Information Sciences最新文献_第10页

An adaptive helper and equivalent objective evolution strategy for constrained optimization 约束优化的自适应辅助和等效目标演化策略

IF 8.1 1区计算机科学 0 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Sciences

Pub Date : 2024-10-10 DOI: 10.1016/j.ins.2024.121536

Tao Xu , Hongyang Chen , Jun He

The matrix adaptation evolution strategy is a simplified covariance matrix adaptation evolution strategy with reduced computational cost. Using it as a search engine, several algorithms have been recently proposed for constrained optimization and have shown excellent performance. However, these algorithms require the simultaneous application of multiple techniques to handle constraints, and also require gradient information. This makes them inappropriate for handling non-differentiable functions. This paper proposes a new matrix adaption evolutionary strategy for constrained optimization using helper and equivalent objectives. The method optimizes two objectives but with no need for special handling of infeasible solutions and without gradient information. A new mechanism is designed to adaptively adjust the weights of the two objectives according to the convergence rate. The efficacy of the proposed algorithm is evaluated using computational experiments on the IEEE CEC 2017 Constrained Optimization Competition benchmarks. Experimental results demonstrate that it outperforms current state-of-the-art evolutionary algorithms. Furthermore, this paper provides sufficient conditions for the convergence of helper and equivalent objective evolutionary algorithms and proves that using helper objectives can reduce the likelihood of premature convergence.

矩阵适应演化策略是一种简化的协方差矩阵适应演化策略，计算成本较低。利用它作为搜索引擎，最近提出了几种用于约束优化的算法，并显示出卓越的性能。然而，这些算法需要同时应用多种技术来处理约束条件，还需要梯度信息。这使得它们不适合处理无差异函数。本文提出了一种新的矩阵自适应进化策略，用于使用辅助目标和等效目标的约束优化。该方法可优化两个目标，但无需特殊处理不可行解，也不需要梯度信息。该方法设计了一种新机制，可根据收敛速度自适应地调整两个目标的权重。通过在 IEEE CEC 2017 受限优化竞赛基准上进行计算实验，评估了所提算法的功效。实验结果表明，该算法优于目前最先进的进化算法。此外，本文还提供了辅助目标和等效目标进化算法收敛的充分条件，并证明使用辅助目标可以降低过早收敛的可能性。

{"title":"An adaptive helper and equivalent objective evolution strategy for constrained optimization","authors":"Tao Xu , Hongyang Chen , Jun He","doi":"10.1016/j.ins.2024.121536","DOIUrl":"10.1016/j.ins.2024.121536","url":null,"abstract":"<div><div>The matrix adaptation evolution strategy is a simplified covariance matrix adaptation evolution strategy with reduced computational cost. Using it as a search engine, several algorithms have been recently proposed for constrained optimization and have shown excellent performance. However, these algorithms require the simultaneous application of multiple techniques to handle constraints, and also require gradient information. This makes them inappropriate for handling non-differentiable functions. This paper proposes a new matrix adaption evolutionary strategy for constrained optimization using helper and equivalent objectives. The method optimizes two objectives but with no need for special handling of infeasible solutions and without gradient information. A new mechanism is designed to adaptively adjust the weights of the two objectives according to the convergence rate. The efficacy of the proposed algorithm is evaluated using computational experiments on the IEEE CEC 2017 Constrained Optimization Competition benchmarks. Experimental results demonstrate that it outperforms current state-of-the-art evolutionary algorithms. Furthermore, this paper provides sufficient conditions for the convergence of helper and equivalent objective evolutionary algorithms and proves that using helper objectives can reduce the likelihood of premature convergence.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"690 ","pages":"Article 121536"},"PeriodicalIF":8.1,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142434455","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}

引用次数: 0

Hierarchical graph contrastive learning framework based on quantum neural networks for sentiment analysis 基于量子神经网络的层次图对比学习框架用于情感分析

IF 8.1 1区计算机科学 0 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Sciences

Pub Date : 2024-10-10 DOI: 10.1016/j.ins.2024.121543

Keliang Jia, Fanxu Meng, Jing Liang

Existing multi-modal sentiment analysis (MSA) methods typically achieve interaction by connecting different layers or designing special structures, but rarely consider the synergistic effects among data. Moreover, most sentiment analysis research tends to focus solely on single sentiment polarity analysis, without considering the intensity and directional attributes of emotions. Addressing these issues, we propose a framework called Hierarchical Graph Contrastive Learning based on Quantum Neural Network (HGCL-QNN) to remedy these shortcomings. Specifically, a graph structure is established within and between modalities. In the quantum fuzzy neural network module, fuzzy quantum encoding is implemented by using complex-valued, then quantum superposition and entanglement are utilized to consider the intensity and directional attributes of emotions while analyzing emotional polarity. In the quantum multi-modal fusion neural network module, methods such as amplitude encoding and quantum entanglement are employed to further integrate information from different modalities, thereby enhancing the model's power to express emotional information. To enhance the model's understanding of fine-grained and global features, and to better align and integrate features from different modalities, hierarchical graph contrastive learning is employed on different levels. The experimental results demonstrate that HGCL-QNN outperforms the existing baseline methods on MOSI and MOSEI datasets, achieving significant efficacy improvements.

