检索结果-内蒙古大学图书馆

10th IEEE Conference on Big Data Security on Cloud, BigDataSecurity 2024

作者： Liu, Jingnian Liang, Wei Chen, Yuxiang Jin, Kai Hunan University of Science and Technology Hunan Key Laboratory for Service Computing and Novel Software Technology School of Computer Science and Engineering Xiangtang China

ISBN: (纸本)9798350389524

Medical image anomaly detection refers to machine learning techniques to analyze and identify lesions and abnormalities in them. However, in medical images, anomaly samples are usually sparse, which can lead to supervised learning not being able to obtain sufficient training. While existing unsupervised learning is mainly based on reconstruction and generation as well as feature embedding, these methods do not take into account the problem of data type bias during domain migration and often face the problem of loose judgment boundaries. To this end, this paper proposes a novel unsupervised learning method, SyNet, based on noisy anomaly synthesis. First, the problem at domain migration is well solved by aggregating the range of features within the feature map, fusing the multiscale features at the intermediate level, and adding feature adapters. Then, anomaly samples are synthesized by adding noise in the feature space, which is more efficient and stable than generating samples in the image space. Finally, normal and abnormal data are distinguished by training the discriminator. In this paper, experiments are conducted on ISIC and brain MRI datasets, and SyNet achieves nearly 20% improvement in AUC, ACC and other metrics compared with the current mainstream methods, and also has good inference efficiency. © 2024 IEEE.

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Recommending Missed Citations Identified by Reviewers: A New Task, Dataset and Baselines 30

Recommending Missed Citations Identified by Reviewers: A New...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Long, Kehan Li, Shasha Wang, Pancheng Bao, Chenlong Tang, Jintao Wang, Ting School of Computer National University of Defense Technology China Key Laboratory of Software Engineering for Complex Systems National University of Defense Technology China

ISBN: (纸本)9782493814104

Citing comprehensively and appropriately has become a challenging task with the explosive growth of scientific publications. Current citation recommendation systems aim to recommend a list of scientific papers for a given text context or a draft paper. However, none of the existing work focuses on already included citations of full papers, which are imperfect and still have much room for improvement. In the scenario of peer reviewing, it is a common phenomenon that submissions are identified as missing vital citations by reviewers. This may lead to a negative impact on the credibility and validity of the research presented. To help improve citations of full papers, we first define a novel task of Recommending Missed Citations Identified by Reviewers (RMC) and construct a corresponding expert-labeled dataset called CitationR. We conduct an extensive evaluation of several state-of-the-art methods on CitationR. Furthermore, we propose a new framework RMCNet with an Attentive Reference Encoder module mining the relevance between papers, already-made citations, and missed citations. Empirical results prove that RMC is challenging, with the proposed architecture outperforming previous methods in all metrics. We release our dataset and benchmark models to motivate future research on this challenging new task. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Paper

来源：评论

学校读者我要写书评

暂无评论

IEARACDA: Predicting circRNA-disease associations based on information enhancement and arctangent rank approximation

IEARACDA: Predicting circRNA-disease associations based on i...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Pang, Shanchen Song, Zheqi Liu, Tiyao Wu, Wenhao Zhang, Yuanyuan Wang, Shudong College of Computer Science and Technology Qingdao Institute of Software Qingdao China Qingdao University of Technology School of Information and Control Engineering Qingdao China

