检索结果-内蒙古大学图书馆

International Conference on Advances in Electronics, Computers and Communications (ICAECC)

作者： Bharathi Mohan G R Prasanna Kumar Elakkiya R V Druva Veer Computer Science and Engineering Amrita Vishwa Vidyapeetham School of Computing Chennai Computer Science and Engineering Birla Institute of Technology and Science School of Computing Dubai UAE

ISBN: (数字)9798350305449

ISBN: (纸本)9798350305456

Duplicate question detection is a crucial task in question-and-answer platforms, aiming to improve user experience and minimize redundancy. This project utilizes NLP techniques and machine learning models to predict whether a pair of questions are duplicates accurately. The project involves comprehensive data pre-processing, including tokenization, stopword removal, and lemmatization, to prepare the textual data for analysis. Here we are using various feature engineering techniques, such as bag-of-words, TF-IDF, and word embedding, are applied to extract meaningful representations of the questions. Multiple machine learning models, including Logistic Regression, Random Forest, SVM, XGBoost, and a Combined model, are trained on labeled data. Evaluation metrics focus on precision and recall minimizing misclassification costs. New sample data is used to validate the models. Predictions are made to classify question pairs as duplicates or non-duplicates, showcasing the model's generalization capabilities. The results highlight the potential of NLP techniques and machine learning models in accurately identifying duplicate question pairs, improving knowledge sharing, and reducing redundancy. This project provides valuable insights and methodologies for future research in duplicate question detection, contributing to the field of NLP and enhancing user experiences on question-and-answer platforms.

关键词： Support vector machines Costs Redundancy Predictive models Feature extraction Data models User experience

来源：评论

学校读者我要写书评

暂无评论

Enhancing Medical Image Security: A Deep Learning Approach with Cloud-based Color Space Scrambling

引用

International Journal of Information Technology (Singapore) 2024年第8期16卷 5041-5054页

作者： Cherian, Aswathy K. Simpson, Serin V. Vaidhehi, M. Marimuthu, Ramaprabha Shankar, M. Department of Computing Technologies SRM Institute of Science and Technology Kattankulathur Chennai India School of Computer Science and Engineering Presidency University Bangalore India Veltech Dr. Rangarajan Dr. Sagunthala Research Institute of Science and Technology Avadi Chennai India Department of Computer Science and Technology Madanapalle Institute of Technology and Science Angallu Andhra Pradesh Madanapalle India

Progress in wisdom medicine has been driven by advancements in big data, cloud computing, and artificial intelligence, enabling the accumulation of valuable information and insights. However, the increasing reliance on cloud-based storage and transmission of medical images has raised significant concerns regarding information security. The risk of unauthorized access to patients' private data poses a considerable obstacle to medical research advancement. Thus, safeguarding patient data in cloud environments is imperative. Color space-based scrambling algorithms (CSSA) are gaining traction for multimedia data encryption due to their compatibility with JPEG and reduced processing requirements. However, traditional CSSA methods rely on colorful images to optimize security, limiting their applicability in fields like medical image processing where color images may be scarce. This study seeks to integrate CSSA image encryption with Multilayer Perceptron (MLP)-based techniques for securing medical images. Additionally, a noise-based data augmentation method is developed to address data scarcity in medical image analysis. Security analysis and temporal complexity assessments are employed to evaluate the effectiveness of the proposed MLP-CSSA deep learning model in encrypting medical images. Results demonstrate robust security in encrypting both grayscale and color medical images, with the proposed MLP-CSSA method outperforming existing encryption techniques. © Bharati Vidyapeeth's Institute of Computer Applications and Management 2024.

关键词： Colour space-based scrambling algorithms (CSSA) Deep learning Encryption techniques JPEG standard Medical images Multilayer Perceptron’s (MLP)

来源：评论

学校读者我要写书评

暂无评论

Securing Multi-Tenant Cloud Environments with Fully Homomorphically Encrypted Secure Multiparty Computation

Securing Multi-Tenant Cloud Environments with Fully Homomorp...

