检索结果-内蒙古大学图书馆

EDRLNet: privacy preservation in online social networks via deep reinforcement learning and graph convolutional networks

引用

International Journal of Information Technology (Singapore) 2025年 1-10页

作者： Gowda, Nithish Ranjan Venkatesh Venugopal, K.R. Department of Computer Science and Engineering University Visvesvaraya College of Engineering Bangalore University Bengaluru India Associate Professor Department Computer Science and Engineering University Visvesvaraya College of Engineering Bangalore University Bangalore India University Visvesvaraya College of Engineering Bangalore University Bangalore India

Online social networks (OSN) have become extremely popular in the past few decades. Prominent OSN companies, such as Facebook and Twitter, control a huge amount of information on its users as well as interactions that may cause people to have major privacy concerns. The OSN information is combined with other online sources to reveal sensitive information. As a result, there is a great need to anonymize the OSN data before providing it to a third party. Thus, in this study, we propose a Deep Reinforcement Learning (DRL)-based rewiring operation that entails changing the relationships (edges) between nodes (users) without changing their degree to protect users’ privacy. To accomplish this, we introduce a novel technique for Enhanced DRL Networks called “EDRLNet” which combines an Optimized Policy Network (OPN) and Graph Convolutional Network (GCN) that preserves the essential characteristics of OSN data while anonymizing it to prevent privacy breaches. This approach employs a GCN to extract embeddings (features) from the edges and nodes. Furthermore, OPN applies the graph feature representation learned by the GCN model to policy learning. Additionally, the hyperparameters of OPN are optimized by using the Addax optimization algorithm (AOA). The suggested scheme’s performance is assessed using PageRank with mean-squared error (MSE), Root mean-squared error (RMSE), mean absolute error (MAE), Privacy score, Utility Score, Clustering coefficient, privacy and information loss. © Bharati Vidyapeeth's Institute of computer Applications and Management 2025.

关键词： Addax optimization algorithm Anonymization Deep reinforcement learning Graph convolutional network Online social network Privacy preservation

来源：评论

学校读者我要写书评

暂无评论

Energy Efficient Routing for Underwater Sensor Network Using Eco-Aware Routing Protocol in Comparison with opportunistic Routing Protocol 2

Energy Efficient Routing for Underwater Sensor Network Using...

引用

2nd IEEE International Conference on Emerging Research in Computational science, ICERCS 2024

作者： Ayyadurai, M. Kp, Barath Nivash Computer Science and Engineering Rajalakshmi Engineering College Chennai India

ISBN: (纸本)9798331534950

Under Water Sensor Networks (UWSN) are promising technologies for oceanography, environmental monitoring, disaster prevention, and resource exploration applications;however, they do suffer from low bandwidth, high delay, and energy limitations. Since recharging or replacing sensor batteries is impractical, achieving longer network lifetimes and reliable data transmission relies on advanced routing protocols suited to underwater environments. In this work, two related routing protocols, Eco Aware Routing Protocol (EARP) and Opportunistic Routing Protocol (ORP), are compared in terms of their energy efficiency. EARP reduces energy consumption and maximizes network lifetime through residual energy-based node selection, where ORP exploits the stochastic nature of underwater communications by randomly choosing several candidate forwarders for packet relaying. The results indicate that EARP performs better compared to ORP with regard to energy consumption and PDR while taking an average time of less delay end-to-end. EARP's energy-aware algorithms adapt to changing conditions and prioritize only nodes with maximum residual energy, thereby optimizing data forwarding and preventing network partitioning. MATLAB- based simulations show that EARP would actually outperform the setup in most of the conditions with respect to node density. Actually, the simulation results obtained elucidate EARP's ability to mitigate key challenges of UWSNs, thus enhancing energy efficiency and packet delivery delay. © 2024 IEEE.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

Transforming Fetal Health Monitoring: Advanced Machine Learning Innovations for Accurate Prognosis 7

Transforming Fetal Health Monitoring: Advanced Machine Learn...