现有的多模态情感分析（MSA）方法通常通过连接不同层或设计特殊结构来实现交互，但很少考虑数据之间的协同效应。此外，大多数情感分析研究往往只关注单一的情感极性分析，而不考虑情感的强度和方向属性。针对这些问题，我们提出了一个名为 "基于量子神经网络的层次图对比学习"（HGCL-QNN）的框架来弥补这些不足。具体来说，我们在模态内部和模态之间建立了图结构。在量子模糊神经网络模块中，利用复值实现模糊量子编码，然后利用量子叠加和纠缠来考虑情绪的强度和方向属性，同时分析情绪极性。在量子多模态融合神经网络模块中，利用振幅编码和量子纠缠等方法进一步整合不同模态的信息，从而增强模型表达情绪信息的能力。为了增强模型对细粒度和全局特征的理解，并更好地排列和整合来自不同模态的特征，在不同层次上采用了分层图对比学习。实验结果表明，HGCL-QNN 在 MOSI 和 MOSEI 数据集上的表现优于现有的基线方法，实现了显著的功效提升。

{"title":"Hierarchical graph contrastive learning framework based on quantum neural networks for sentiment analysis","authors":"Keliang Jia, Fanxu Meng, Jing Liang","doi":"10.1016/j.ins.2024.121543","DOIUrl":"10.1016/j.ins.2024.121543","url":null,"abstract":"<div><div>Existing multi-modal sentiment analysis (MSA) methods typically achieve interaction by connecting different layers or designing special structures, but rarely consider the synergistic effects among data. Moreover, most sentiment analysis research tends to focus solely on single sentiment polarity analysis, without considering the intensity and directional attributes of emotions. Addressing these issues, we propose a framework called Hierarchical Graph Contrastive Learning based on Quantum Neural Network (HGCL-QNN) to remedy these shortcomings. Specifically, a graph structure is established within and between modalities. In the quantum fuzzy neural network module, fuzzy quantum encoding is implemented by using complex-valued, then quantum superposition and entanglement are utilized to consider the intensity and directional attributes of emotions while analyzing emotional polarity. In the quantum multi-modal fusion neural network module, methods such as amplitude encoding and quantum entanglement are employed to further integrate information from different modalities, thereby enhancing the model's power to express emotional information. To enhance the model's understanding of fine-grained and global features, and to better align and integrate features from different modalities, hierarchical graph contrastive learning is employed on different levels. The experimental results demonstrate that HGCL-QNN outperforms the existing baseline methods on MOSI and MOSEI datasets, achieving significant efficacy improvements.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"690 ","pages":"Article 121543"},"PeriodicalIF":8.1,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142434458","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}

引用次数: 0

Event-based integral sliding mode control for leader-follower consensus with perturbed agent dynamics 基于事件的积分滑模控制，实现具有扰动代理动态的领导者-追随者共识

IF 8.1 1区计算机科学 0 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Sciences

Pub Date : 2024-10-10 DOI: 10.1016/j.ins.2024.121535

Tara Swaraj , Krishanu Nath , Manas Kumar Bera , Rajiv Kumar Mishra , Sudipta Chakraborty

This manuscript proposes the design of an event-triggered integral sliding mode (ET-ISM)-based distributed controller for multi-agent systems (MASs) to achieve consensus between the leader and followers. Specifically, we consider the MASs, having multi-input-multi-output (MIMO) linear dynamics with unknown bounded perturbations. The information flow in the network is modelled by the directed graph. The design of the continuous-time integral sliding mode (ISM) controller is discussed first, followed by an ET-ISM strategy using a novel triggering rule to avoid periodic communication between the leader and agents. Unlike the earlier works, our proposed method uses an event function entirely devoid of non-differentiable terms to define the triggering condition. The stability of robust closed-loop dynamics of the network is guaranteed using Lyapunov stability theory, and the existence of practical sliding motion (PSM) is established by calculating the band for PSM as well as the band of convergence of the disagreement vector. The Zeno-free behaviour of the closed-loop system is also ensured to show that the sampling is well-behaved means the triggering approach generates a finite number of events. Finally, we take up a numerical example to discuss the design process of the proposed controller and present the simulation results along with a detailed analysis.

