process of hierarchical decision-making is elaborately depicted. Meanwhile, an adaptive event-triggered mechanism (AETM) is designed to alleviate the computational and communication burdens. On this basis, by combining the integral reinforcement learning (IRL) algorithm with the neural network (NN), the Hamilton-Jacobi-Bellman (HJB) equation based on the AETM is solved to obtain the approximately optimal control law and achieve the Stackelberg-Nash equilibrium (SNE). Utilizing Lyapunov stability theory, the uniform ultimate boundedness (UUB) of the system states and the NN weight errors is rigorously proved. Finally, comparative simulation results validate the effectiveness and practicality of the proposed method.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"715 ","pages":"Article 122218"},"PeriodicalIF":8.1,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143867879","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-22DOI: 10.1016/j.ins.2025.122215
Peng Wu , Li Pan
Clustering has been extensively studied in data mining and machine learning, with numerous applications across domains. In this paper, we propose the Gaussian Mixture Autoencoder (GMAE), a deep clustering method that integrates a probabilistic Autoencoder (AE) with a Gaussian Mixture Model (GMM). GMAE trains the GMM to model the latent representation distribution of the AE and further regularizes the aggregated posterior distribution by minimizing a KL divergence-based loss. To prevent degenerate solutions and enhance clustering performance, a negative mutual information loss is introduced in the model. Additionally, a package of strategies, including an initialization method, an adjusted loss function and an alternating iterative method, is designed to optimize the loss function effectively. Beyond clustering, GMAE can generate diverse, realistic samples for any target cluster, as it trains a decoder with reconstruction loss and adopts the GMM to regularize the latent representation distribution. Experiments on five cross-domain benchmarks demonstrate superior performance over state-of-the-art clustering methods.
{"title":"Deep unsupervised clustering by information maximization on Gaussian mixture autoencoders","authors":"Peng Wu , Li Pan","doi":"10.1016/j.ins.2025.122215","DOIUrl":"10.1016/j.ins.2025.122215","url":null,"abstract":"<div><div>Clustering has been extensively studied in data mining and machine learning, with numerous applications across domains. In this paper, we propose the Gaussian Mixture Autoencoder (GMAE), a deep clustering method that integrates a probabilistic Autoencoder (AE) with a Gaussian Mixture Model (GMM). GMAE trains the GMM to model the latent representation distribution of the AE and further regularizes the aggregated posterior distribution by minimizing a KL divergence-based loss. To prevent degenerate solutions and enhance clustering performance, a negative mutual information loss is introduced in the model. Additionally, a package of strategies, including an initialization method, an adjusted loss function and an alternating iterative method, is designed to optimize the loss function effectively. Beyond clustering, GMAE can generate diverse, realistic samples for any target cluster, as it trains a decoder with reconstruction loss and adopts the GMM to regularize the latent representation distribution. Experiments on five cross-domain benchmarks demonstrate superior performance over state-of-the-art clustering methods.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"714 ","pages":"Article 122215"},"PeriodicalIF":8.1,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143860002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Best-worst method (BWM) has been extended in various uncertain scenarios owing to fewer comparisons and better reliability. This article utilizes hesitant multiplicative (HM) sets (HMSs) to express reference comparisons (RCs) and develops a novel HM BWM (HMBWM). We first define the multiplicative consistency for HM preference relation (HMPR). A fast and effective approach is designed to derive the priority weights (PWs) from an HMPR. To extend BW into HMPR, the score value of each criterion is computed to identify the best and worst criteria. Then, the PWs are acquired through constructing a 0–1 mixed goal programming model based on the HM RCs (HMRCs). The consistency ratio is given to judge the multiplicative consistency of HMRCs. An approach is proposed to enhance the consistency when the HMRCs are unacceptably consistent. Thereby, a novel HMBWM is proposed. On basis of HMBWM, this article further develops a novel method for group decision making (GDM) with HMPRs. The decision makers’ weights are determined by consistency ratio and the group PWs of alternatives are obtained by minimum relative entropy model. Four examples show that HMBWM possesses higher consistency and the proposed GDM method has stronger distinguishing ability, less computation workload and fewer modifications of elements.
