检索结果-内蒙古大学图书馆

2nd IEEE International Conference on Control, Electronics and computer technology, ICCECT 2024

作者： Zhang, Peng Ma, Ru Zhang, Fuzhi School of Computer Science and Technology Beijing Institute of Technology Beijing China School of Information Science and Engineering Yanshan University Hebei Qinhuangdao China

ISBN: (纸本)9798350380958

Aiming at the problems that an individual bundle recommendation approach is prone to producing recommendation bias and has poor generality, we propose an ensemble approach for bundle recommendation based on performance evaluation. First, we propose an ensemble bundle recommendation framework, which integrates the predictive power of multiple base recommenders to improve the bundle recommendation performance. Second, we train the base recommenders in the ensemble on the training set and endow them with importance weights according to their performance evaluation on the validation set. Finally, we propose two bundle recommendation strategies for the ensemble and aggregate the recommendation lists generated with these two strategies in a weighted manner. Experiments show that the average improvement of recall@K and NDCG@K for our ensemble approach over state-of-the-art methods is 3.82% and 7.39%, respectively. © 2024 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Frequency and Edge Guided Network for Weakly Supervised Salient Object Detection 9

Frequency and Edge Guided Network for Weakly Supervised Sali...

引用

9th International Conference on Intelligent Informatics and BioMedical sciences, ICIIBMS 2024

作者： Liu, Yuyan Tao, Chenhao Shanghai Institute of Technology School of Computer Science and Information Engineering Shanghai China University of Shanghai for Science and Technology Shanghai China

ISBN: (纸本)9798350363043

Recently, due to the large amount of manpower and material resources required for fully supervised salient object detection in pixel-level label labeling, some weakly supervised salient object detection (SOD) based on scribble annotations have been proposed to reduce the burden. However, since scribble annotations are sparse annotations, they lack salient object structure information, making it difficult for the model to accurately distinguish the foreground from the background, especially on the object boundary. This paper proposes a boundary and frequency domain guided learning framework to obtain information to help the network locate the object, improve the recognition rate, refine the edge and strengthen the capture of structural details. In addition, the edge and frequency domain are fused to make up for the lack of lowlevel features and make full use of the dominant attributes of different modes. Extensive experiments show that our model achieves competitive performance against the state-of-the-art weakly supervised SOD methods, demonstrating the superiority of effectiveness of our proposed network. © 2024 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Motion path planning method based on map reference points

引用

Multimedia Tools and Applications 2024年 1-21页

作者： Yu, Wenbin Fan, Cheng Zhang, Chengjun Chen, Yadang Li, Yangsong Zhang, Yifan Yin, Na Cen, Xiaolin School of Software Nanjing University of Information Science and Technology Jiangsu Nanjing210044 China School of Computer Science Nanjing University of Information Science and Technology Jiangsu Nanjing210044 China Wuxi Institute of Technology Nanjing University of Information Science and Technology Jiangsu Wuxi214000 China Nanjing University of Information Science and Technology Jiangsu Nanjing210044 China School of Computer Science Northeastern University Boston02101 United States Computer Science and Engineering The University of New South Wales SydneyNSW02101 Australia

In recent years, significant progress has been made in the development of the Rapidly-exploring Random Trees (RRT) algorithm, a probabilistic sampling-based search approach. This algorithm effectively addresses complex challenges in path planning within high-dimensional spaces characterized by intricate constraints. However, the RRT algorithm faces certain limitations, including sluggish computational performance in handling intricate calculations, susceptibility to becoming trapped in confined path regions, and difficulties in generating randomized paths across a given map. To overcome these issues, this study introduces a motion path planning method based on map reference points. Initially, the algorithm selects strategic map reference points, followed by the utilization of an enhanced RRT algorithm for fundamental path selection. Real-world scenario data is collected to empirically evaluate the algorithm’s performance. The outcomes of the experiments reveal that, within the same experimental setting, the proposed method outperforms conventional motion path planning algorithms by effectively circumventing obstacles. Furthermore, it substantially reduces the resource wastage caused by global random sampling in the RRT algorithm, leading to expedited path planning and improved pathfinding success rates. Additionally, the proposed approach enhances the RRT algorithm’s adaptability to some extent, particularly in negotiating narrow passages and intricate environments. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

I/O Efficient Early Bursting Cohesive Subgraph Discovery in Massive Temporal Networks

