检索结果-内蒙古大学图书馆

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Yu Zhang Jianqiang Zhang Gongpeng Song Qin Lu Shuo Zhao Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Shandong Branch of China Mobile Communication Group Design Institute Co. Jinan China

ISBN: (数字)9798331515966

ISBN: (纸本)9798331515973

Image-text matching is an important problem at the intersection of computer vision and natural language processing. It aims to establish the semantic link between image and text to achieve high-quality semantic alignment between the two modalities. However, the existing methods have the problem that the meaning expressed in the image or the complex narrative in the text cannot be fully understood due to insufficient feature extraction. Moreover, due to the essential modal differences between images and texts, how to effectively and accurately align the semantic contents in images and texts has become the key of research. In order to solve the above problems, this paper proposes a method based on feature enhancement and relationship interaction. When processing images, the proposed method fuses labeled features, region features and location features to represent images. When processing text, a combination of Bi-GRU and self-attention mechanism is used to represent the text. In order to further align the semantic content in images and texts accurately, this paper improves two relational interaction mechanisms by identifying connection relationships and learning association relationships. Thus, the relation enhanced embedding is obtained. Finally, it calculated the similarity of the enhanced embedding to judge the matching degree of the image and text. Extensive experiments on the public datasets Flickr30K and MSCOCO demonstrate the effectiveness of our method.

关键词： computer vision Fuses Image matching Semantics Feature extraction Natural language processing

来源：评论

学校读者我要写书评

暂无评论

An ontology-supported inquiry learning technique

Journal of Network Intelligence

引用

Journal of Network Intelligence 2021年第1期6卷 16-28页

作者： Xue, Xingsi Zhang, Jie Yang, Chaofan Fujian Provincial Key Laboratory of Big Data Mining and Applications Fujian University of Technology No.33 Xuefu South Road University Town Minhou Fuzhou Fujian350118 China School of Computer Science and Engineering Yulin Normal University No.299 Education Middle Road Yulin City Guanxi Province537000 China School of Computer Science and Mathematics Intelligent Information Processing Research Center Fujian University of Technology No.33 Xueyuan Road University Town Minhou Fuzhou Fujian350118 China Guangxi Key Laboratory of Automatic Detecting Technology and Instruments Guilin University of Electronic Technology No.1 Jinji Road Guilin Guangxi541004 China

Although Inquiry-based Learning (IBL) offers compelling opportunities for the engineering education, one of the challenges to the implementation of it is that the learner always lacks of the background knowledge. To provide him with the necessary knowledge when making the inquiry, this paper proposes an Ontology Supported Inquiry Learning technique (OSIL), which works on the basis of two ontologies, i.e. curriculum ontology and the learner’s ontology. The experimental results show that the constructed curriculum ontology and learner ontology is of high-quality, and the answers recommended by OSIL almost perfectly match with the true answers. From the experimental results, we can see that it is a feasible way of using the software engineering and prototype evolution method to construct the curriculum and learner ontology, and the individual inquiry based on two ontologies is able to provide the learners with high-quality personalized services. © 2021, Taiwan Ubiquitous information. All rights reserved.

关键词： Curricula

来源：评论

学校读者我要写书评

暂无评论

A Prediction Method for Atmospheric Monitoring Data Based on Transformer and GAN

A Prediction Method for Atmospheric Monitoring Data Based on...

引用

IEEE information technology, Networking, Electronic and Automation Control Conference

作者： Hongge Zhao Jingli Wang Tixiang Zhang Huijuan Hao Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Jinan Eco-environmental Bureau Pingyin Branch Office Jinan China Shandong Unite Environmental Technology Co Jinan China

ISBN: (数字)9798350370805

ISBN: (纸本)9798350370812

With the rapid development of urbanization and industrialization, atmospheric pollution has become a major environmental problem of global concern. The problem of atmospheric pollution not only destroys the ecological en-vironment of the earth, but also has a serious impact on people's health and development. Therefore, accurate prediction of atmospheric monitoring data is of great significance to environmental management as well as public health. To improve the accuracy of prediction, we propose a prediction method for atmospheric monitoring data based on Trans-former and Generative Adversarial Network (GAN). Firstly, data pre-processing is performed on the collected atmospheric monitoring data, and then, constructing a prediction model based on Transformer and GAN for atmospheric monitoring data, aiming at extracting the data features of atmospheric monitoring data by Transformer and generating the data by using the generator of GAN. In addition, we improve the position encoding (PE) of the Transformer in order to it can efficiently extract temporal features of atmospheric monitoring data. Finally, the effectiveness of the proposed method is verified by data from three air monitoring stations in a city.