本文提出了一种基于事件触发积分滑动模式（ET-ISM）的多代理系统（MAS）分布式控制器的设计方案，以实现领导者与追随者之间的共识。具体来说，我们考虑的是具有未知有界扰动的多输入多输出（MIMO）线性动力学的 MAS。网络中的信息流由有向图模拟。我们首先讨论了连续时间积分滑动模式（ISM）控制器的设计，然后讨论了使用新型触发规则的 ET-ISM 策略，以避免领导者与代理之间的周期性通信。与之前的研究不同，我们提出的方法使用完全不含无差别项的事件函数来定义触发条件。利用李雅普诺夫稳定性理论保证了网络鲁棒闭环动力学的稳定性，并通过计算实际滑动运动（PSM）的频带以及分歧向量的收敛频带确定了实际滑动运动（PSM）的存在。我们还确保了闭环系统的无 Zeno 行为，表明采样行为良好意味着触发方法会产生有限数量的事件。最后，我们以一个数值示例讨论了所提控制器的设计过程，并给出了仿真结果和详细分析。

{"title":"Event-based integral sliding mode control for leader-follower consensus with perturbed agent dynamics","authors":"Tara Swaraj , Krishanu Nath , Manas Kumar Bera , Rajiv Kumar Mishra , Sudipta Chakraborty","doi":"10.1016/j.ins.2024.121535","DOIUrl":"10.1016/j.ins.2024.121535","url":null,"abstract":"<div><div>This manuscript proposes the design of an event-triggered integral sliding mode (ET-ISM)-based distributed controller for multi-agent systems (MASs) to achieve consensus between the leader and followers. Specifically, we consider the MASs, having multi-input-multi-output (MIMO) linear dynamics with unknown bounded perturbations. The information flow in the network is modelled by the directed graph. The design of the continuous-time integral sliding mode (ISM) controller is discussed first, followed by an ET-ISM strategy using a novel triggering rule to avoid periodic communication between the leader and agents. Unlike the earlier works, our proposed method uses an event function entirely devoid of non-differentiable terms to define the triggering condition. The stability of robust closed-loop dynamics of the network is guaranteed using Lyapunov stability theory, and the existence of practical sliding motion (PSM) is established by calculating the band for PSM as well as the band of convergence of the disagreement vector. The Zeno-free behaviour of the closed-loop system is also ensured to show that the sampling is well-behaved means the triggering approach generates a finite number of events. Finally, we take up a numerical example to discuss the design process of the proposed controller and present the simulation results along with a detailed analysis.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"690 ","pages":"Article 121535"},"PeriodicalIF":8.1,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142434453","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}

引用次数: 0

Revocable certificate-based broadcast signcryption scheme for edge-enabled IIoT 基于可撤销证书的边缘启用 IIoT 广播签名加密方案

IF 8.1 1区计算机科学 0 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Sciences

Pub Date : 2024-10-10 DOI: 10.1016/j.ins.2024.121540

Yan Gao , Lunzhi Deng , Shuai Feng , Huan Liu , Binhan Li , Na Wang

In edge computing-enabled Industrial Internet of Things (edge-enabled IIoT), edge computing facilitates data processing with reduced latency, enhanced reliability, and optimized real-time resource use. With the rapid increase in connected devices, ensuring secure data sharing among them is crucial. Broadcast signcryption technology is an excellent choice for achieving data confidentiality and authentication while enhancing operational efficiency. With devices potentially expiring, aging, or suffering damage, the prompt revocation of their decryption privileges is essential. Currently, there is a lack of research that simultaneously integrates broadcast signcryption with a revocation mechanism. In this paper, we propose a revocable certificate-based broadcast signcryption (RCB-BSC) scheme tailored for edge-enabled IIoT. In our work, base station generates a signcrypted ciphertext for multiple edge computing nodes (ECNs), and sends it to edge service (ES). The ES revokes the access of illegitimate ECNs and generates a new ciphertext for those ECNs with non-revoked decryption privileges, allowing them to decrypt the message using their private keys. Under the random oracle model (ROM), our scheme achieves plaintext confidentiality, ensures anonymity of receivers, and authenticates the legitimacy of the broadcaster. Moreover, the performance analysis shows our scheme excels in computation and communication efficiency, making it ideal for edge-enabled IIoT.