引用

Journal of computer science & technology 2022年第6期37卷 1337-1355页

作者： Yuan Li Jie Dai Xiao-Lin Fan Yu-Hai Zhao Guo-Ren Wang School of Information Science and Technology North China University of TechnologyBeijing 100144China School of Computer Science and Engineering Northeastern UniversityShenyang 110169China School of Computer Beijing Institute of TechnologyBeijing 100081China

Temporal networks are an effective way to encode temporal information into graph data *** the bursting cohesive subgraph(BCS),which accumulates its cohesiveness at the fastest rate,is an important problem in temporal *** BCS has a large number of applications,such as representing emergency events in social media,traffic congestion in road networks and epidemic outbreak in ***,existing methods demand the BCS lasting for a time interval,which neglects the timeliness of the *** this paper,we design an early bursting cohesive subgraph(EBCS)model based on the k-core to enable identifying the burstiness as soon as *** find the EBCS,we first construct a time weight graph(TWG)to measure the bursting level by integrating the topological and temporal ***,we propose a global search algorithm,called GS-EBCS,which can find the exact EBCS by iteratively removing nodes from the ***,we propose a local search algorithm,named LS-EBCS,to find the EBCS by first expanding from a seed node until obtaining a candidate k-core and then refining the k-core to the result subgraph in an optimal time ***,considering the situation that the massive temporal networks cannot be completely put into the memory,we first design an I/O method to build the TWG and then develop I/O efficient global search and local search algorithms,namely I/O-GS and I/O-LS respectively,to find the EBCS under the semi-external *** experiments,conducted on four real temporal networks,demonstrate the efficiency and effectiveness of our proposed *** example,on the DBLP dataset,I/O-LS and LS-EBCS have comparable running time,while the maximum memory usage of I/O-LS is only 6.5 MB,which is much smaller than that of LS-EBCS taking 308.7 MB.

关键词： early bursting cohesive subgraph(EBCS) I/O efficient algorithm semi-external model temporal network

来源：评论

学校读者我要写书评

暂无评论

Efficient Authentication System Using Wavelet Embeddings of Otoacoustic Emission Signals

引用

computer Systems science & engineering 2023年第5期45卷 1851-1867页

作者： V.Harshini T.Dhanwin A.Shahina N.Safiyyah A.Nayeemulla Khan Department of Information Technology Sri Sivasubramaniya Nadar College of EngineeringChennai603110India School of Computer Science and Engineering Vellore Institute of TechnologyChennai600127India

Biometrics,which has become integrated with our daily lives,could fall prey to falsification attacks,leading to security *** our paper,we use Transient Evoked Otoacoustic Emissions(TEOAE)that are generated by the human cochlea in response to an external sound stimulus,as a biometric *** are robust to falsification attacks,as the uniqueness of an individual’s inner ear cannot be *** this study,we use both the raw 1D TEOAE signals,as well as the 2D time-frequency representation of the signal using Continuous Wavelet Transform(CWT).We use 1D and 2D Convolutional Neural Networks(CNN)for the former and latter,respectively,to derive the feature *** corresponding lower-dimensional feature maps are obtained using principal component analysis,which is then used as features to build classifiers using machine learning techniques for the task of person identification.T-SNE plots of these feature maps show that they discriminate well among the *** the various architectures explored,we achieve a best-performing accuracy of 98.95%and 100%using the feature maps of the 1D-CNN and 2D-CNN,respectively,with the latter performance being an improvement over all the earlier *** performance makes the TEOAE based person identification systems deployable in real-world situations,along with the added advantage of robustness to falsification attacks.

关键词： Person identification system cochlea:transient evoked otoacoustic emission wavelet transform convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Maximizing Profit of the Service Providers through Efficient Content Placement and Retrieval in an IoT-enabled Edge-Cloud Platform 18

Maximizing Profit of the Service Providers through Efficient...

引用

18th IEEE International Conference on Advanced Networks and Telecommunications Systems, ANTS 2024

作者： Gandhi, Preet Mahanta, Abhishriya Ghose, Manojit Sahu, Aryabartta Indian Institute of Technology Department of Computer Science and Engineering Guwahati India Indian Institute of Technology School of Computing and Electrical Engineering Mandi India Indian Institute of Information Technology Department of Computer Science and Engineering Guwahati India