关键词： Accuracy Pollution Atmospheric modeling Urban areas Predictive models Generative adversarial networks Transformers Feature extraction Data models Monitoring

来源：评论

学校读者我要写书评

暂无评论

Pixel-Level Polyp Segmentation Network Based on Parallel Feature Enhancement and Attention Mechanism

SSRN

引用

SSRN 2024年

作者： Xu, Haiping Wang, Jie Li, Zuoyong Cheng, Xuesong Teng, Shenghua College of Computer and Data Science Minjiang University Fuzhou China School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu China College of Electronic and Information Engineering Shandong University of Science and Technology Qingdao China Fujian Key Laboratory of Medical Big Data Engineering Fujian Provincial Hospital Fuzhou China

Polyp segmentation, a crucial tool in medical diagnostics, effectively locates and segments polyps, thereby aiding clinicians in enhancing both detection efficiency and accuracy. To address the persistent challenges of imprecise polyp localization and ambiguous boundary segmentation, we propose a pixel-level polyp segmentation network based on a parallel feature enhancement and attention mechanism (PFEA-Net). The proposed PFEA-Net follows an enconder-decoder architecture. Given that polyps vary greatly in size and shape, PFEA-Net first utilizes the HarDNet68 as the encoder to extract multi-level feature representations. Then, a parallel feature enhancement module is designed to enrich the feature map of each level by integrating semantic information derived from higher-level features. Meanwhile, a parallel partial decoder module is used to introduce global contextual information. Finally, the reverse attention module refines the boundary features. PFEA-Net effectively extracts high-level representations to enhance feature depiction, capturing both detailed and semantic information in the images. Rigorous experiments conducted on five publicly available polyp datasets using six evaluation metrics conclusively demonstrate that the proposed network exhibits superior robustness and achieves more precise segmentation. Compared to other state-of-the-art algorithms, PFEA-Net distinguishes itself with exceptional performance. © 2024, The Authors. All rights reserved.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

Hybridizing infeasibility driven and constrained-domination principle with MOEA/D for constrained multiobjective evolutionary optimization

Lecture Notes in Computer Science (including subseries Lectu...

引用

Lecture Notes in computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 2014年 8916卷 249-261页

作者： Lin, Huibiao Fan, Zhun Cai, Xinye Li, Wenji Wang, Sheng Li, Jian Zhang, Chengdian School of Engineering Shantou University Guangdong515063 China College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics NanjingJiangsu210016 China College of Science Shantou University Guangdong515063 China Guangdong Provincial Key Laboratory of Digital Signal and Image Processing Techniques Shantou University Guangdong515063 China

This paper presents a novel multiobjective constraint handling approach, named as MOEA/D-CDP-ID, to tackle constrained optimization problems. In the proposed method, two mechanisms, namely infeasibility driven (ID) and constrained-domination principle (CDP) are embedded into a prominent multiobjective evolutionary algorithm called MOEA/D. Constraineddomination principle defined a domination relation of two solutions in constraint handling problem. Infeasibility driven preserves a proportion of marginally infeasible solutions to join the searching process to evolve offspring. Such a strategy allows the algorithm to approach the constraint boundary from both the feasible and infeasible side of the search space, thus resulting in gaining a Pareto solution set with better distribution and convergence. The efficiency and effectiveness of the proposed approach are tested on several well-known benchmark test functions. In addition, the proposed MOEA/D-CDP-ID is applied to a real world application, namely design optimization of the two-stage planetary gear transmission system. Experimental results suggest that MOEA/D-CDP-ID can outperform other state-of-the-art algorithms for constrained multiobjective evolutionary optimization. © Springer International Publishing Switzerland 2014.

关键词： Constraint handling

来源：评论

学校读者我要写书评

暂无评论

Alzheimer's disease distinction based on gait feature analysis 22

Alzheimer's disease distinction based on gait feature analys...