在支持边缘计算的工业物联网（edge-enabled IIoT）中，边缘计算通过减少延迟、提高可靠性和优化实时资源使用来促进数据处理。随着联网设备的迅速增加，确保这些设备之间的安全数据共享至关重要。广播式签名加密技术是在提高运行效率的同时实现数据保密性和身份验证的绝佳选择。由于设备有可能过期、老化或损坏，因此及时撤销其解密权限至关重要。目前，还缺乏将广播式签名加密与撤销机制同时集成的研究。在本文中，我们提出了一种基于证书的可撤销广播加密（RCB-BSC）方案，该方案专为边缘启用的物联网量身定制。在我们的工作中，基站为多个边缘计算节点（ECN）生成签名加密密文，并将其发送给边缘服务（ES）。ES 会撤销非法 ECN 的访问权限，并为那些拥有未撤销解密权限的 ECN 生成新的密文，允许它们使用自己的私钥解密信息。在随机甲骨文模型（ROM）下，我们的方案实现了明文保密性，确保了接收者的匿名性，并验证了广播者的合法性。此外，性能分析表明，我们的方案在计算和通信效率方面表现出色，因此非常适合支持边缘的物联网。

{"title":"Revocable certificate-based broadcast signcryption scheme for edge-enabled IIoT","authors":"Yan Gao , Lunzhi Deng , Shuai Feng , Huan Liu , Binhan Li , Na Wang","doi":"10.1016/j.ins.2024.121540","DOIUrl":"10.1016/j.ins.2024.121540","url":null,"abstract":"<div><div>In edge computing-enabled Industrial Internet of Things (edge-enabled IIoT), edge computing facilitates data processing with reduced latency, enhanced reliability, and optimized real-time resource use. With the rapid increase in connected devices, ensuring secure data sharing among them is crucial. Broadcast signcryption technology is an excellent choice for achieving data confidentiality and authentication while enhancing operational efficiency. With devices potentially expiring, aging, or suffering damage, the prompt revocation of their decryption privileges is essential. Currently, there is a lack of research that simultaneously integrates broadcast signcryption with a revocation mechanism. In this paper, we propose a revocable certificate-based broadcast signcryption (RCB-BSC) scheme tailored for edge-enabled IIoT. In our work, base station generates a signcrypted ciphertext for multiple edge computing nodes (ECNs), and sends it to edge service (ES). The ES revokes the access of illegitimate ECNs and generates a new ciphertext for those ECNs with non-revoked decryption privileges, allowing them to decrypt the message using their private keys. Under the random oracle model (ROM), our scheme achieves plaintext confidentiality, ensures anonymity of receivers, and authenticates the legitimacy of the broadcaster. Moreover, the performance analysis shows our scheme excels in computation and communication efficiency, making it ideal for edge-enabled IIoT.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"690 ","pages":"Article 121540"},"PeriodicalIF":8.1,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142434397","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}

引用次数: 0

Fuzzy serial-parallel stochastic configuration networks based on nonconvex dynamic membership function optimization 基于非凸动态成员函数优化的模糊串并联随机配置网络

IF 8.1 1区计算机科学 0 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Sciences

Pub Date : 2024-10-10 DOI: 10.1016/j.ins.2024.121501

Jinghui Qiao , Jiayu Qiao , Peng Gao , Zhe Bai , Ningkang Xiong

A fuzzy series–parallel stochastic configuration networks (F-SPSCN) is proposed based on the application of nonconvex optimization in fuzzy systems. Firstly, the kernel density estimation method is used to fit the distribution of original input data to generate dynamic nonconvex membership functions, which enhances the fuzzy system ability to handle uncertain industrial data. Then the parameters of the nonconvex membership functions are optimized based on Majorization-Minimization algorithm and Generalized Projective Gradient Descent algorithm. The optimized membership matrices and fuzzy outputs are used as inputs of the serial-parallel stochastic configuration networks to improve the overall prediction accuracy of the model. Finally, the prediction accuracy of the F-SPSCN model has been verified by performing prediction experiments with two different functions and four benchmark datasets. The F-SPSCN model demonstrates superior performance compared to other models in predicting the magnetic separation recovery ratio (MSRR) of hydrogen-based mineral phase transformation (HMPT) process for refractory iron ore.

基于非凸优化在模糊系统中的应用，提出了一种模糊串并联随机配置网络（F-SPSCN）。首先，利用核密度估计方法拟合原始输入数据的分布，生成动态非凸成员函数，从而提高模糊系统处理不确定工业数据的能力。然后，基于 Majorization-Minimization 算法和广义投影梯度下降算法对非凸会员函数的参数进行优化。优化后的成员矩阵和模糊输出被用作串并联随机配置网络的输入，以提高模型的整体预测精度。最后，通过对两种不同函数和四个基准数据集进行预测实验，验证了 F-SPSCN 模型的预测准确性。与其他模型相比，F-SPSCN 模型在预测难选铁矿石氢基矿物相转化（HMPT）过程的磁选回收率（MSRR）方面表现出更优越的性能。

