检索结果-内蒙古大学图书馆

22nd International Symposium on Network Computing and Applications, NCA 2024

作者： Dixit, Aarju Das, Debasis Department of Computer Science and Engineering Indian Institute of Technology Jodhpur India

ISBN: (纸本)9798331510183

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder with an annual global economic impact of approximately $1 trillion. Early diagnosis is crucial to mitigate disease progression, yet current detection methodologies lack consistency and accuracy in delineating disease stages. To address these challenges, we introduce NeuroSense, a novel smartwatch-based early detection framework for AD. NeuroSense employs advanced activity recognition algorithms to passively monitor ambulatory patterns, behaviors, and sleep cycles via Inertial Measurement Unit (IMU) and photoplethysmogram (PPG) sensors from smartwatches (i.e., Apple Watch and Fitbit), including metrics like walking speed, sleep posture, and heart rate variability. NeuroSense leverages enhanced shallow neural networks integrating cross-sectional entropy to dynamically adjust feature weights, improving sensitivity to subtle variations in AD biomarkers. Additionally, the SqueezeNet architecture has been optimized by introducing deeper fire modules and expanding filter sizes, enabling high-resolution feature extraction with reduced computational overhead. These models are further refined with a cloud computing infrastructure, allowing model updation, ensuring adaptive learning based on continuous data influx. To validate its efficacy, NeuroSense was tested on AI-generated datasets, achieving a detection accuracy of 94.4%, substantially higher than existing approaches. This novel framework provides a robust, scalable, and non-invasive solution for continuous AD monitoring, representing a significant advancement in the early detection and management of neurodegenerative disorders. ©2024 IEEE.

关键词： Sleep research

来源：评论

学校读者我要写书评

暂无评论

An approach to on-stream DDoS blitz detection using machine learning algorithms

引用

Materials Today: Proceedings 2023年 80卷 3492-3499页

作者： Manjula, H.T. Neha Mangla Atria Institute Of Technology Bengaluru India Department Of Computer Science & Engineering CMRIT Bengaluru India Department of Information Science & Engineering Atria Institute of Technology Bengaluru India

Distributed Denial of service (DDoS) attacks is an enormous threat to today's cyber world, cyber networks are compromised by the attackers to distribute attacks in a large volume by denying the service to legitimate users. The toughest and challenging task in today's network and network security engineers is to identify compromised traffic (attacked) and legitimate (normal) traffic. The main goal of the paper is to detect DDos attacks using classification algorithms. To achieve the goal the proposed system uses attacking tool to initiate attacks using Loic attacking tool with the data set extracted from open source tool Wireshark and transferring the dataset to apache Spark for detection analysis. The system also uses Apache spark machine learning algorithms (MLib), classification algorithms to classify the dataset. We use Naive Bayes, KNN and Random forest classification algorithms to classify normal traffic and attacked traffic. Our system is capable of detecting attacks with respect to any traffic protocols ICMP, TCP, or UDP. The accuracy of detection is compared on three classification algorithms and noted that random forest gives the accuracy of 96.75%. © 2021

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

Label Flipping Attacks on Federated Learning: GAN-Based Poisoning and Countermeasures 2

Label Flipping Attacks on Federated Learning: GAN-Based Pois...

引用

2nd IEEE International Conference on Data Science and Network Security, ICDSNS 2024

作者： Abdullah, Muhammad Bapu, Vidhatri Akash, K.A. Khan, Abdul Mannan Bhargavi, M.S. Bangalore Institute of Technology Department of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9798350373110

Federated Learning (FL) is a promising approach for training machine learning models in a decentralized manner to preserve data privacy. However, FL is vulnerable to various attacks, including label-flipping attacks where malicious clients manipulate labels to degrade global model accuracy. This research investigates the efficacy of label-flipping attacks using synthetic images generated by a Generative Adversarial Network (GAN). The GAN architecture and learning procedure are adapted to address challenges related to dataset limitations and ensure realistic data generation for the attacks. The study analyzes the impact of label-flipping attacks on FL model performance. The results demonstrate that these attacks can evade conventional defenses and significantly reduce model accuracy. To mitigate this threat, a novel defense mechanism is proposed to identify and isolate poisoned clients based on abnormally high gradient values. This research highlights the security weaknesses of FL and proposes a defense mechanism against label-flipping attacks. This paves the way for more secure and trustworthy FL applications while preserving data privacy. © 2024 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Overlapping Community Detection based on Facets of Social Network: An Empirical Analysis 4

Overlapping Community Detection based on Facets of Social Ne...