引用

2024 IEEE International Conference on Recent Advances in science and engineering Technology, ICRASET 2024

作者： Ponmalar, A. Pandiarajan, Rajkumar Sudha, I. Ramesh, P.S. Jagannathan, J. R.M.K. Engineering College Tamilnadu Chennai India School of Computing Kalasalingam Academy Of Research and Education Department of Information and Technology Krishnankoil Virudhunagar India Saveetha Institute of Medical and Technical Sciences Simats University Saveetha School of Engineering Department of Computer Science and Engineering Tamil Nadu Chennai India Vel Tech Rangarajan Dr.Sagunthala R&d Institute of Science and Technology Department of Computer Science and Engineering Tamil Nadu Chennai India Vellore Institute of Technology School of Computer Science Engineering and Information Systems Department of Software and Systems Engineering Katpadi Vellore India

ISBN: (纸本)9798350388602

The aim of this work is to examine the application of FHE-SMPC as a viable security model for protecting the tenant isolation in multi-tenant cloud model. In multi-tenant environments, where various customers use the same cloud system, controlling the data leakage and protecting privacy is a major issue. Actual safeguards fail to meet the needs of data protection during processing, therefore, the introduction of highly elaborate encryption processes must happen. This research work focuses on the use of FHE technique and SMPC such that data can undergo processing while the plaintext remains unknown to any party. A kind of approach is based on the acquisition of synthetic and real data sets, system-level architecture, and cryptographic algorithms using open-source libraries with cloud service providers. The results of such implementation indicate that despite the fact that the FHE-SMPC system has a considerable computational overhead in comparison with classical schemes it is very secure and protected against different types of cryptographic attacks. The problem of scalability is still an open issue, but the complex is quite efficient with various data types that is why it can be used in various cloud applications. This paper offers a pertinent solution for security of data that is housed in multitenant environment and will go along way in creating fertile ground for research and innovation in secure clouding. © 2024 IEEE.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

1,2-Norm and CUR Decomposition based Sparse Online Active Learning for Data Streams with Streaming Features

1,2-Norm and CUR Decomposition based Sparse Online Active Le...

引用

2024 IEEE International Conference on Big Data, BigData 2024

作者： Chen, Zhong He, Yi Wu, Di Zuo, Liudong Li, Keren Zhang, Wenbin Deng, Zhiqiang Southern Illinois University School of Computing United States William & Mary School of Data Science United States Southwest University College of Computer and Information Science United States California State University Dominguez Hills Computer Science Department United States University of Alabama at Birmingham Department of Mathematics United States Florida International University School of Computing and Information Sciences United States Louisiana State University Department of Civil and Environmental Engineering United States

ISBN: (纸本)9798350362480

Aiming at learning from a sequence of data instances over time, online learning has attracted increasing attention in the big data era. As two important variants, sparse online learning has been extensively explored by facilitating sparse constraints for online models such as truncated gradient, 1-norm regularization, 1-ball projection, and regularized dual averaging;while online active learning aims to build an online prediction model with a limited number of labeled instances, deploying the so called query strategies to select informative instances over time. However, most existing studies consider sparse online learning or online active learning with fixed feature spaces, whereby in real practice the features may be dynamically evolved over time. To the end, we propose a novel unified one-pass online learning framework named OASF for simultaneously online active learning and sparse online learning tailored for data streams described by open feature spaces, where new features can emerge constantly, and old features may be vanished over various time spans. Specifically, we technically develop an effective online CUR matrix decomposition based on the 1,2 mixed norm constraint for simultaneously selecting important up-to-date samples in a sliding window and facilitating stable and meaningful features in open feature spaces over time. If the loss function is simultaneously Lipschitz and convex, a sub-linear regret bound of our proposed algorithm is guaranteed with. Extensive experiments that are conducted with multiple streaming datasets have demonstrated the effectiveness of the proposed OASF compared with state-of-the-art online active learning and sparse online learning methods. © 2024 IEEE.

关键词： Active learning

来源：评论

学校读者我要写书评

暂无评论

Towards Integrated Spatial Crowdsourcing: Online Privacy-Preserving Selection

引用

IEEE Transactions on Dependable and Secure computing 2025年

作者： Wang, Hengzhi Ma, Minghe Cui, Laizhong Liu, Jiangchuan Shenzhen University College of Computer Science and Software Engineering Shenzhen China Simon Fraser University School of Computing Science Burnaby Canada

We study an intriguing and practical scenario of online Spatial Crowdsourcing (SC), in which workers have the flexibility to perform tasks using various methods, such as walking, driving, or utilizing Unmanned Aerial Vehicles (UAVs). This results in workers having heterogeneous, arbitrary, and non-stationary utilities over time. We refer to this scenario as integrated SC. Unfortunately, existing studies are limited in addressing integrated SC settings due to two aspects: (1) these studies are based on the assumption that workers' utilities are independently and identically distributed and follow a stationary distribution like Gaussian, which does not hold in integrated SC;(2) their approaches fail to provide personalized privacy preservation for different workers. Motivated by these limitations, we closely investigate the heterogeneous utility and personalized privacy requirement in integrated SC and propose an Online Personalized Privacy-preserving Selection framework (OPPS). In this framework, we present an online selection policy that balances the exploration-exploitation trade-off given heterogeneous utilities and develop a built-in privacy policy that ensures differential privacy guarantee. We then demonstrate that our framework effectively addresses the trade-off by deriving a sublinear, privacy-related upper bound on regret that scales as O(√T). Extensive numerical simulations based on real-world drone datasets are conducted to validate the effectiveness of our framework compared with state-of-the-art approaches. © 2004-2012 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

FRNet: A Full-Resolution Convolutional Neural Network for OCTA Vascular Segmentation 30

FRNet: A Full-Resolution Convolutional Neural Network for OC...