引用

22nd IEEE International Conference on E-Health Networking, Application and Services, HEALTHCOM 2020

作者： You, Zhiyang You, Zeng Li, Yilong Zhao, Shipeng Ren, Huixia Hu, Xiping Gansu Provincial Key Laboratory of Wearable Computing School of Information Science and Engineering Lanzhou University Lanzhou China Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China University of Wisconsin-Madison Computer Sciences Department Madison United States Department of Neurology Shenzhen People's Hospital The 2nd Clinical Medical College of Jinan University Shenzhen China Clinical Medicine Postdoctoral Research Station Jinan University Guangzhou China

ISBN: (纸本)9781728162676

Alzheimer's disease(AD) is a neurodegenerative disease that progresses slowly but worsens gradually, also, the most common kinds of dementia. Clinically, the diagnosis of AD is mainly based on rating scales and neuroimaging technology which is invasive, costly and time-consuming. Other than that, the clinical pathology has become irreversible when neuroimaging characteristics appear. It is imperative to develop new noninvasive methods for early diagnosis of AD. Several studies indicated the probable association of cognitive decline with gait changes might shed light on potential features for distinction of AD. This paper aims to exploit the feasibility of gait features for early diagnosis of mild cognitive impairment(MCI) and AD by using machine learning methods. A device-free AD detection system is built, with a natural undisturbed gait collecting system and a well-performed Long Short-Term Memory(LSTM) based model, in this article. Moreover, it can serve as a simplified, non-invasive, and highly accurate clinical auxiliary tool for early diagnosis and distinction of AD. Experimental results showed a 90.48%, 92.00%, and 88.24% in accuracy, sensitivity, and specificity respectively for distinguishing AD by using the method with LSTM based model. Furthermore, the gait cycle and stride length in MCI or AD were more variable than in healthy controls through redefining and calculating the gait features with skeleton data obtained by Kinect devices. © 2021 IEEE.

关键词： Neuroimaging

来源：评论

学校读者我要写书评

暂无评论

A new formal modeling method for web service composition

International Journal of Signal Processing, Image Processing...

引用

International Journal of Signal processing, Image processing and Pattern Recognition 2015年第7期8卷 341-350页

作者： Li, Lin Wang, Cheng College of Computer Science and Technology Wuhan University of Science and Technology WuhanHubei China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System WuhanHubei China Department of Common Required Courses Hubei Institute of Fine Arts China

With the rapid development of e-commerce, the application of web service composition is becoming increasingly important in many areas. If we apply unchecked web service composition, then we can only discover problems afterwards, and we have to devote unnecessary energy to repair. Here, formal modeling serves to be a premise to check. This paper brings up a formal modeling method for web service composition based on BDL (behavioral description language). The structure of this essay is to analyze the description of BPEL process, and we can get the formal model from the BPEL-to-BDL mapping relation. © 2015 SERSC.

关键词： Modeling languages

来源：评论

学校读者我要写书评

暂无评论

SCTAR: A Multi-Layer BiLSTM-Based Chinese Short Text Similarity Computation Model with Attention Mechanism

SCTAR: A Multi-Layer BiLSTM-Based Chinese Short Text Similar...

引用

International Conference on computer Supported Cooperative Work in Design

作者： Shuo Zhao Yiyun Xing Jianqiang Zhang Gongpeng Song Qin Lu Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Engineering Research Center of Big Data Applied Technology Faculty of Computer Science and Technology Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computer Networks Shandong Fundamental Research Center for Computer Science Jinan China Shandong Branch of China Mobile Communication Group Design Institute Co. Jinan China

ISBN: (数字)9798350349184

ISBN: (纸本)9798350349191

Text semantic similarity is a crucial research area in Natural Language processing (NLP). However, traditional methods for calculating the similarity of short Chinese texts often fall short in accuracy and other crucial aspects. The Bidirectional Long Short-Term Memory Network (BiLSTM) has demonstrated remarkable performance in computing the similarity of short Chinese texts by effectively capturing long-range dependencies and semantic information. Additionally, the multi-head attention mechanism considers the interaction between different text locations and semantic information, enhancing the model’s representative capacity. Building upon this foundation, our paper proposes an enhanced model known as SCTbilstmAttRdrop (SCTAR). This model is constructed based on a multilayer BiLSTM architecture and incorporates an SE-gated convolutional module and a convolutional multi-head attention-aware model. We conducted extensive evaluations using two Chinese short text datasets, Chinese-SNLI and CCKS2018_Task3. The experimental results unequivocally demonstrate that our SCTAR model surpasses other common methods in terms of accuracy, precision, recall, and F1 scores when tasked with computing the similarity of Chinese short texts.

