检索结果-内蒙古大学图书馆

6th Northern Lights Deep Learning Conference, NLDL 2025

作者： Brüsch, Thea Wickstrøm, Kristoffer Knutsen Schmidt, Mikkel N. Alstrøm, Tommy Sonne Jenssen, Robert Department of Applied Mathematics and Computer Science Technical University of Denmark Denmark Department of Physics and Technology UiT The Arctic University of Norway Norway Norwegian Computing Center Oslo Norway

Time series data is fundamentally important for many critical domains such as healthcare, finance, and climate, where explainable models are necessary for safe automated decision-making. To develop explainable artificial intelligence in these domains therefore implies explaining salient information in the time series. Current methods for obtaining saliency maps assumes localized information in the raw input space. In this paper, we argue that the salient information of a number of time series is more likely to be localized in the frequency domain. We propose FreqRISE, which uses masking-based methods to produce explanations in the frequency and time-frequency domain, and outperforms strong baselines across a number of tasks. The source code is available here: https://***/theabrusch/FreqRISE. © NLDL *** rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Advancements in Machine Learning and Deep Learning Techniques for Sickle Cell Disease Detection using Digital Morphology

Advancements in Machine Learning and Deep Learning Technique...

引用

2024 International Conference on Computer and Applications, ICCA 2024

作者： Abdul, Rahiman S.K. George, E Ben Anand, V.K. Al Abri, Eman Said Al Rahbi, Muna Saif Nadaf, Kausar Niazi Ahmed Al-Abdali, Al Waleed College of Computing and Information Sciences University of Technology and Applied Sciences Muscat Oman Saud Bahwan Group Health Care Department Muscat Oman

ISBN: (纸本)9798350367560

Sickle Cell Disease (SCD) is still a serious public health concern, particularly in areas with inadequate access to healthcare. The management of the condition depends on an early and precise diagnosis;yet, traditional diagnostic techniques, such manual blood smear examinations, are frequently laborious, subjective, and prone to variability. The major developments in deep learning (DL) and machine learning (ML) approaches that are redefining SCD detection especially with regard to the application of digital morphology are examined in this review paper. The sickle-shaped red blood cells from microscopic pictures can effectively categorized by utilizing convolutional neural networks (CNNs), transfer learning, and ensemble learning techniques. This method offers better accuracy, speed, and scalability over conventional approaches. These models allow for more accurate assessments of cell morphology in addition to improving the accuracy of SCD diagnosis, lowering the possibility of human error. Additionally, issues like interpretability and data imbalance, offering answers via creative model designs and pre-processing methods. This research demonstrates the revolutionary potential of machine learning and deep learning in the healthcare industry, where intelligent, automated technologies may supplement clinical knowledge and improve outcomes for SCD patients worldwide. By combining these technologies, scalable and easily available diagnostic tools could be developed in the future, leading to more fair healthcare delivery. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Improving Quality of Service Using a Innovative Energy Effective Reinforcement Based Path Finding in Mobile Ad Hoc Network 2

Improving Quality of Service Using a Innovative Energy Effec...

引用

2nd International Conference on computing and Data Analytics, ICCDA 2024

作者： Singaravelan, M. Saleem Raja, A. Thangam, S. School of Computing Vel Tech Rangarajan Dr. Sagunthala R & D Institute of Science and Technology Avadi Chennai India College of Computing and information Sciences University of Technology and Applied Sciences Shinas Oman Amrita school of computing Department of computer science and engineering Bengaluru India

ISBN: (数字)9798331542573

ISBN: (纸本)9798331542573

RL stands for reinforcement learning, is a component of the broader field of machine learning. It provides a framework its features help to select an efficient action using previously learned information of interaction and its environment. Mobile nodes are grouped to form a network for communicating themselves termed autonomous network with mobile nodes (MANET). This is a dynamic and independent network requiring no administrative control. The mobile nodes in the network are to be part or apart at any time that causes the topology changes. The frequent topology changes make it hard to determine the destination route. The route discovery in the dynamic topology that consumes more node energy and, minimize the network lifetime. The energy usage and network lifetime are a vital part of the MANET. This paper proposes a new global routing protocol using a reinforcement learning algorithm called modified energy and signal strength Q-learning ad hoc on-demand distance vector algorithm (ESQAODV). This process conducts a result comparison to a local routing algorithm known as AODVSARSA. The proposed algorithm proved a better result with reference to average distribution ratio compared to the other, and complete delay metrics © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Advancing Industrial Cybersecurity: Machine Learning-Based Detection and Mitigation of IIoT Attacks 5

Advancing Industrial Cybersecurity: Machine Learning-Based D...