ISBN: (纸本)9798350386226

Circular RNAs (circRNAs) are non-coding RNA molecules that play a significant role in cell regulation and disease occurrence. In recent years, the use of computational methods to predict circRNAs associated with diseases has become a research focus. However, these methods are often constrained by the noise interference and an insufficient exploration of the underlying information in known associations. Therefore, this paper proposes an information enhancement and arctangent rank approximation based method to predict potential circRNA-disease associations (IEARACDA). Firstly, information enhancement methods are used to make circRNA and disease similarity information more complete and to reduce the sparsity of known association. Subsequently, the arctangent rank approximation method is employed to mitigate the potential bias issues that may arise from the nuclear norm, and optimization is carried out using the inexact augmented Lagrange multiplier (IALM) method. Finally, potential associations are predicted through heterogeneous graph inference method. The experimental results demonstrate that IEARACDA exhibits superior performance compared to existing methods. Furthermore, case studies for hepatocellular carcinoma and lung cancer provide additional confirmation of the accuracy and practical significance of the model predictions. © 2024 IEEE.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Time Series Data Cleaning under Expressive Constraints on Both Rows and Columns 40

Time Series Data Cleaning under Expressive Constraints on Bo...

引用

40th IEEE International Conference on Data engineering, ICDE 2024

作者： Ding, Xiaoou Li, Genglong Wang, Hongzhi Wang, Chen Song, Yichen School of Computer Science and Technology Harbin Institute of Technology China Tsinghua University National Engineering Laboratory for Big Data Software EIRI China

ISBN: (纸本)9798350317152

Time series data generated by thousands of sensors are suffering data quality problems. Traditional constraint-based techniques have greatly contributed to data cleaning applications. However, cleaning methods that support expressive constraints on time series data remain insufficient. Given the notable characteristics of time series data, existing cleaning approaches are challenged to provide good repair solutions. To address the challenges, we propose a novel data cleaning method for time series which incorporates expressive constraints that support arithmetic operations between attributes and time context. In the violation detection phase, we introduce specialized violation degree quantification functions and design a violation cell discovery algorithm to identify errors hidden in time series data. In the data repairing phase, we formalize the cleaning task as a constrained optimization problem and develop a novel repair objective function that considers both modification costs and conformance degrees of constraints. We effectively reduce the repair search space through the evaluation of time-context constraints and propose a bidirectional repairing algorithm. We also provide theoretical analysis of the proposed repairing method. Experimental results on three real-world IoT datasets across five metrics demonstrate that our proposed method outperforms seven state-of-the-art cleaning techniques specialized for time series data. Specifically, we achieve a 60% improvement in repairing effectiveness and a 70% reduction in time costs with our designed cleaning strategy. © 2024 IEEE.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

An elastic framework construction method based on task migration in edge computing

An elastic framework construction method based on task migra...

引用

作者： Pu, Yonglin Li, Ziyang Yu, Jiong Lu, Liang Guo, Binglei School of Software Nanjing University of Information Science and Technology Nanjing China School of Software Xinjiang University Urumqi China College of Computer Science and Technology Civil Aviation University of China Tianjin China School of Computer Engineering Hubei University of Arts and Science Xiangyang China

Edge computing (EC) serves as an effective technology, empowering end-users to attain high bandwidth and low latency by offloading tasks with high computational demands from mobile devices to edge servers. However, a major challenge arises when the processing load fluctuates continuously, leading to a performance bottleneck due to the inability to rescale edge node (EN) resources. To address this problem, the approach of task migration is introduced, and EN load prediction model, the resource constrained model, optimal communication overhead model, optimal task migration model, and energy consumption model are built to form a theoretical foundation from which to propose a task migration based resilient framework construction method in EC. With the aid of the domino effect and the combined effect of task migration, a dynamic node-growing algorithm (DNGA) and a dynamic node-shrinking algorithm (DNSA), both based on the task migration strategy, are proposed. Specifically, the DNGA smoothly expands the EN scale when the processing load increases, while the DNSA shrinks the EN scale when the processing load decreases. The experimental results show that for standard benchmarks deployed on an elastic framework, the proposed method realizes a smooth scaling mechanism in the EC, which reduces the latency and improves the reliability of data processing. © 2023 John Wiley & Sons, Ltd.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Dual Cross-Stage Partial Learning for Detecting Objects in Dehazed Images 24

Dual Cross-Stage Partial Learning for Detecting Objects in D...