引用

7th IEEE Pune Section International Conference, PuneCon 2024

作者： Nancy Jenifer, A. Jabasheela, L. Subramani, Kavitha Vinmathi, M.S. Sathiya, V. Maheswari, M. Panimalar Engineering College Computer Science and Engineering Chennai India

ISBN: (纸本)9798331527822

Early detection of pregnancy-related illnesses, such as congenital heart defects, neural tube defects, Down syndrome, and Inborn Errors of Metabolism (IEM), is vital for ensuring the health of both the fetus and the expectant mother. Leveraging advanced ML techniques-specifically Decision Trees, KNN, and Random Forest-can significantly enhance the accuracy of predicting these conditions. Decision Trees provide a transparent, interpretable model that visualizes decision-making through a series of binary splits, enabling healthcare professionals to understand the rationale behind predictions. In contrast, KNN operates on the principle of proximity, classifying new data points based on the 'k' nearest neighbors in a multidimensional space, making it effective in scenarios with complex, non-linear relationships. Meanwhile, Random Forest improves predictive performance by aggregating the outputs of multiple Decision Trees, thus mitigating overfitting and enhancing reliability. This ensemble approach draws on the diversity of trees, each trained on random subsets of data, allowing it to capture intricate interactions among variables. Together, these ML techniques utilize extensive historical data to autonomously predict pregnancy-related conditions, forming a sophisticated decision support system that complements traditional screening methods. By integrating these advanced algorithms, healthcare providers can achieve more accurate early diagnoses, ultimately improving maternal and fetal health outcomes and paving the way for targeted interventions. © 2024 IEEE.

关键词： Fetal monitoring

来源：评论

学校读者我要写书评

暂无评论

Improving Early Detection of Diabetic Retinopathy: A Hybrid Deep Learning Model Focused on Lesion Identification 1

Improving Early Detection of Diabetic Retinopathy: A Hybrid ...

引用

1st International Conference on Innovations in Communications, Electrical and computer engineering, ICICEC 2024

作者： Eluri, Rama Krishna Reddy, Y. Gnaneswar Valicharla, Karunakumar Prakash, K. Divya Sudheer, B. Narasaraopeta Engineering College Computer Science and Engineering Narasaraopet India

ISBN: (纸本)9798350376517

Diabetic Retinopathy has been found to be the leading cause of sight impairment in most parts of the globe, particularly in diabetic patients. An early detection of DR in retinal images could considerably reduce the chances of blindness. The present paper develops a hybrid deep learning approach using DenseNet, InceptionV3, and ResNet architecture combined with Random Forest Classifier, XGBClassifier, Decision Tree Classifier, and LightGBM Classifier. Preprocessing techniques consisting of green channel extraction, top-hat/bottom-hat transformations and data augmentation are used. The comparison of the model performance metrics such as precision, recall, F1 score, and AUC are done. It can be seen that the hybrid model has a strong difference in classifying different stages of DR. Comparison of a number of models is also made that comes up with the best model. Even more remains for the near future: improve sensitivity of the model and extend this approach to larger datasets. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Segmentation of MR Spinal Images Using a 2 - Stage Semantic Framework

Segmentation of MR Spinal Images Using a 2 - Stage Semantic ...

引用

2024 IEEE International Women in engineering (WIE) Conference on Electrical and computer engineering, WIECON-ECE 2024

作者： Kumarasamy, Karpagam College of Engineering Guindy Computer Science and Engineering Chennai India

ISBN: (数字)9798331535476

ISBN: (纸本)9798331535476

Segmentation of Spine with the 2 - Stage Semantic framework involves the Vertebrae and the Inter-vertebral discs components of the spine, helps in the identification of the spinal disorders. The methods using the Convolutional Neural Network does not precisely identify the interdependencies within the Vertebral and the Inter-vertebral disc components of the spine. The proposed method helps achieve Segmentation of MR Spinal images in two stages. The Coarse-Grained Segmentation utilizes the 3D-Graph Convolution Segmentation Network and the Segmentation Enhancement is carried out using the 2D-ResUNet. In Coarse-Grained Segmentation, the Spatial Pooling technique transforms the image into a graph-based representation, where individual spinal structures are represented as nodes. This graph undergoes three layers of graph convolutions to extract Semantic Features. Following this, the Inverse pooling process is applied to transform the graph back into a Visual Semantic Feature map. Experiments are carried out using the T2-weighted volumetric MRI data of 215 individuals[1], and achieved DSC values 88.56 ± 3.36%, 87.59 ± 3.54%, and 88.07 ± 5.25% for the Vertebrae, IVDs and the complete set of nineteen spinal components respectively. © 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

Identification of Epileptic Seizures using Recurrent Neural Networks and Time Series Transformer 7

Identification of Epileptic Seizures using Recurrent Neural ...