{"title":"Fuzzy serial-parallel stochastic configuration networks based on nonconvex dynamic membership function optimization","authors":"Jinghui Qiao , Jiayu Qiao , Peng Gao , Zhe Bai , Ningkang Xiong","doi":"10.1016/j.ins.2024.121501","DOIUrl":"10.1016/j.ins.2024.121501","url":null,"abstract":"<div><div>A fuzzy series–parallel stochastic configuration networks (F-SPSCN) is proposed based on the application of nonconvex optimization in fuzzy systems. Firstly, the kernel density estimation method is used to fit the distribution of original input data to generate dynamic nonconvex membership functions, which enhances the fuzzy system ability to handle uncertain industrial data. Then the parameters of the nonconvex membership functions are optimized based on Majorization-Minimization algorithm and Generalized Projective Gradient Descent algorithm. The optimized membership matrices and fuzzy outputs are used as inputs of the serial-parallel stochastic configuration networks to improve the overall prediction accuracy of the model. Finally, the prediction accuracy of the F-SPSCN model has been verified by performing prediction experiments with two different functions and four benchmark datasets. The F-SPSCN model demonstrates superior performance compared to other models in predicting the magnetic separation recovery ratio (MSRR) of hydrogen-based mineral phase transformation (HMPT) process for refractory iron ore.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"690 ","pages":"Article 121501"},"PeriodicalIF":8.1,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142530610","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}

引用次数: 0

A data anonymization methodology for security operations centers: Balancing data protection and security in industrial systems 安全运营中心的数据匿名化方法：平衡工业系统中的数据保护和安全

IF 8.1 1区计算机科学 0 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Sciences

Pub Date : 2024-10-10 DOI: 10.1016/j.ins.2024.121534

Giacomo Longo , Francesco Lupia , Alessio Merlo , Francesco Pagano , Enrico Russo

In an era where industrial Security Operations Centers (SOCs) are paramount to enabling cybersecurity, they can unintentionally become enablers of intellectual property theft through the data they analyze and retain. The above issue requires finding solutions to strike a balance between data protection and security. This paper proposes a real-time data anonymization framework designed to operate directly within network devices. Using an extensive case study, our approach demonstrates how valuable intellectual property associated with industrial processes can be protected without compromising the effectiveness of behavioral anomaly detection systems. The methodology is designed to be nonintrusive, reversible, and seamlessly portable on existing security solutions. We evaluated these properties through comprehensive experimental testing, which showed both the method's effectiveness in securing intellectual property and its suitability for continuous real-time operation.

在工业安全运营中心（SOC）对实现网络安全至关重要的时代，它们可能会通过分析和保留数据，无意中成为知识产权盗窃的帮凶。要解决上述问题，就必须找到解决方案，在数据保护和安全之间取得平衡。本文提出了一种实时数据匿名化框架，旨在直接在网络设备内运行。通过广泛的案例研究，我们的方法展示了如何在不影响行为异常检测系统有效性的情况下保护与工业流程相关的宝贵知识产权。我们设计的方法具有非侵入性、可逆性和可在现有安全解决方案上无缝移植的特点。我们通过全面的实验测试对这些特性进行了评估，结果表明该方法不仅能有效保护知识产权，而且适合连续实时运行。

引用次数: 0

Non-blocking functional dependency discovery from data streams 从数据流中发现非阻塞功能依赖关系

IF 8.1 1区计算机科学 0 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Sciences

Pub Date : 2024-10-10 DOI: 10.1016/j.ins.2024.121531

Loredana Caruccio, Stefano Cirillo, Vincenzo Deufemia, Giuseppe Polese

With the proliferation of sensors and IoT technologies, there is an increasing need to analyze information from data streams that they produce dynamically. However, the volume and velocity of this data require algorithms that mine knowledge as data are read from streams. The capability of dynamically extracting functional dependencies (fds) from data streams would not only permit to assess and improve the quality of data, but also provide knowledge on the evolution of data correlations within streams, allowing to understand the relevance that each feature has in predicting unknown features. In this paper, we propose a new discovery algorithm, namely COD3, which allows to continuous discovery fds holding on a data stream, as the data are read from it. COD3 represents the first proposal to use a non-blocking architectural model for discovering fds from data streams. Furthermore, we present novel data structures and a validation method to handle dynamic discovery and reduce data load inbound streams. Experimental evaluations demonstrate its effectiveness on both adapted real-world datasets and real data streams, such as those from air quality sensors. Moreover, by integrating COD3 with Bleach, a well-known fd-based data stream cleansing framework, we demonstrate its effectiveness in a real-world use case.