{"title":"Hesitant multiplicative best and worst method for multi-criteria group decision making","authors":"Shu-Ping Wan , Xi-Nuo Chen , Jiu-Ying Dong , Yu Gao","doi":"10.1016/j.ins.2025.122214","DOIUrl":"10.1016/j.ins.2025.122214","url":null,"abstract":"<div><div>Best-worst method (BWM) has been extended in various uncertain scenarios owing to fewer comparisons and better reliability. This article utilizes hesitant multiplicative (HM) sets (HMSs) to express reference comparisons (RCs) and develops a novel HM BWM (HMBWM). We first define the multiplicative consistency for HM preference relation (HMPR). A fast and effective approach is designed to derive the priority weights (PWs) from an HMPR. To extend BW into HMPR, the score value of each criterion is computed to identify the best and worst criteria. Then, the PWs are acquired through constructing a 0–1 mixed goal programming model based on the HM RCs (HMRCs). The consistency ratio is given to judge the multiplicative consistency of HMRCs. An approach is proposed to enhance the consistency when the HMRCs are unacceptably consistent. Thereby, a novel HMBWM is proposed. On basis of HMBWM, this article further develops a novel method for group decision making (GDM) with HMPRs. The decision makers’ weights are determined by consistency ratio and the group PWs of alternatives are obtained by minimum relative entropy model. Four examples show that HMBWM possesses higher consistency and the proposed GDM method has stronger distinguishing ability, less computation workload and fewer modifications of elements.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"715 ","pages":"Article 122214"},"PeriodicalIF":8.1,"publicationDate":"2025-04-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143867883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-20DOI: 10.1016/j.ins.2025.122221
Renjie Lv , Wenwen Chang , Guanghui Yan , Muhammad Tariq Sadiq , Wenchao Nie , Lei Zheng
Deep learning has shown promising results in motor imagery brain-computer interfaces. However, most existing methods fail to account for the topological relationships between electrodes and the nonlinear features of electroencephalogram (EEG) signals. To address this, we propose a model combining Gramian Angular Fields (GAF) and Phase-Locking Value (PLV) with a parallel convolutional neural network (CNN). GAF captures time-domain nonlinear features, while PLV represents spatial features based on electrode topology. Comparative experiments between the end-to-end parallel CNN model and the model with spatiotemporal feature representation demonstrate that considering both time-domain correlations and electrode topology significantly enhances model performance. Furthermore, when separately evaluating the temporal and spatial features of EEG signals, the results confirm that jointly considering spatiotemporal features leads to a substantial improvement. On the Physionet dataset, our model achieves an accuracy of 99.73% in binary classification tasks and 83.37% in four-class classification tasks, showing improvement over the comparison algorithms used in the paper.
{"title":"Enhanced classification of motor imagery EEG signals using spatio-temporal representations","authors":"Renjie Lv , Wenwen Chang , Guanghui Yan , Muhammad Tariq Sadiq , Wenchao Nie , Lei Zheng","doi":"10.1016/j.ins.2025.122221","DOIUrl":"10.1016/j.ins.2025.122221","url":null,"abstract":"<div><div>Deep learning has shown promising results in motor imagery brain-computer interfaces. However, most existing methods fail to account for the topological relationships between electrodes and the nonlinear features of electroencephalogram (EEG) signals. To address this, we propose a model combining Gramian Angular Fields (GAF) and Phase-Locking Value (PLV) with a parallel convolutional neural network (CNN). GAF captures time-domain nonlinear features, while PLV represents spatial features based on electrode topology. Comparative experiments between the end-to-end parallel CNN model and the model with spatiotemporal feature representation demonstrate that considering both time-domain correlations and electrode topology significantly enhances model performance. Furthermore, when separately evaluating the temporal and spatial features of EEG signals, the results confirm that jointly considering spatiotemporal features leads to a substantial improvement. On the Physionet dataset, our model achieves an accuracy of 99.73% in binary classification tasks and 83.37% in four-class classification tasks, showing improvement over the comparison algorithms used in the paper.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"714 ","pages":"Article 122221"},"PeriodicalIF":8.1,"publicationDate":"2025-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143860003","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-18DOI: 10.1016/j.ins.2025.122212
Shizhan Lu , Zeshui Xu , Zhu Fu , Longsheng Cheng , Tongbin Yang
Hesitant fuzzy sets find extensive application in specific scenarios involving uncertainty and hesitation. In the context of set theory, the concept of inclusion relationship holds significant importance as a fundamental definition. Consequently, as a type of sets, hesitant fuzzy sets necessitate a clear and explicit definition of the inclusion relationship. Based on the discrete form of hesitant fuzzy membership degrees, this study proposes multiple types of inclusion relationships for hesitant fuzzy sets. Subsequently, this paper introduces foundational propositions related to hesitant fuzzy sets, as well as propositions concerning families of hesitant fuzzy sets. Furthermore, this research presents foundational propositions regarding parameter reduction of hesitant fuzzy information systems. An example and an algorithm are provided to demonstrate the parameter reduction processes. Lastly, a multi-strength intelligent classifier is proposed for diagnosing the health states of complex systems.