引用

7th International Conference on Circuit Power and Computing Technologies, ICCPCT 2024

作者： Rahul Chiranjeevi, V. Keerthana, H. Keerthana, G. Keerthana, S.P. Rajalakshmi Engineering College Computer Science and Engineering Chennai India

ISBN: (数字)9798350372816

ISBN: (纸本)9798350372816

In recent years epileptic seizures have become a major impact on large number of people worldwide. Over 60 million people have been affected globally due to this neurological condition. These seizures are caused by excessive discharge of electrical signals in the brain. To identify the excess discharge of signals, a device called electroencephalogram (EEG) is used. This device captures the signals by placing multiple sensors in scalp of the brain. However, the signals from the brain vary in time with different individuals thereby requiring more data. Also, differences are found in the signals between various stages of seizures. To overcome these issues and to detect the seizure automatically from EEG signal a novel Ensemble approach is proposed using Recurrent Neural Networks (RNN) and Time Series Transformer (TST). Raw EEG signals are pre-processed to remove undesirable noises from the obtained signals for further processing. The TST model was utilized to extract characteristics from EEG recordings. Furthermore, a recurrent neural network incorporating a long input layer, fully connected layer, activation layer was constructed in parallel to train the features, classify the EEG data, and identify the seizures. The effectiveness of the network was further examined and compared using various activation channels. The findings demonstrated that the suggested network could achieve a reasonably accurate prediction using a mere 18 channels. Experiments were carried out on the CHBMIT and Bonn datasets. Proposed method achieved an accuracy of 94.31% with an F-Score of 91.93%. Furthermore, classification of seizures has been improved compared to state of art methods. © 2024 IEEE.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

AI-Powered Dynamic Images: A New Frontier in Graphical Password Authentication 2

AI-Powered Dynamic Images: A New Frontier in Graphical Passw...

引用

2nd IEEE International Conference on Emerging Research in Computational science, ICERCS 2024

作者： Jananee, V. Ragava Krishnan, N.S. Rakesh, V.S. Rajalakshmi Engineering College Computer Science and Engineering Chennai India

ISBN: (数字)9798331534967

ISBN: (纸本)9798331534950

Graphical Password Authentication is an emerging and advanced technique in password authentication for hypothesis that provides a stronger and more secure password compared to the traditional text password, which is thus deemed superior. These methods of graphical authentication, however, have benefits that they have designed for current systems that still incorporate vulnerabilities like shoulder surfing, screen recording, brute force, and other attacks. The present research works on the existing graphical encryption systems that indicate the strengths and weaknesses the systems have. Enter a brand-new method-the AI-based dynamic image-which brings in more security for graphical passwords on authentication process as an authentication difference. It is an artificial intelligence back-end, in which with every access attempt, the picture is automatically varying but presents easy visible contents which give a higher level of enhanced security. © 2024 IEEE.

关键词： Visualization Scientific computing Heuristic algorithms Authentication Passwords Generative adversarial networks User experience Recording Security Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Cardiovascular Disease Prediction Using Deep Learning 11

Cardiovascular Disease Prediction Using Deep Learning

引用

11th International Conference on Reliability, Infocom Technologies and Optimization, ICRITO 2024

作者： Khanna, Vaneet Punjab Engineering College Department of Computer Science and Engineering India

ISBN: (纸本)9798350350357

Cardiovascular disease (CVD), ahead of all other causes of death worldwide in this era. There is an immediate need for accurate, reliable, and practically applicable ways of early detection and treatment of diseases, and it connects a number of hazards for cardiovascular disease. One common method for analyzing massive amounts of historical information in healthcare is data extraction. To assist physicians with CVD prediction, they employ a number of data mining and deep learning (DL) techniques to navigate complex medical data. This review paper critically examines the application of deep learning techniques in predicting cardiovascular diseases, emphasizing data collection, preprocessing, model selection, performance metrics, and challenges. The data-driven nature of DL models allows for the analysis of diverse patient information, contributing to more accurate risk assessments. The review achieves by discussing the challenges and limitations, emphasizing the importance of collaborative efforts to connect DL's potential in enhancing cardiovascular disease prediction and improving patient outcomes. © 2024 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Classification of Lung Diseases Using Transfer Learning with Chest X-Ray Images 2

Classification of Lung Diseases Using Transfer Learning with...