引用

30th International Conference on Mechatronics and Machine Vision in Practice, M2VIP 2024

作者： Wang, Liang Gao, Dongxu Gao, Hongwei Jiang, Guozhang Ju, Zhaojie Shenyang Ligong University School of Automation and Electrical Engineering Shenyang China University of Portsmouth School of Computing United Kingdom Wuhan University of Science and Technology Key Laboratory of Metallurgical Equipment and Control Technology Ministry of Education Wuhan China

ISBN: (纸本)9798350391916

Optical coherence tomography angiography (OCTA) is an advanced, non-invasive imaging technique. Automatic segmentation of vascular networks in retinal OCTA images is essential for the early diagnosis and progression assessment of various vision-related diseases. However, most existing methods for OCTA image vessel segmentation rely on encoder-decoder architectures, which typically involve a high number of parameters and result in slower inference speeds. In this paper, we introduce FRNet V2, an accurate and efficient neural network specifically designed for retinal vessel segmentation in OCTA images. This is achieved by integrating a modified recursive ConvNeXt V2 block and a recursive HAAM attention mechanism into a full-resolution convolutional network framework. We evaluate our proposed method on two large public datasets: ROSSA and OCTA-500. Experimental results demonstrate that our network achieves performance on par with other methods, while boasting significantly fewer parameters and faster inference speeds. This makes it highly suitable for practical industrial applications. © 2024 IEEE.

关键词： Optical coherence tomography

来源：评论

学校读者我要写书评

暂无评论

A Weighted Ensemble Causal Discovery Method for Effective Connectivity Estimation 9

A Weighted Ensemble Causal Discovery Method for Effective Co...

引用

9th IEEE Smart World Congress, SWC 2023

作者： Zhang, Qiqi Zhang, Yingwei Ding, Yanhui Chen, Yiqiang Song, Shuchao Wu, Shuang Shandong Normal University School of Information Science and Engineering Jinan China Shandong Academy of Intelligent Computing Technology Jinan China Chinese Academy of Sciences Institute of Computing Technology Beijing China Beijing Key Laboratory of Mobile Computing and Pervasive Device Beijing China

ISBN: (纸本)9798350319804

Exploring and explaining the effective connectivity (EC) between brain regions can help us understand the mechanisms behind neurodegenerative diseases such as Alzheimer's disease, thus helping us to diagnose patients better. Causal discovery techniques have been widely used in this domain recently, but diverse methods may produce conflicting outcomes because of the differences in underlying principles and search methods. To solve this problem, we integrate the causal relationships generated by different algorithms to obtain high-confidence inferences, resulting in a more reliable weighted ensemble causal discovery (WECD) method with higher accuracy. First of all, WECD defines a distribution metric function to measure the weights of different individual methods. Then, WECD adaptively chooses the critical threshold to obtain the confident edges set based on the generated directed edge confident metric curve. Finally, the directed acyclic graph condition is used to refine the confident edges set and acquire the final causal graph. We validate the effectiveness of WECD with both the simulated dataset and real functional near-infrared spectroscopy (fNIRS) dataset collected on AD patients, MCI patients and healthy controls. By comparing with other existing methods, experimental results demonstrate the potential of WECD as an effective tool for estimating brain EC. © 2023 IEEE.

关键词： Alzheimer's disease causal discovery effective connectivity ensemble learning functional near-infrared spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Enhancing Military Surveillance: Advanced Aircraft Image Identification using Hybrid Deep Neural Networks 2

Enhancing Military Surveillance: Advanced Aircraft Image Ide...