{"title":"Foundational theories of hesitant fuzzy sets and hesitant fuzzy information systems and their applications for multi-strength intelligent classifiers","authors":"Shizhan Lu , Zeshui Xu , Zhu Fu , Longsheng Cheng , Tongbin Yang","doi":"10.1016/j.ins.2025.122212","DOIUrl":"10.1016/j.ins.2025.122212","url":null,"abstract":"<div><div>Hesitant fuzzy sets find extensive application in specific scenarios involving uncertainty and hesitation. In the context of set theory, the concept of inclusion relationship holds significant importance as a fundamental definition. Consequently, as a type of sets, hesitant fuzzy sets necessitate a clear and explicit definition of the inclusion relationship. Based on the discrete form of hesitant fuzzy membership degrees, this study proposes multiple types of inclusion relationships for hesitant fuzzy sets. Subsequently, this paper introduces foundational propositions related to hesitant fuzzy sets, as well as propositions concerning families of hesitant fuzzy sets. Furthermore, this research presents foundational propositions regarding parameter reduction of hesitant fuzzy information systems. An example and an algorithm are provided to demonstrate the parameter reduction processes. Lastly, a multi-strength intelligent classifier is proposed for diagnosing the health states of complex systems.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"714 ","pages":"Article 122212"},"PeriodicalIF":8.1,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143859932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-04-18DOI: 10.1016/j.ins.2025.122208
Yu Wang, Wei Chen
Hybrid models for influential node identification have gained attention for integrating local, semilocal, and global information. These models regularly use location to account for global information, however, seldom take further consideration of the nonlinear feedback contribution and non-redundant bridging ability. In information dissemination, the nonlinear feedback contribution can enhance information reliability through diverse feedback validation, and the non-redundant bridging ability can foster broad access and allocation of heterogeneous information by connecting multiple independent nodes. Additionally, most hybrid models overlook centrality of receivers in the second-order dissemination, which can affect the scope and speed of information dissemination. Moreover, identification of bottom ranked nodes is often ignored, despite that the optimization of these nodes can enhance network efficiency. This work presents a novel hybrid model that incorporates hybrid centrality of receivers in the second-order dissemination. Specifically, hybrid centrality is formulated by simultaneously considering the location, nonlinear feedback contribution, and non-redundant bridging ability. Receivers in the second-order dissemination are then collected, and node importance is determined based on their hybrid centrality. Extensive experiments on 9 real-world and 3 synthetic networks show that our model outperforms state-of-the-art models in node ranking, top-k and bottom-k nodes identification. Robustness is also validated via varying infection probabilities.
{"title":"Identifying influential nodes based on hybrid centrality of receivers in the second-order dissemination","authors":"Yu Wang, Wei Chen","doi":"10.1016/j.ins.2025.122208","DOIUrl":"10.1016/j.ins.2025.122208","url":null,"abstract":"<div><div>Hybrid models for influential node identification have gained attention for integrating local, semilocal, and global information. These models regularly use location to account for global information, however, seldom take further consideration of the nonlinear feedback contribution and non-redundant bridging ability. In information dissemination, the nonlinear feedback contribution can enhance information reliability through diverse feedback validation, and the non-redundant bridging ability can foster broad access and allocation of heterogeneous information by connecting multiple independent nodes. Additionally, most hybrid models overlook centrality of receivers in the second-order dissemination, which can affect the scope and speed of information dissemination. Moreover, identification of bottom ranked nodes is often ignored, despite that the optimization of these nodes can enhance network efficiency. This work presents a novel hybrid model that incorporates hybrid centrality of receivers in the second-order dissemination. Specifically, hybrid centrality is formulated by simultaneously considering the location, nonlinear feedback contribution, and non-redundant bridging ability. Receivers in the second-order dissemination are then collected, and node importance is determined based on their hybrid centrality. Extensive experiments on 9 real-world and 3 synthetic networks show that our model outperforms state-of-the-art models in node ranking, top-<em>k</em> and bottom-<em>k</em> nodes identification. Robustness is also validated via varying infection probabilities.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"714 ","pages":"Article 122208"},"PeriodicalIF":8.1,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143864195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号