引用

5th International Conference on IoT Based Control Networks and Intelligent Systems, ICICNIS 2024

作者： Kumar, S Gouri Kiran Prasad, Raavi Satya College of Computing and Information Sciences University of Technology and Applied Sciences Muscat Oman Dhanekula Institute of Engineering & Technology Department of Computer Science & Engineering A.P. Vijayawada India

ISBN: (纸本)9798331518097

A threatening entry or exploitation of vulnerabilities within an industrial network or system connected to IoT devices is an IIoT (Industrial Internet of Things) attack. Due to manufacturing delays, broken equipment, lost confidential data, and fines from the government, cyberattacks on IIoT systems can cause significant financial losses. Rapid reaction and recovery are made possible by efficient detection systems, which can lessen these losses. Many researchers have applied machine learning (ML) approaches to predict IIoT attacks in advance. Industrial networks can be highly effectively and dynamically protected against various cyber threats by utilizing MLthrough IIoT attack distinguishing. This ML methodology is a vital component of today's cybersecurity toolkit for IIoT contexts owing to its capacity for handling massive info, adjusting for novel threats, & delivering insights in real-time. Industrial networks can be highly effectively and dynamically protected against cyber threats by utilizing ML to IIoT attack identification. ML is a vital component of today's cybersecurity toolkit for IIoT contexts for its capacity to handle massive quantities of info, adjust to emerging risks, & deliver insights in real time. IIoT data can contain noise, missing values, or irrelevant information, negatively affecting model performance. Identifying and extracting relevant features from IIoT data was highly complex and time-consuming. The methodology's effectiveness is heavily related to features with high quality utilized. This paper discusses the various techniques, their limitations, and the gaps studied during the literature survey. Among the machine learning approaches, the model should adapt to the new infrastructure attacks, which is effectively possible only through meta-heuristic learning algorithms. © 2024 IEEE.

关键词： Cyber attacks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Classification Algorithm Recommendation in Automated Machine Learning: A Meta-Learning Approach Using Multivariate Sparse Group Lasso

引用

Computer Modeling in Engineering & Sciences 2025年第2期142卷 1611-1636页

作者： Irfan Khan Xianchao Zhang Ramesh Kumar Ayyasamy Saadat M.Alhashmi Azizur Rahim School of Software Dalian University of TechnologyDalian116620China Faculty of Information and Communication Technology Universiti Tunku Abdul RahmanKampar31900Malaysia College of Computing and Informatics University of SharjahSharjah27272United Arab Emirates Department of Computer Systems Engineering University of Engineering and Applied SciencesSwat19200Pakistan

The rapid growth of machine learning(ML)across fields has intensified the challenge of selecting the right algorithm for specific tasks,known as the Algorithm Selection Problem(ASP).Traditional trial-and-error methods have become impractical due to their resource *** Machine Learning(AutoML)systems automate this process,but often neglect the group structures and sparsity in meta-features,leading to inefficiencies in algorithm recommendations for classification *** paper proposes a meta-learning approach using Multivariate Sparse Group Lasso(MSGL)to address these *** method models both within-group and across-group sparsity among meta-features to manage high-dimensional data and reduce multicollinearity across eight meta-feature *** Fast Iterative Shrinkage-Thresholding Algorithm(FISTA)with adaptive restart efficiently solves the non-smooth optimization *** validation on 145 classification datasets with 17 classification algorithms shows that our meta-learning method outperforms four state-of-the-art approaches,achieving 77.18%classification accuracy,86.07%recommendation accuracy and 88.83%normalized discounted cumulative gain.

关键词： Meta-learning machine learning automated machine learning classification meta-features

来源：评论

学校读者我要写书评

暂无评论

Dynamic generative R-CNN

引用

Neural computing and Applications 2025年第10期37卷 7107-7120页

作者： Saffarini, Rasha Khamayseh, Faisal Awwad, Yousef Sabha, Muath Eleyan, Derar IT Engineering Arab American University Jenin Palestine Department of Computer Science Palestine Polytechnic University Hebron Palestine Department of Computer System Engineering Palestine Technical University Tulkarm Palestine Department of Multimedia Technology Arab American University Jenin Palestine Department of Applied Computing Palestine Technical University Tulkarm Palestine