随着传感器和物联网技术的普及，人们越来越需要从它们动态产生的数据流中分析信息。然而，这些数据的数量和速度要求算法能够在从数据流中读取数据的同时挖掘知识。从数据流中动态提取功能依赖性（fds）的能力不仅能评估和提高数据质量，还能提供数据流内数据相关性演变的知识，从而了解每个特征在预测未知特征时的相关性。在本文中，我们提出了一种新的发现算法，即 COD3，它可以在读取数据的过程中连续发现数据流中的 fds。COD3 是首个使用非阻塞架构模型从数据流中发现 fds 的提案。此外，我们还提出了新颖的数据结构和验证方法，以处理动态发现并减少入站数据流的数据负载。实验评估证明了它在适应真实世界数据集和真实数据流（如来自空气质量传感器的数据流）上的有效性。此外，通过将 COD3 与著名的基于 fd 的数据流清理框架 Bleach 相集成，我们证明了它在实际应用案例中的有效性。

{"title":"Non-blocking functional dependency discovery from data streams","authors":"Loredana Caruccio, Stefano Cirillo, Vincenzo Deufemia, Giuseppe Polese","doi":"10.1016/j.ins.2024.121531","DOIUrl":"10.1016/j.ins.2024.121531","url":null,"abstract":"<div><div>With the proliferation of sensors and IoT technologies, there is an increasing need to analyze information from data streams that they produce dynamically. However, the volume and velocity of this data require algorithms that mine knowledge as data are read from streams. The capability of dynamically extracting functional dependencies (<span>fd</span>s) from data streams would not only permit to assess and improve the quality of data, but also provide knowledge on the evolution of data correlations within streams, allowing to understand the relevance that each feature has in predicting unknown features. In this paper, we propose a new discovery algorithm, namely COD3, which allows to continuous discovery <span>fd</span>s holding on a data stream, as the data are read from it. COD3 represents the first proposal to use a non-blocking architectural model for discovering <span>fd</span>s from data streams. Furthermore, we present novel data structures and a validation method to handle dynamic discovery and reduce data load inbound streams. Experimental evaluations demonstrate its effectiveness on both adapted real-world datasets and real data streams, such as those from air quality sensors. Moreover, by integrating <span>COD</span>3 with Bleach, a well-known <span>fd</span>-based data stream cleansing framework, we demonstrate its effectiveness in a real-world use case.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"690 ","pages":"Article 121531"},"PeriodicalIF":8.1,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142434457","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Adaptive sequential three-way decisions for dynamic time warping 动态时间扭曲的自适应顺序三向决策

IF 8.1 1区计算机科学 0 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Sciences

Pub Date : 2024-10-10 DOI: 10.1016/j.ins.2024.121541

Jianfeng Xu , Ruihua Wang , Yuanjian Zhang , Weiping Ding

Dynamic time warping (DTW) algorithm is widely used in diversified applications due to its excellent anti-deformation and anti-interference in measuring time-series based similarity. However, the high time complexity of DTW restrains the applicability of real-time case. The existing DTW acceleration studies suffer from a loss of accuracy. How to accelerate computation while maintaining satisfying computational accuracy remains challenging. Motivated by sequential three-way decisions, this paper develops a novel model with adaptive sequential three-way decisions for dynamic time warping (AS3-DTW). Firstly, we systematically summarize distance differences under the context of adjacent tripartite search spaces for DTW, and propose five patterns of granularity adjustments of the search spaces. Furthermore, we present the corresponding calculation method of DTW adjacent tripartite search spaces distances difference. Finally, we construct a novel dynamism on adaptively adjusting time warping by combining sequence-based multi-granularity with sequential three-way decisions. Experimental results show that AS3-DTW effectively achieves promising trade-off between computational speed and accuracy on multiple UCR datasets when compared with the state-of-the-art algorithms.

动态时间扭曲（DTW）算法在测量基于时间序列的相似性时具有出色的抗变形和抗干扰能力，因此被广泛应用于多种领域。然而，DTW 的高时间复杂性限制了其在实时情况下的适用性。现有的 DTW 加速研究都存在精度损失的问题。如何在加速计算的同时保持令人满意的计算精度仍是一个挑战。受顺序三向决策的启发，本文为动态时间扭曲（AS3-DTW）建立了一个具有自适应顺序三向决策的新模型。首先，我们系统地总结了 DTW 相邻三方搜索空间下的距离差异，并提出了五种搜索空间粒度调整模式。此外，我们还提出了 DTW 相邻三方搜索空间距离差的相应计算方法。最后，我们通过将基于序列的多粒度与序列三向决策相结合，构建了一种自适应调整时间扭曲的新动力。实验结果表明，与最先进的算法相比，AS3-DTW 在多个 UCR 数据集上有效地实现了计算速度和精确度之间的折衷。