关键词： Accuracy Attention mechanisms Computational modeling Semantics Bidirectional control Feature extraction Nonhomogeneous media

来源：评论

学校读者我要写书评

暂无评论

EQ-Net: Elastic Quantization Neural Networks

arXiv

引用

arXiv 2023年

作者： Xu, Ke Han, Lei Tian, Ye Yang, Shangshang Zhang, Xingyi School of Artificial Intelligence Anhui University Hefei China Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education Hefei China Institutes of Physical Science and Information Technology Anhui University Hefei China School of Computer Science and Technology Anhui University Hefei China

Current model quantization methods have shown their promising capability in reducing storage space and computation complexity. However, due to the diversity of quantization forms supported by different hardware, one limitation of existing solutions is that usually require repeated optimization for different scenarios. How to construct a model with flexible quantization forms has been less studied. In this paper, we explore a one-shot network quantization regime, named Elastic Quantization Neural Networks (EQ-Net), which aims to train a robust weight-sharing quantization supernet. First of all, we propose an elastic quantization space (including elastic bit-width, granularity, and symmetry) to adapt to various mainstream quantitative forms. Secondly, we propose the Weight Distribution Regularization Loss (WDR-Loss) and Group Progressive Guidance Loss (GPG-Loss) to bridge the inconsistency of the distribution for weights and output logits in the elastic quantization space gap. Lastly, we incorporate genetic algorithms and the proposed Conditional Quantization-Aware Accuracy Predictor (CQAP) as an estimator to quickly search mixed-precision quantized neural networks in supernet. Extensive experiments demonstrate that our EQ-Net is close to or even better than its static counterparts as well as state-of-the-art robust bit-width methods. Code can be available at https://***/xuke225/EQNet. Copyright © 2023, The Authors. All rights reserved.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

From Data Integrity to Global Model Integrity for Federated Learning: An MHT-based Approach

From Data Integrity to Global Model Integrity for Federated ...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Nannan Zhang Yao Zhao Y. Neil Qu Bruce Gu Keshav Sood Longxiang Gao Shui Yu Key Laboratory of Computing Power Network and Information Security Ministry of Education Shandong Computer Science Center (National Supercomputer Center in Jinan) Qilu University of Technology (Shandong Academy of Sciences) Jinan China Shandong Provincial Key Laboratory of Computing Power Internet and Service Computing Shandong Fundamental Research Center for Computer Science Jinan China School of IT Deakin University VIC Australia School of Computer Science University of Technology Sydney NSW Australia

ISBN: (数字)9798350351255

ISBN: (纸本)9798350351262

Federated Learning (FL) is a distributed machine learning (ML) approach that enables multiple edge nodes to collaboratively train ML models by sharing model parameters, thus addressing privacy concerns. However, in highly distributed, dynamic, and volatile FL environments, the global model is vulnerable to various corruptions. For instance, edge nodes might falsely claim that the received global model is incomplete, or the channel that transmits the global model is untrustworthy. Effectively verifying the integrity of the global model poses a critical challenge. To tackle this issue, we introduce a method for verifying model integrity called Federated learning global Model Integrity Verification (FMIV). It leverages Merkle Hash Tree (MHT) to generate integrity proofs of the global model during verification. To improve security, we integrate random security codes during proof generation. FMIV is capable of verifying the global model updated by the central server and shared with untrusted edge nodes, while efficiently identifying the edge node caching the incomplete global model. Furthermore, we conduct theoretical analysis and extensive experiments to validate the performance of FMIV. Compared to the two state-of-the-art approaches, FMIV consistently exhibits a notable improvement in verification efficiency and effectiveness in detecting model corruption.

关键词： Analytical models Privacy Codes Federated learning Computational modeling Image edge detection Data integrity Data models Security Servers

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：