引用

24th IEEE International Conference on Data Mining, ICDM 2024

作者： Zhao, Jinbiao Zhang, Zhao Ren, Jiahuan Zhang, Haijun Zhao, Zhongqiu Wang, Meng School of Computer Science and Information Engineering Hefei University of Technology Hefei China Yunnan Key Laboratory of Software Engineering Yunan China Harbin Institute of Technology Department of Computer Science Shenzhen China

ISBN: (纸本)9798331506681

Performing an object detection task after the restoration of a hazy image, or rather detecting with the network backbone directly, will result in the inclusion of information mixed with dehazing, which tends to interfere with detection performance. To address these issues, we propose a novel framework for detecting objects in dehazed images via Dual Cross-Stage Partial Learning (DCSP). Specifically, we introduce a Cross-Stage Partial (CSP) module for extracting clean feature information after dehazing. Secondly, to enhance data integrity, we employ a skip-input strategy to supplement information related to object detection features that may be lost during the dehazing task, while avoiding the gradient vanishing problem. In addition, CSP is also introduced to facilitate comprehensive learning of multiple feature representations. Finally, to avoid the inclusion of irrelevant dehazing information in detection, we apply a Ground-Truth Flow at detection network (at dark3), for fine feature information calibration. Additionally, we created a synthetic fog dataset to expand the training data for DCSP. Experimental results on both synthetic and real-world datasets demonstrate the effectiveness and accuracy of the proposed method. The code is available at https://***/zhaojinbiao/DCSP. © 2024 IEEE.

关键词： Demulsification

来源：评论

学校读者我要写书评

暂无评论

FCA-based θ-iceberg core decomposition in graphs

引用

Journal of Ambient Intelligence and Humanized Computing 2024年第2期15卷 1423-1428页

作者： Hao, Fei Xinchang, Khamphaphone Park, Doo-Soon Key Laboratory of Modern Teaching Technology Ministry of Education Xi’an China School of Computer Science Shaanxi Normal University Xi’an710119 China Department of Computer Science and Engineering Soonchunhyang University Asan31538 Korea Republic of Department of Computer Software Engineering Soonchunhyang University Asan31538 Korea Republic of

Complex networking analysis is a powerful technique for understanding both complex networks and big graphs in ubiquitous computing. Particularly, there are several novel metrics, such as k-clique and k-core are proposed in order to study the relative importance of nodes in complex networks. Among of those metrics, k-core analysis is an effective approach for simplifying graphical structure. However, the relation between k and the scale of networks is not explored in most existing literature. Toward this end, this paper formulate a new research problem, θ-Iceberg Core decomposition in graphs, which is able to incorporate a parameter θ (0 © Springer-Verlag GmbH Germany, part of Springer Nature 2017.

关键词： Formal concept analysis

来源：评论

学校读者我要写书评

暂无评论

EPipe: Pipeline Inference Framework with High-quality Offline Parallelism Planning for Heterogeneous Edge Devices 24

EPipe: Pipeline Inference Framework with High-quality Offlin...

引用

43rd International Conference on computer-Aided Design, ICCAD 2024

作者： Xiong, Yi Liu, Weihong Zhang, Rui Zu, Yulong Zhu, Zongwei Zhou, Xuehai School of Computer Science and Technology University of Science and Technology of China China Suzhou Institute for Advanced Research University of Science and Technology of China China School of Software Engineering University of Science and Technology of China China