{"title":"Adaptive sequential three-way decisions for dynamic time warping","authors":"Jianfeng Xu , Ruihua Wang , Yuanjian Zhang , Weiping Ding","doi":"10.1016/j.ins.2024.121541","DOIUrl":"10.1016/j.ins.2024.121541","url":null,"abstract":"<div><div>Dynamic time warping (DTW) algorithm is widely used in diversified applications due to its excellent anti-deformation and anti-interference in measuring time-series based similarity. However, the high time complexity of DTW restrains the applicability of real-time case. The existing DTW acceleration studies suffer from a loss of accuracy. How to accelerate computation while maintaining satisfying computational accuracy remains challenging. Motivated by sequential three-way decisions, this paper develops a novel model with adaptive sequential three-way decisions for dynamic time warping (AS3-DTW). Firstly, we systematically summarize distance differences under the context of adjacent tripartite search spaces for DTW, and propose five patterns of granularity adjustments of the search spaces. Furthermore, we present the corresponding calculation method of DTW adjacent tripartite search spaces distances difference. Finally, we construct a novel dynamism on adaptively adjusting time warping by combining sequence-based multi-granularity with sequential three-way decisions. Experimental results show that AS3-DTW effectively achieves promising trade-off between computational speed and accuracy on multiple UCR datasets when compared with the state-of-the-art algorithms.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"690 ","pages":"Article 121541"},"PeriodicalIF":8.1,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142434456","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}

引用次数: 0

Enhancing network robustness with structural prior and evolutionary techniques 用结构先验和进化技术增强网络鲁棒性

IF 8.1 1区计算机科学 0 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Sciences

Pub Date : 2024-10-09 DOI: 10.1016/j.ins.2024.121529

Jie Huang , Ruizi Wu , Junli Li

Robustness optimization in complex networks is a critical research area due to its implications for the reliability and stability of various systems. However, existing algorithms encounter two key challenges: the lack of integration of prior network knowledge, leading to suboptimal solutions, and high computational costs, which hinder their practical application. To address these challenges, this paper introduces Eff-R-Net, an efficient evolutionary algorithm framework aimed at enhancing the robustness of complex networks through accelerated evolution. Eff-R-Net leverages global and local network information, featuring a novel three-part composite crossover operator. Prior network knowledge is incorporated in mutation and local search operators to expedite the construction of networks with superior robustness. Additionally, a simplified method for calculating robustness enhances efficiency, while adaptive hyper-parameters dynamically adjust operators execution probabilities for optimal evolution. Extensive evaluations on both synthetic (Scale-Free, Erdös-Rényi, and Small-World) and three infrastructure real-world networks demonstrate the superiority of Eff-R-Net. The algorithm improves robustness by 12.8% and reduces computational time by 25.4% compared to state-of-the-art algorithm in real-world network experiments. These findings underscore Eff-R-Net's versatility and potential in enhancing network robustness across different domains.

复杂网络的鲁棒性优化是一个重要的研究领域，因为它对各种系统的可靠性和稳定性都有影响。然而，现有的算法遇到了两个主要挑战：一是缺乏对先前网络知识的整合，导致次优解；二是计算成本高，阻碍了算法的实际应用。为了应对这些挑战，本文介绍了 Eff-R-Net，这是一种高效的进化算法框架，旨在通过加速进化提高复杂网络的鲁棒性。Eff-R-Net 利用全局和局部网络信息，采用新颖的三部分复合交叉算子。在突变和局部搜索算子中加入了先验网络知识，以加快构建具有卓越鲁棒性的网络。此外，计算鲁棒性的简化方法提高了效率，而自适应超参数可动态调整算子的执行概率，以实现最佳进化。在合成网络（Scale-Free、Erdös-Rényi 和 Small-World）和三个基础设施真实世界网络上进行的广泛评估证明了 Eff-R-Net 的优越性。在真实世界网络实验中，与最先进的算法相比，该算法的鲁棒性提高了 12.8%，计算时间减少了 25.4%。这些发现凸显了 Eff-R-Net 在提高不同领域网络鲁棒性方面的多功能性和潜力。

{"title":"Enhancing network robustness with structural prior and evolutionary techniques","authors":"Jie Huang , Ruizi Wu , Junli Li","doi":"10.1016/j.ins.2024.121529","DOIUrl":"10.1016/j.ins.2024.121529","url":null,"abstract":"<div><div>Robustness optimization in complex networks is a critical research area due to its implications for the reliability and stability of various systems. However, existing algorithms encounter two key challenges: the lack of integration of prior network knowledge, leading to suboptimal solutions, and high computational costs, which hinder their practical application. To address these challenges, this paper introduces Eff-R-Net, an efficient evolutionary algorithm framework aimed at enhancing the robustness of complex networks through accelerated evolution. Eff-R-Net leverages global and local network information, featuring a novel three-part composite crossover operator. Prior network knowledge is incorporated in mutation and local search operators to expedite the construction of networks with superior robustness. Additionally, a simplified method for calculating robustness enhances efficiency, while adaptive hyper-parameters dynamically adjust operators execution probabilities for optimal evolution. Extensive evaluations on both synthetic (Scale-Free, Erdös-Rényi, and Small-World) and three infrastructure real-world networks demonstrate the superiority of Eff-R-Net. The algorithm improves robustness by 12.8% and reduces computational time by 25.4% compared to state-of-the-art algorithm in real-world network experiments. These findings underscore Eff-R-Net's versatility and potential in enhancing network robustness across different domains.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"690 ","pages":"Article 121529"},"PeriodicalIF":8.1,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142422898","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}