Different techniques have been developed for object detection and recognition. These techniques can be divided into single-shot and two-shot methods. Single-shot methods focus on real-time applications, while two-shot methods are used in applications requiring higher accuracy. However, different versions of the two-shot techniques produce limited results in terms of accuracy and speed, or both. Therefore, this study proposes a novel model called dynamic generative R-CNN (DGR-CNN) that reduces the number of proposed regions using a dynamic programming model that applies the graph similarity method over graph-based image segmentation. Additionally, the proposed model employs DCGAN technique to improve detection performance. DGR-CNN reduces the overall detection and classification time and enhances the detection accuracy. The PASCAL VOC2007 and MS COCO datasets were utilized to evaluate the model. The results showed that DGR-CNN significantly reduces the number of candidate regions compared to the selective search algorithm employed in R-CNN and fast R-CNN. Although fast R-CNN utilizes 2000 regions and faster R-CNN utilizes 300 regions, DGR-CNN reduces the number of regions to approximately 130. The mean average precision of the proposed method was 75.1% on the PASCAL VOC2007, while fast and faster R-CNN scored 66.9% and 69.9%, respectively. Moreover, the DGR-CNN model significantly improved the classification accuracy when tested on the MS COCO dataset, achieving an MAP of 68.76%, compared with 32.64% and 42.3% for fast and faster R-CNN. This increase in accuracy was achieved without significantly compromising the speed compared with faster R-CNN. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

A Novel Congestion Control Scheme Using Fuzzy Logic Systems to Enhance the Path Selection Criteria in Routing Protocols for Low-Power and Lossy Networks on the Internet of Things

引用

SN Computer Science 2024年第5期5卷 610页

作者： Aimtongkham, Phet Musikawan, Pakarat Kongsorot, Yanika So-In, Chakchai Applied Network Technology (ANT) Department of Computer Science College of Computing Khon Kaen University Khon Kaen 40002 Thailand Advanced Intelligent Interdisciplinary Integration (AIII) Department of Computer Science College of Computing Khon Kaen University Khon Kaen 40002 Thailand

In routing protocols for low-power and lossy (RPL)-based Internet of Things (IoT) networks, congestion control is essential for ensuring efficient and dependable communications with energy awareness. As the number of devices and the amount of data traffic increase, congestion typically occurs, resulting in degraded network performance, i.e., increased packet losses and delays with decreased energy efficiency. Many parameters relating to congestion behaviors have been considered and dynamically changed, as have the possible transmission paths either within a single destination-oriented directed acyclic graph (DoDAG) or across different DoDAGs. Thus, this research investigates congestion mitigation methods for RPL-IoT networks based on a fuzzy logic system (FLS) by proposing a novel FLS scheme to determine the selection criteria for conducting path management in three FLS components: (1) inside/outside the DoDAG selection process, (2) inside the path selection process within a single DoDAG (the so-called inside DoDAG) for high or to-be-high loads, and (3) inside the path selection process across the DoDAGs (the so-called outside DoDAG) during severe congestion applying the concept of a multi-DoDAG node (MDN) to determine both a suitable MDN and data traffic for alternate path selection. Simulation results demonstrate that the proposed method effectively controls congestion levels, decreases packet losses and latency, and attains improved energy efficiency in comparison with the state-of-the-art congestion control schemes in RPL-IoT networks. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Congestion controls Fuzzy logic systems Internet of Things Objective function Path selection Routing protocol for low-power and lossy networks

来源：评论

学校读者我要写书评

暂无评论

Towards Swarm Intelligence Ontology for Formal Description of Metaheuristics Algorithms 4

Towards Swarm Intelligence Ontology for Formal Description o...

引用

4th International Conference on computing, Networking, Telecommunications and Engineering Sciences Applications, CoNTESA 2023

作者： Tulicic, Domagoj Horvat, Marko Lovrencic, Sandra University of Applied Science in Information Technology Department of Computer System and Networks Zagreb Croatia University of Zagreb Faculty of Electrical Engineering and Computing Department of Applied Computing Zagreb Croatia University of Zagreb Faculty of Organization and Informatics Foundations of Information Sciences Department of Theoretical and Applied Varaždin Croatia

ISBN: (纸本)9798350382754

Swarm intelligence is a class of nature-inspired metaheuristic algorithms, that is specifically derived from biological systems in nature with an emphasis on their social interactions. These algorithms have been primarily developed to solve complex combinatorial optimization problems and are extremely popular among researchers in metaheuristic algorithms. This is evident from the numerous published scientific papers describing algorithms inspired by biological systems, ranging from insects such as ants and bees to the largest animals on the planet like whales. All these algorithms share some common concepts and relationships that can be explicitly identified, named, and formalized. These shared concepts and relationships can be organized and represented through the construction of an ontology. The goal of this paper is to present a formal ontology that standardizes vocabulary, as well as a set of rules enabling precise communication and reasoning about these algorithms. Ultimately, this will improve our understanding, analysis, research processes, and utilization of these computational approaches. © 2023 IEEE.