ISBN: (纸本)9798400710773

Pipeline parallelism is essential for edge computing as it effectively consolidates the limited resources of edge devices, enabling the deployment of large Deep Neural Network (DNN) models and accelerating inference processes without compromising the performance of models. Accurate computation and communication latency estimation on heterogeneous edge devices is essential for searching for a superior parallelism plan. However, existing heterogeneous pipeline inference approaches either incur substantial resource wastage during online parallelism planning, as they utilize profiling strategies that occupy physical devices;or rely on cost models with inadequate representational capabilities, leading to inaccurate predictions, thereby harming the result of pipeline planning. This paper proposes EPipe, a novel pipeline inference framework that supports high-quality offline planning in heterogeneous edge environments. EPipe integrates two core components: the Task-Device Co-analyzer (TDC) and the Multi-pipeline Parallelism Planner (MPP). TDC utilizes an undirected connected graph to depict the compatibility of DNNs across device groups and precisely estimates inference and communication latencies through fine-grained modeling. Based on TDC, MPP utilizes a dynamic programming-based genetic algorithm to explore multi-pipeline solutions, extending beyond traditional single-pipeline methods. A comprehensive experimental evaluation on an edge testbed confirms the effectiveness of EPipe, demonstrating significant speedups in inference tasks for both task streams and single tasks. © 2024 Copyright is held by the owner/author(s).

关键词： Edge detection

来源：评论

学校读者我要写书评

暂无评论

RLSM: A Strategy Matrix based Microservice Splitting Method for Reinforcement Learning

RLSM: A Strategy Matrix based Microservice Splitting Method ...

引用

2024 International Conference on Intelligent Systems and Robotics, CISR 2024

作者： Liu, Yuqiang Yu, Jianyong Han, Xue Liu, Yuqi The School of Computer Science and Engineering Hunan University of Science and Technology Xiangtan China The School of Hunan Software Vocational and Technical University Xiangtan China

We introduce a novel microservice splitting approach, the Policy Matrix-based Reinforcement Learning Splitting Method (RLSM), designed to overcome the limitations of traditional service splitting schemes by providing a solution that is both fine-grained and efficient, with a strong emphasis on automation. This method utilizes dynamic link tracking and static code analysis techniques to analyze business modules, and extract entities using data flow graphs. We use these entities as agents in reinforcement learning to construct a reinforcement learning environment model, and optimize and update the value of entities by constructing a policy matrix, replacing the Q table in traditional Q-learning algorithms. Finally, we obtained a set of entity categories and used the K-means algorithm to cluster the entities of these categories, with each clustered cluster being a split individual microservice. The experimental results show that this splitting scheme not only achieves automated splitting, but also maintains a stable compliance rate of about 97% for microservices. This is enough to demonstrate that RLSM is more efficient and flexible in completing service design and evaluation. © Published under licence by IOP Publishing Ltd.

关键词： Data flow graphs

来源：评论

学校读者我要写书评

暂无评论

Deletion and Recovery Scheme of Electronic Health Records Based onMedical Certificate Blockchain

引用

computers, Materials & Continua 2023年第7期76卷 849-859页

作者： Baowei Wang Neng Wang Yuxiao Zhang Zenghui Xu Junhao Zhang School of Computer and Software Nanjing University of Information Science and TechnologyNanjing210044JiangsuChina Engineering Research Center of Digital Forensics Ministry of EducationNanjing210044JiangsuChina

The trusted sharing of Electronic Health Records(EHRs)can realize the efficient use of medical data *** speaking,EHRs are widely used in blockchain-based medical data *** are valuable private assets of patients,and the ownership belongs to *** recent research has shown that patients can freely and effectively delete the EHRs stored in hospitals,it does not address the challenge of record sharing when patients revisit *** order to solve this problem,this paper proposes a deletion and recovery scheme of EHRs based on Medical Certificate *** paper uses cross-chain technology to connect the Medical Certificate Blockchain and the Hospital Blockchain to real-ize the recovery of deleted *** the same time,this paper uses the Medical Certificate Blockchain and the InterPlanetary File System(IPFS)to store Personal Health Records,which are generated by patients visiting different medical *** addition,this paper also combines digital watermarking technology to ensure the authenticity of the restored electronic medical *** the combined effect of blockchain technology and digital watermarking,our proposal will not be affected by any other rights throughout the *** analysis and security analysis illustrate the completeness and feasibility of the scheme.

关键词： Electronic health records cross-chain medical certificate blockchain data deletion and recovery

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：