引用

4th International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies, ICAECT 2024

作者： Das, Soumita Biswas, Anupam National Institute of Technology Silchar Department of Computer Science and Engineering India

ISBN: (纸本)9798350343670

Detection of overlapping communities is a challenging problem that has drawn a lot of research interest. This is motivated by the fact that in real-world networks, individuals frequently join multiple groups subsequently, resulting in overlapping communities. In this paper, we presented a comprehensive analysis of numerous state-of-the-art overlapping community detection algorithms in order to understand the relative efficiency of the corresponding algorithms in handling specific issues. We consider issues like the facets of the social networks that are used for overlapping community detection, time complexity, accuracy, and quality. However, the accuracy and quality metrics are not sufficient to evaluate the comparative performance of community detection algorithms because these measures use an indirect approach for comparing the algorithms. Therefore, we have additionally used a direct evaluation metric namely, topological variance for performance analysis of the community detection algorithms. Experiments are conducted on several widely used real world networks. This study allows us to identify the algorithms that work well in different scenarios. As a result, we arrive at findings that direct our algorithm selection procedure in accordance with predetermined goals. © 2024 IEEE.

关键词： Direct comparative analysis Evaluation metrics Facets of social network Overlapping community

来源：评论

学校读者我要写书评

暂无评论

Integrating Deep Learning with Ensemble Approach for Anomaly Detection in Network Traffic 4

Integrating Deep Learning with Ensemble Approach for Anomaly...

引用

4th IEEE International Conference on Mobile Networks and Wireless Communications, ICMNWC 2024

作者： Sangeetha, V. Naidu, R. China Appala Bhat, Avinash Kulkarni, Pranav Ramaiah Institute of Technology Department of Computer Science and Engineering Bangalore India

ISBN: (纸本)9798350352931

Intrusion Detection Systems (IDS) is frequently automated using expert systems and applied machine learning methods. Because of the interplay between different industrial control systems and the Internet environment that is inherent to IoT technology, critical infrastructure applications of communication technologies are vulnerable to attacks. Conventional machine learning techniques used in network anomaly detection are ineffective due to the massive network traffic present in crucial Cyber-Physical Systems. As a result, the detection and categorization of abnormalities at the network and host levels are successfully utilizing recently developed machine learning algorithms, with a focus on deep learning. It is very crucial in today's network traffic to distinguish between normal and malicious activities. With the rise of cyber threats, machine learning models have become popular for IDS due to their adaptability and high detection rates. This paper introduces a hybrid Network Intrusion Detection System (NIDS) that combines Long Short-Term Memory (LSTM) networks and XGBoost classifiers. By leveraging LSTM's temporal learning and XGBoost's boosting capabilities. The proposed model achieves 99% detection accuracy on the NF-BoT-IoT dataset, making it a robust solution for real-time anomaly detection. © 2024 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Parallel Complexity of Geometric Bipartite Matching 44

Parallel Complexity of Geometric Bipartite Matching

引用

44th IARCS Annual Conference on Foundations of Software technology and Theoretical computer Science, FSTTCS 2024

作者： Bhore, Sujoy Equbal, Sarfaraz Gurjar, Rohit Department of Computer Science & Engineering Indian Institute of Technology Bombay India

ISBN: (纸本)9783959773553

In this work, we study the parallel complexity of the geometric minimum-weight bipartite perfect matching (GWBPM) problem in 2. Here our graph is the complete bipartite graph G on two sets of points A and B in 2 (|A| = |B| = n) and the weight of each edge (a, b) ∈ A × B is the p distance (for some integer p ≥ 2) between the corresponding points, i.e., ||a - b||p. The objective is to find a minimum weight perfect matching of A ∪ B. In their seminal work, Mulmuley, Vazirani, and Vazirani (STOC 1987) showed that the weighted perfect matching problem on general bipartite graphs is in RNC. Almost three decades later, Fenner, Gurjar, and Thierauf (STOC 2016) showed that the problem is in Quasi-NC. Both of these results work only when the weights are of O(log n) bits. It is a long-standing open question to show the problem to be in NC. First, we show that in a geometric bipartite graph under the p metric for any p ≥ 2, unless we take Ω(n) bits of approximation for weights, we cannot distinguish the minimum-weight perfect matching from other perfect matchings. This means that we cannot hope for an MVV-like NC/RNC algorithm for solving GWBPM exactly (even when vertex coordinates are small integers). Next, we give an NC algorithm (assuming vertex coordinates are small integers) that solves GWBPM up to 1/poly(n) additive error, under the lp metric for any p ≥ 2. © 2024 Sujoy Bhore, Sarfaraz Equbal, and Rohit Gurjar.