引用

2nd International Conference on Automation, computing and Renewable Systems, ICACRS 2023

作者： Trivedi, Naresh Kumar Maheshwari, Himani Tiwari, Raj Gaurang Agarwal, Ambuj Kumar Gautam, Vinay Chitkara University Institute of Engineering and Technology Chitkara University Punjab India School of Computing Graphic Era Hill University Uttrakhand Dehradun India Department of Computer Science and Engineering Sharda School of Engineering and Technology Sharda University Greater Noida India

ISBN: (纸本)9798350340235

Machine learning is a powerful tool that may significantly improve the judgment of businesses and individuals alike. The use of machine learning methods in data analytics is now commonplace, especially with the development of data science. Identifying enemy fighter planes is crucial for military applications, as it plays a role in operational planning. The challenge is in correctly determining the orientation of the unknown airplane. The 30 different types of aircraft in this database were retrieved from the Kaggle repository. An 80:20 or 70:30 split of the image database was used to create a training set and test set, respectively. The normalized feature sets allowed the LSTM, CNN, BiLSTM, ANN, Autoencoder, and Hybrid Model deep learning models to be developed and trained to their full potential. The test aircraft image was fed into the hybrid deep neural networks so that they could be categorized. The developed automatic aircraft image identification system achieved a classification accuracy of 98.85%. © 2023 IEEE.

关键词： Aircraft Deep Learning Image Classification MachineLearning

来源：评论

学校读者我要写书评

暂无评论

Automated Detection of Violence in Detached Areas using Hybrid Deep Learning Models: A YOLO-5 and CNN Approach 2

Automated Detection of Violence in Detached Areas using Hybr...

引用

2nd International Conference on Automation, computing and Renewable Systems, ICACRS 2023

作者： Gautam, Vinay Maheshwari, Himani Tiwari, Raj Gaurang Agarwal, Ambuj Kumar Trivedi, Naresh Kumar Chitkara University Institute of Engineering and Technology Chitkara University Punjab India School of Computing Graphic Era Hill University Uttrakhand Dehradun India Department of Computer Science and Engineering Sharda School of Engineering and Technology Sharda University Greater Noida India

ISBN: (纸本)9798350340235

The human violence rate increases gradually every year and it's increasing at a rate of 7% annually. It has been observed in many studies that violence takes place suddenly in detached areas. The major hurdle to stopping these acts is a delay in information exchange. To tackle these issues, automated detection techniques were employed. CCTV footage was employed to identify objects with computer vision terminology and actions were taken based on correctly identified objects. These CCTV cameras are extremely helpful to resolve such cases and take appropriate action. The violence and non-violent acts were identified based on the analysis of footage. Here, in this research, an effective and appropriate method was employed to detect the violent and non-violent acts from the CCTV footage. In this context, a deep learning model was proposed in the research that comprises two deep learning models especially YOLO-5 and Convolutional neural network (CNN). Here, initially, the images were preprocessed to enhance the quality of CCTV footage and afterward, YOLO was employed to identify objects from the available CCTV footage and later, the classification was done using CNN. The performance hybrid model was evaluated on different parameters and compared with other state-of-the-art techniques. After comparison, it has been observed that the proposed model outperformed with an accuracy rate of 98.63%. © 2023 IEEE.

关键词： Closed-Circuit Television (CCTV) Computer Vision Computer Vision Deep Learning Deep Neural Networks Human Violence Machine Learning Object Detection

来源：评论

学校读者我要写书评

暂无评论

A Meta Distribution-Based Fine-Grained Analysis for Contention-based WiFi Backscatter Networks 21

A Meta Distribution-Based Fine-Grained Analysis for Contenti...

引用

21st International Conference on Networking, Sensing and Control, ICNSC 2024

作者： Wang, Yulei Zhao, Qinglin Feng, Li Zhou, MengChu Shen, Meng Luo, Yu Sun, Yi School of Computer Science and Engineering Macau University of Science and Technology 999078 China College of Electronics and Information Engineering South-Central Minzu University Wuhan430074 China New Jersey Institute of Technology Department of Electrical and Computer Engineering NewarkNJ07102 United States School of Cyberspace Science and Technology Beijing Institute of Technology Beijing100081 China School of Computer Science Guangdong University of Technology Guangzhou510006 China Institute of Computing Technology Chinese Academy of Sciences Beijing100045 China

ISBN: (纸本)9798350365221

WiFi backscatter communication has gained many applications, but its performance characteristics remain to be analyzed. While existing research has investigated the success probability of backscatter tags in contention-based WiFi backscatter networks (CWBNs), it has focused solely on the first-order statistic of the signal-to-interference-plus-noise ratio (SINR). In this paper, we present a meta distribution-based fine-grained analysis that provides high-order statistics of SINR and characterizes the disparity among backscatter transmission links in CWBNs. Leveraging stochastic geometry, we, for the first time, derive mathematical expressions for the b-th moments of conditional success probability and its meta distribution. The extensive Monte-Carlo simulation results validate the accuracy of our proposed theoretical model and demonstrate its outstanding value to help us understand the overall performance of CWBNs. © 2024 IEEE.

关键词： Wi-Fi

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：