引用

2nd International Conference on Emerging Trends in Information Technology and engineering, ic-ETITE 2024

作者： Kavitha, S. Shudapreyaa, R.S. Prakash, P. Vaibhav, S. Viswa, V. Yogavarshan, S. Kongu Engineering College Computer Science and Engineering Erode India

ISBN: (纸本)9798350328202

Millions of people worldwide are impacted by lung illnesses, which are a major cause of morbidity and mortality. An accurate and timely diagnosis is essential for managing and treating conditions effectively. A useful non-invasive technique for identifying and categorizing a range of lung disorders, chest X- ray imaging enables prompt treatment and better patient outcomes. This research describes an extensive study that develops and evaluates a classification based on deep learning model using the Chest X-ray images Dataset from National Institutes of Health (NIH), which covers a wide spectrum of thoracic anomalies. Astrong model could be developed thanks to the dataset, which included 112,120 images with annotations taken from radiological reports utilizing approaches from Natural Language Processing (NLP) techniques. Pre-trained architectures like ResNet and specially created Convolutional Neural Networks (CNNs), trained with a weighted loss technique to manage data imbalance, are among the models that have been put into practise. Through extensive exploratory data analysis, preparation procedures, model training, and evaluation, the research offers insights into the performance and accuracy of each illness class's classification. At 83.57%, ResNet50 has the highest accuracy, followed by Custom CNN at 78.25%. ResNet18's accuracy was 72.67%, whereas VGG19's accuracy was 67.12%. ResNet50 is the most accurate of these models, while VGG19 performs less accurately. © 2024 IEEE.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

A Secure Framework for WSN-IoT Using Deep Learning for Enhanced Intrusion Detection

引用

computers, Materials & Continua 2024年第10期81卷 471-501页

作者： Chandraumakantham Om Kumar Sudhakaran Gajendran Suguna Marappan Mohammed Zakariah Abdulaziz S.Almazyad School of Computer Science&Engineering Vellore Institute of TechnologyChennai CampusChennai600127India School of Electronics&Engineering Vellore Institute of TechnologyChennai CampusChennai600127India Department of Computer Sciences and Engineering College of Applied ScienceKing Saud UniversityRiyadh11543Saudi Arabia Department of Computer Engineering College of Computer and Information SciencesKing Saud UniversityRiyadh11543Saudi Arabia

The security of the wireless sensor network-Internet of Things(WSN-IoT)network is more challenging due to its randomness and self-organized *** detection is one of the key methodologies utilized to ensure the security of the *** intrusion detection mechanisms have issues such as higher misclassification rates,increased model complexity,insignificant feature extraction,increased training time,increased run time complexity,computation overhead,failure to identify new attacks,increased energy consumption,and a variety of other factors that limit the performance of the intrusion system *** this research a security framework for WSN-IoT,through a deep learning technique is introduced using Modified Fuzzy-Adaptive DenseNet(MF_AdaDenseNet)and is benchmarked with datasets like NSL-KDD,UNSWNB15,CIDDS-001,Edge IIoT,Bot *** this,the optimal feature selection using Capturing Dingo Optimization(CDO)is devised to acquire relevant features by removing redundant *** proposed MF_AdaDenseNet intrusion detection model offers significant benefits by utilizing optimal feature selection with the CDO *** results in enhanced Detection Capacity with minimal computation complexity,as well as a reduction in False Alarm Rate(FAR)due to the consideration of classification error in the fitness *** a result,the combined CDO-based feature selection and MF_AdaDenseNet intrusion detection mechanism outperform other state-of-the-art techniques,achieving maximal Detection Capacity,precision,recall,and F-Measure of 99.46%,99.54%,99.91%,and 99.68%,respectively,along with minimal FAR and Mean Absolute Error(MAE)of 0.9%and 0.11.

关键词： Deep learning intrusion detection fuzzy rules feature selection false alarm rate accuracy wireless sensor networks

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：