关键词： Consensus algorithm

来源：评论

学校读者我要写书评

暂无评论

Validating δ-Currency Using Model Checking

Validating δ-Currency Using Model Checking

引用

International Conference on Data Science and Network engineering, ICDSNE 2023

作者： Prabhu, Shreekanth M. Department of Computer Science and Engineering CMR Institute of Technology Bengaluru India

ISBN: (纸本)9789819967544

In recent years, there is a surfeit of digital currencies, virtual currencies, and cryptocurrencies. These currencies serve as alternatives to fiat currencies in the form of physical currencies or deposits in banks. Some of the common characteristics that differentiate these currencies are how they gain and maintain their value, the anonymity of transactions, and considerations of security, data integrity, and transaction performance in a distributed computing scenario. Digital Currencies are generally issued by Central Banks raising privacy concerns. Cryptocurrencies work outside such formal mechanisms raising concerns about volatility and the possibility of loss. In this paper, we discuss a new variant called δ-Currency which attempts to navigate these concerns and arrive at an alternative mechanism that addresses privacy and financial security considerations in a novel manner. The δ-Currency is currently at a conceptual stage. We make use of Model Checking to validate the architecture, design principles, and implementation approach of δ-Currency. Model Checking focuses on fundamental building blocks of computation and is generally used to validate complex systems operating in an environment that enables a high degree of concurrency. In this paper, we make use of Model Checking to arrive at a common vocabulary, abstractions, and framework to understand seemingly disparate systems which nevertheless achieve the same objective. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Model checking

来源：评论

学校读者我要写书评

暂无评论

Exploring Transformer-Based Approaches for Hyperspectral Image Classification: A Comparative Analysis 3rd

Exploring Transformer-Based Approaches for Hyperspectral Ima...

引用

3rd International Conference on Machine Vision and Augmented Intelligence, MAI 2023

作者： Arya, Rajat Kumar Srivastava, Rajeev Department of Computer Science and Engineering Indian Institute of Technology-BHU Varanasi India

ISBN: (纸本)9789819743582

Hyperspectral imaging (HSI) is a technique that captures and analyzes multiple spectral bands for each pixel in an image. Convolutional neural networks (CNNs) effectively extract local features for HSI classification but struggle with capturing sequential spectral signatures. To address this, we conducted a comparative study to assess the performance of transformers, which can handle sequential information, in HSI classification. We reviewed a few models in the literature, analyzing their architecture and performance. Using the Indian Pines dataset, we evaluated models like ViT, SSRN-ViT, SpectralFormer, and Double-ViT. Our results show that Double-ViT models outperform transformers for HSI classification. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Multimodal Hate Speech Detection using Fine-Tuned Llama 2 Model

Multimodal Hate Speech Detection using Fine-Tuned Llama 2 Mo...

引用

2024 International Conference on Intelligent Algorithms for Computational Intelligence Systems, IACIS 2024

作者： Sasidaran, Keerthana Geetha, J. Ramaiah Institute of Technology Department of Computer Science and Engineering Bangalore India

ISBN: (纸本)9798350360660

Hate speech detection is crucial for secure online environments, especially with recent policy changes on social media. This study enhances the Llama 2 model by fine-tuning it with a comprehensive Twitter dataset to improve hate speech detection accuracy. The system's frontend allows users to upload or record audio, upload videos, or provide YouTube URLs, with processing managed by the Google Speech Recognition API and the Whisper library. The backend integrates advanced speech recognition and video processing technologies. The refined Llama 2 model achieved an F1 score of 0.94, with accuracy, precision, and recall metrics surpassing other state-of-the-art models. These results demonstrate the model's high effectiveness and reliability in hate speech detection. This research highlights the potential of transformer-based models in enhancing online safety and inclusivity. Future work will focus on expanding the dataset, improving model generalization, supporting additional languages, and enhancing system scalability to better handle evolving hate speech forms. © 2024 IEEE.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

Evaluating Neural Network Parameter Obliteration in Catastrophic Forgetting

Evaluating Neural Network Parameter Obliteration in Catastro...

引用

2024 IEEE Region 10 Symposium, TENSYMP 2024

作者： Sehra, Udit Dutta, Rishi Raj Pandit, Mohammad Khalid National Institute of Technology Hamirpur Department of Computer Science & Engineering India

ISBN: (纸本)9798350364866

One of the major weaknesses of Neural Networks is their inability to learn multiple functions (tasks) sequentially. This phenomenon is observed when a data distribution shift is encountered in continual learning, leading to catastrophic forgetting. Such ability is foremost important to develop models with general intelligence. In this paper, we evaluate how different parts of the neural network contribute to overall forgetting by investigating the training dynamics during different methods of continual learning. We evaluate and analyze the learning dynamics in catastrophic forgetting on frequently used classification models on CIFAR-10 and Fashion-MNIST datasets. Our findings reveal a consistent trend in continual learning scenarios, both in domain-incremental and class-incremental settings. Specifically, gamma means demonstrate stability in earlier layers but exhibit increased plasticity in later layers, indicating heightened adaptability to changes in data distributions. © 2024 IEEE.

关键词： Neural network models

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：