引用次数: 0

Enhanced tensor based embedding anchor learning for multi-view clustering 基于张量的增强型嵌入式锚点学习，用于多视角聚类

IF 8.1 1区计算机科学 0 COMPUTER SCIENCE, INFORMATION SYSTEMS

Information Sciences

Pub Date : 2024-10-09 DOI: 10.1016/j.ins.2024.121532

Beihua Yang, Peng Song, Yuanbo Cheng, Shixuan Zhou, Zhaowei Liu

Existing anchor graph based multi-view clustering methods can overcome the problem of high computational cost in traditional multi-view clustering methods. However, the anchor points selected from high-dimensional data often contain irrelevant noise and outliers, which would affect the clustering performance. To address this issue, we propose an embedding anchor based multi-view clustering method, called enhanced tensor based embedding anchor learning (ETEAL). Specifically, we unify the learning process of latent embedding space, anchor points, and anchor graphs into a common framework, which eliminates noise and redundant data in the original space and enhances the synergistic optimization between the components. Meanwhile, we employ an enhanced tensor strategy to constrain the embedding anchor graphs, which exploits the higher-order relationships between views and recovers the global low-rank property of the embedding anchor graphs. Finally, we develop an anchor graph fusion strategy, which significantly reduces the huge overhead of traditional graph fusion that requires the construction of complete graphs. Experimental results on eight benchmark datasets show that the proposed method significantly outperforms other state-of-the-art methods in terms of scalability and clustering accuracy.

现有的基于锚点图的多视图聚类方法可以克服传统多视图聚类方法计算成本高的问题。然而，从高维数据中选取的锚点往往包含无关噪声和异常值，这会影响聚类性能。针对这一问题，我们提出了一种基于嵌入锚点的多视图聚类方法，即增强型基于张量的嵌入锚点学习（ETEAL）。具体来说，我们将潜在嵌入空间、锚点和锚图的学习过程统一到一个共同的框架中，从而消除了原始空间中的噪声和冗余数据，增强了各组成部分之间的协同优化。同时，我们采用了一种增强的张量策略来约束嵌入锚图，这种策略利用了视图之间的高阶关系，恢复了嵌入锚图的全局低阶属性。最后，我们开发了一种锚图融合策略，大大减少了传统图融合需要构建完整图的巨大开销。在八个基准数据集上的实验结果表明，所提出的方法在可扩展性和聚类准确性方面明显优于其他最先进的方法。

{"title":"Enhanced tensor based embedding anchor learning for multi-view clustering","authors":"Beihua Yang, Peng Song, Yuanbo Cheng, Shixuan Zhou, Zhaowei Liu","doi":"10.1016/j.ins.2024.121532","DOIUrl":"10.1016/j.ins.2024.121532","url":null,"abstract":"<div><div>Existing anchor graph based multi-view clustering methods can overcome the problem of high computational cost in traditional multi-view clustering methods. However, the anchor points selected from high-dimensional data often contain irrelevant noise and outliers, which would affect the clustering performance. To address this issue, we propose an embedding anchor based multi-view clustering method, called enhanced tensor based embedding anchor learning (ETEAL). Specifically, we unify the learning process of latent embedding space, anchor points, and anchor graphs into a common framework, which eliminates noise and redundant data in the original space and enhances the synergistic optimization between the components. Meanwhile, we employ an enhanced tensor strategy to constrain the embedding anchor graphs, which exploits the higher-order relationships between views and recovers the global low-rank property of the embedding anchor graphs. Finally, we develop an anchor graph fusion strategy, which significantly reduces the huge overhead of traditional graph fusion that requires the construction of complete graphs. Experimental results on eight benchmark datasets show that the proposed method significantly outperforms other state-of-the-art methods in terms of scalability and clustering accuracy.</div></div>","PeriodicalId":51063,"journal":{"name":"Information Sciences","volume":"690 ","pages":"Article 121532"},"PeriodicalIF":8.1,"publicationDate":"2024-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142422900","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}

引用次数: 0