ISBN: (纸本)9798350391725

In today's 5G era, the amount of data generated by the Internet of Things (IoT) devices is enormous. Data is processed and stored in the cloud under a traditional cloud computing architecture, and real-time processing and retrieval of these data is time-consuming. Therefore, an edge-cloud framework comes in handy where the content generated is placed on the edge nodes of the network, and it can reach the endusers in a time-bound manner. A majority of the previous research works on content placement under the edge-cloud architecture primarily aimed at improving the latency of the system. However, the approach proposed in this paper aims to maximize the profit of the service provider by effectively placing the content in a hierarchically organized edge-cloud network. We propose two strategies and compare them with random and greedy approaches for both synthetic and real data sets. The experimental results demonstrate the effectiveness of the proposed strategies, and the profit of the service providers increases by 38.81% (maximum) and 10.51% (on average) for the synthetic dataset and by 65.74% (maximum) and 22.83% (on average) for the real-world dataset when compared with the greedy approach. © 2024 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

CVS-FLN: a novel IoT-IDS model based on metaheuristic feature selection and neural network classification model

引用

Multimedia Tools and Applications 2024年第39期83卷 86557-86591页

作者： Geetha, R. Jegatheesan, A. Dhanaraj, Rajesh Kumar Vijayalakshmi, K. Nayyar, Anand Arulkumar, V. Velmurugan, J. Thavasimuthu, Rajendran Department of Computing Technologies SRM Institute of Science and Technology Kattankulathur Campus Chengalpattu Chennai India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Tamil Nadu Chennai India Pune India Department of Computer Science and Engineering Ramco Institute of Technology Tamil Nadu Rajapalayam India Graduate School Faculty of Information Technology Duy Tan University Da Nang550000 Viet Nam School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Vellore India

The Internet of Things (IoT) is one of the technologies that will be used all over the world in the future, and its security and privacy features are the primary concerns. However, the most critical limitation to overcome before the IoT's widespread use is addressing its security concerns. One of the most critical tasks to address the security problems posed by the IoT is the detection of network intrusions. Intrusion Detection Systems (IDS) for the IoT face substantial challenges because of the functional and physical diversity of the devices. In the rapidly growing IoT industry, the use of IDS is essential for ensuring security since many devices communicate efficiently. IDS are crucial for continual monitoring and responding to any security risks, ensuring the integrity and reliability of interconnected IoT networks. The primary objective of this research is to develop a feature-selection-based intrusion classification model. Therefore, we develop an IoT-IDS feature selection-based classification model to detect intrusions. We propose a metaheuristic algorithm, the Chaotic Vortex Search (CVS) algorithm, for feature selection. The Fast-Learning Network (FLN), an artificial neural network (ANN) model, is additionally proposed for classification. We use the IDS datasets, such as CIC IDS-2017 and BoT-IoT, to evaluate our research model. To test and validate the proposed model, extensive experiments were performed, and the outcomes stated that the CVS-FLN model achieved 99.77% accuracy, 99.92% specificity, 99.60% precision, 99.81% detection rate, and a 99.72% F1-score in the CIC IDS dataset, and 99.68% accuracy, 99.30% precision, 99.83% specificity, 99.11% detection rate, and a 99.21% F1-score in evaluating the BoT-IoT dataset. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

Implementation of Elliptic Curve Cryptosystem with Bitcoin Curves on SECP256k1, NIST256p, NIST521p, and LLL

Journal of ICT Standardization

引用

Journal of ICT Standardization 2023年第4期11卷 329-354页

作者： Ulla, Mohammed Mujeer Preethi Khan, Md Sameeruddin Sakkari, Deepak S. School of Computer Science and Engineering Presidency University Karnataka Bangalore India Department of Information Technology Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India Department of Computer Science and Engineering Sri Krishna Institute of Technology Bangalore India

Very recent attacks like ladder leaks demonstrated the feasibility of recovering private keys with side-channel attacks using just one bit of secret nonce. ECDSA nonce bias can be exploited in many ways. Some attacks on ECDSA involve complicated Fourier analysis and lattice mathematics. This paper will enable cryptographers to identify efficient ways in which ECDSA can be cracked on curves NIST256p, SECP256k1, NIST521p, and weak nonce, kind of attacks that can crack ECDSA and how to protect yourself. Initially, we begin with an ECDSA signature to sign a message using the private key and validate the generated signature using the shared public key. Then we use a nonce or a random value to randomize the generated signature. Every time we sign, a new verifiable random nonce value is created, and a way in which the intruder can discover the private key if the signer leaks any one of the nonce values. Then we use Lenstra–Lenstra–Lovasz (LLL) method as a black box, we will try to attack signatures generated from bad nonce or bad random number generator (RAG) on NIST256p, SECP256k1 curves. The combination of nonce generation, post-message signing, and validation in ECDSA helps achieve Uniqueness, Authentication, Integrity, and Non-Repudiation. The analysis is performed by considering all three curves for the implementation of the Elliptic Curve Digital Signature Algorithm (ECDSA). The comparative analysis for each of the selected curves in terms of computational time is done with the leak of nonce and with the Lenstra–Lenstra–Lovasz method to crack ECDSA. The average computational costs to break ECDSA with curves NIST256p, NIST521p, and SECP256k1 are 0.016, 0.34,0.46 respectively which is almost zero depicting the strength of the algorithm. The average computational costs to break ECDSA with curves SECP256K1 and NIST256p using LLL are 2.9 and 3.4 respectively. © 2023 River Publishers.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