关键词： Swarm intelligence

来源：评论

学校读者我要写书评

暂无评论

Towards Tight Bounds for the Graph Homomorphism Problem Parameterized by Cutwidth via Asymptotic Matrix Parameters 51

Towards Tight Bounds for the Graph Homomorphism Problem Para...

引用

51st International Colloquium on Automata, Languages, and Programming, ICALP 2024

作者： Groenland, Carla Mannens, Isja Nederlof, Jesper Piecyk, Marta Rzążewski, Pawel Delft Institute of Applied Mathematics Netherlands Department of Information and Computing Sciences Utrecht University Netherlands Warsaw University of Technology Poland University of Warsaw Poland

ISBN: (纸本)9783959773225

A homomorphism from a graph G to a graph H is an edge-preserving mapping from V (G) to V (H). In the graph homomorphism problem, denoted by Hom(H), the graph H is fixed and we need to determine if there exists a homomorphism from an instance graph G to H. We study the complexity of the problem parameterized by the cutwidth of G, i.e., we assume that G is given along with a linear ordering v1, . . ., vn of V (G) such that, for each i ∈ {1, . . ., n − 1}, the number of edges with one endpoint in {v1, . . ., vi} and the other in {vi+1, . . ., vn} is at most k. We aim, for each H, for algorithms for Hom(H) running in time ckHnO(1) and matching lower bounds that exclude ckH·o(1)nO(1) or ckH(1−Ω(1))nO(1) time algorithms under the (Strong) Exponential Time Hypothesis. In the paper we introduce a new parameter that we call mimsup(H). Our main contribution is strong evidence of a close connection between cH and mimsup(H): an information-theoretic argument that the number of states needed in a natural dynamic programming algorithm is at most mimsup(H)k, lower bounds that show that for almost all graphs H indeed we have cH ≥ mimsup(H), assuming the (Strong) Exponential-Time Hypothesis, and an algorithm with running time exp(O(mimsup(H) · k log k))nO(1). In the last result we do not need to assume that H is a fixed graph. Thus, as a consequence, we obtain that the problem of deciding whether G admits a homomorphism to H is fixed-parameter tractable, when parameterized by cutwidth of G and mimsup(H). The parameter mimsup(H) can be thought of as the p-th root of the maximum induced matching number in the graph obtained by multiplying p copies of H via a certain graph product, where p tends to infinity. It can also be defined as an asymptotic rank parameter of the adjacency matrix of H. Such parameters play a central role in, among others, algebraic complexity theory and additive combinatorics. Our results tightly link the parameterized complexity of a problem to such an asymptoti

关键词： Parameterization

来源：评论

学校读者我要写书评

暂无评论

Enhanced Intrusion Detection with Advanced Deep Features and Ensemble Classifier Techniques

引用

SN Computer Science 2025年第4期6卷 1-17页

作者： Toralkar, Pawan Mainalli, Kavita Allagi, Shridhar Debnath, Sanjoy Kumar Bagchi, Susama Leong, Wai Yie Khan, Muhammad Numan Ali Department of Artificial Intelligence Madanapalle Institute of Technology and Science Andhra Pradesh Madanapalle India Department of Computer Science and Engineering KLEIT: KLE Institute of Technology Karnataka Hubballi India Chitkara University Institute of Engineering and Technology Chitkara University Punjab India INTI International University and Colleges Negeri Sembilan Nilai Malaysia College of Computing and Applied Sciences University of Technology and Applied Sciences Shinas Oman

Intrusion detection is crucial for modern cyber-security, aiming to detect and mitigate unauthorized access and malicious activities within networks. Traditional systems often rely on linear models, which may fail to capture complex patterns in advanced threats. This study proposes a heuristic approach to select the best deep features through an ensemble framework consisting of Random Forest (RF) and Support Vector Machine (SVM) classifiers using soft voting. The novelty of this work is the deep feature extraction through Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Autoencoder for intrusion detection. These extracted features were fed separately for classification. In order to balance the explainability and accuracy, the ensemble classifier consisting of RF and SVM, uses feature importance ranking (RF) and unambiguous decision bounds (SVM) to provide partial interpretability. Evaluating the NSL-KDD and CICDDoS2019 datasets, our ensemble model with CNN features achieves near-perfect accuracy of 99.99% and 99.91% respectively. RNN and Autoencoder models also show high performance, each reaching 99.3% accuracy on the CICDDoS2019 dataset with Soft Voting classifiers. This work highlights the importance of the ensemble method with better feature selection in intrusion detection. The proposed system outperformed the standard methods by providing a robust, scalable solution for organizational security. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolution neural network (CNN) Intrusion detection Random forest (RF) Recurrent neural network (RNN) Voting classification (Voting)

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：