LKAW: A Robust Watermarking Method Based on Large Kernel Convolution and Adaptive Weight Assignment

引用

computers, Materials & Continua 2023年第4期75卷 1-17页

作者： Xiaorui Zhang Rui Jiang Wei Sun Aiguo Song Xindong Wei Ruohan Meng Engineering Research Center of Digital Forensics Ministry of EducationJiangsu Engineering Center of Network MonitoringSchool of Computer and SoftwareNanjing University of Information Science&TechnologyNanjing210044China Wuxi Research Institute Nanjing University of Information Science&TechnologyWuxi214100China Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology(CICAEET) Nanjing University of Information Science&TechnologyNanjing210044China School of Automation Nanjing University of Information Science&TechnologyNanjing210044China School of Instrument Science and Engineering Southeast UniversityNanjing211189China School of Teacher Education Nanjing University of Information Science&TechnologyNanjing210044China School of Computer Science Engineering Nanyang Technological UniversitySingapore

Robust watermarking requires finding invariant features under multiple attacks to ensure correct *** learning has extremely powerful in extracting features,and watermarking algorithms based on deep learning have attracted widespread *** existing methods use 3×3 small kernel convolution to extract image features and embed the ***,the effective perception fields for small kernel convolution are extremely confined,so the pixels that each watermarking can affect are restricted,thus limiting the performance of the *** address these problems,we propose a watermarking network based on large kernel convolution and adaptive weight assignment for loss *** uses large-kernel depth-wise convolution to extract features for learning large-scale image information and subsequently projects the watermarking into a highdimensional space by 1×1 convolution to achieve adaptability in the channel ***,the modification of the embedded watermarking on the cover image is extended to more *** the magnitude and convergence rates of each loss function are different,an adaptive loss weight assignment strategy is proposed to make theweights participate in the network training together and adjust theweight ***,a high-frequency wavelet loss is proposed,by which the watermarking is restricted to only the low-frequency wavelet sub-bands,thereby enhancing the robustness of watermarking against image *** experimental results show that the peak signal-to-noise ratio(PSNR)of the encoded image reaches 40.12,the structural similarity(SSIM)reaches 0.9721,and the watermarking has good robustness against various types of noise.

关键词： Robust watermarking large kernel convolution adaptive loss weights high-frequency wavelet loss deep learning

来源：评论

学校读者我要写书评

暂无评论

AI LipReader-Transcribing Speech from Lip Movements 6

AI LipReader-Transcribing Speech from Lip Movements

引用

6th IEEE International Conference on Emerging Smart Computing and Informatics, ESCI 2024

作者： Ajitha, D. Dutta, Disha Saha, Falguni Giri, Parus Kant, Rohan School of Computer Science and Engineering Vellore Institute of Technology Vellore Department of Software Systems Vellore India School of Computer Science and Engineering Vellore Institute of Technology Department of Database Systems Vellore India School of Computer Science and Engineering Vellore Institute of Technology Vellore Department of Information Security Vellore India

ISBN: (纸本)9798350306613

Lip Reading AI is a discipline that is rapidly changing and has numerous applications in security, accessibility and human-computer interaction. This paper proposes a model which combines Convolutional Neural Networks (CNNs) to capture spatial capabilities, Long Short-Term Memory (LSTM) networks to examine temporal dependencies, and an adaptive interest mechanism. Meticulous preprocessing of the MIRACL VC-l dataset addressing challenges including one of a kind lip moves and occlusions accompanied with the aid of transitioning this study effortlessly to LRS2 dataset to complement lexemic versatility is one of its key function. The effects verify its robustness throughout unique datasets with superior overall performance towards cutting-edge techniques. Ablation checks suggest the crucial significance of every element in phrases of improving lip analyzing accuracy. Our proposed model version additionally suggests flexibility in restricted and naturalistic language situations. © 2024 IEEE.

关键词： Attention Mechanism Convolutional Neural Networks (CNNs) Deep Learning Lip Reading Long Short-Term Memory (LSTM) LRS2 Dataset MIRACLVC-J Dataset

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：