检索结果-内蒙古大学图书馆

HybridViT: An Approach for Alzheimer’s Disease Classification with ADNI Neuroimaging Data

SN computer science 2025年第4期6卷 1-18页

作者： Tekumudi, Vivek Sai Surya Chaitanya Ramanathan, Annamalai Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Chennai India

Alzheimer’s disease (AD) is a prevalent neurological disorder characterized by progressive brain cell degeneration and atrophy, leading to a gradual decline in cognitive and functional abilities. Timely diagnosis is paramount in potentially delaying or preventing the course of this debilitating condition. Magnetic Resonance Imaging presents a non-invasive means for longitudinal monitoring and serves as a crucial biomarker for tracking disease progression. In particular, Structural Magnetic Resonance Imaging (sMRI) enables the quantification of atrophy, which is a reliable indicator for assessing the precise stage and severity of AD-related neurodegeneration. In this research, we employ advanced machine learning techniques to address the challenge of accurate and early diagnosis of AD. We utilize a comprehensive dataset comprising five stages of 2D sMRI Image data, encompassing AD, Cognitively Normal, Mild Cognitive Impairment, Early Mild Cognitive Impairment, and Late Mild Cognitive Impairment classes. To optimize the classification process, we explore a novel approach that combines a Vision Transformer with pre-trained convolutional neural networks. Our study includes binary and multi-class classification tasks, and the performance of 26 pre-trained Keras Deep Learning models is assessed. Notably, The DenseNet121+ViT model achieves 91.26% accuracy in multi-class classification. For binary classification, MobileNetV2+ViT, EfficientNetB4+ViT, and MobileNet+ViT achieve accuracies of 92.33%, 97.82%, and 94.81%, respectively. These results highlight our approach’s potential to improve AD diagnosis accuracy and underscore the importance of deep learning in early detection of neurological disorders. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Alzheimer’s disease (AD) Alzheimer’s disease neuroimaging initiative (ADNI) HybridViT Pre-trained models Vision transformer

来源：评论

学校读者我要写书评

暂无评论

Efficient Text Summarization using Natural Language Processing

Efficient Text Summarization using Natural Language Processi...

引用

2023 International Conference on Data science, Agents and Artificial Intelligence, ICDSAAI 2023

作者： Kumar, S. Anupama Kumar, Y. S Kiran Thejas, P. Rv College of Engineering Department of Master of Computer Applications Bangalore560059 India

ISBN: (纸本)9798350348910

In recent years, a huge volume of textual data has been released through social media and various other domains that can be extracted and analyzed to reproduce useful information. The underlying information from the data should be retrieved and summarized within a specific period for effective analysis. Text summarization is the process of condensing a text document into a briefer version while retaining its key information and meaning and understanding the essence of a document. This process enables the user to save time. It has a wide range of applications in various domains such as news articles, scientific papers, legal documents, social media, and *** are two ways to implement Text summarization: Extractive and Abstractive Summarization. Extractive summarization involves generating summaries from the original text but cannot frame new sentences. Another approach is abstractive summarization, which involves generating new sentences that capture the essence of the original text. Text summarization can significantly reduce the time and effort required to extract important information from large volumes of text. Experimental data can be used to validate the system's effectiveness and fine-tune the various modules involved in the text summarization *** paper explores the implementation of different text summarization approaches to summarize the text data collected from different sources and analyze their performance using the ROUGE score. The abstractive text summarization method is more effective than the extractive summarization for the given set of input text chosen. © 2023 IEEE.

关键词： Syntactics

来源：评论

学校读者我要写书评

暂无评论

Enhancing Intrusion Detection Capabilities in IoT Technologies through Machine Learning

Enhancing Intrusion Detection Capabilities in IoT Technologi...

引用

2024 IEEE International Conference on Computing, applications and Systems, COMPAS 2024

作者： Hasan, Imran All Ahhad, Abdullah Md Rakibul Islam, Sk Mim, Marufa Akhter Habibur Rahman, Md Islamic University Department of Computer Science and Engineering Kushtia Bangladesh Islamic University Department of Tourism and Hospitality Management Kushtia Bangladesh Independent University Department of Computer Science and Engineering Dhaka Bangladesh

ISBN: (纸本)9798331529765

Smart home systems, industrial automation, healthcare systems, and others have been ground sectors developed massively because of the exponential growth that the Internet of Things (IoT) has had. The significant issues that it has presented are security challenges. IoT devices are weakly defended in terms of security and widely susceptible to cyber threats because of interconnectivity. This paper presents novel methods to mitigate such risks, with their implementation focused on intrusion detection systems that can be enhanced with Random Forest classifiers to detect and terminate malicious activities. The approach presented attains excellent performance, with an accuracy of 98.2% in the case of the NSL-KDD dataset and 99.9% for the BoT-IoT dataset. The paper discloses improved accuracy, precision, and recall relative to the existing approaches in the field and that RF classifiers effectively enhance the IDS capability. The importance of flexibility among ML models is therefore highlighted in the real-time detection and reaction to prevent significant damages and data breaches in IoT environments. The paper also discusses the deployment alternatives because of the limitation of resources in IoT devices, with which an architectural implementation of the IDS framework will raise challenges toward real-world IoT applications. Effective methods for overhead and accurate detection efficacy were also recommended. These results are regarded as an addition to the debate on IoT security and have been helpful to researchers, practitioners, and policymakers. This state-of-the-art machine learning technique is protection proposed for the IoT ecosystem not only from present threats but also from futuristic challenges to ensure the most secure IoT systems. Research contributes to innovation and continuous improvement by securing a growing IoT and future advanced systems through cyber collaboration © 2024 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Guardian PI : Raspberry Based Night Survaillance Camera 4

Guardian PI : Raspberry Based Night Survaillance Camera

引用

4th International Conference on Power, Energy, Control and Transmission Systems, ICPECTS 2024

作者： Lalitha, V. Venkat Raj, M. Nirmal Raaghav, N. Ramanan, D. Sri Sairam Engineering College Department of Computer Science and Engineering Chennai India

ISBN: (数字)9798331508845

ISBN: (纸本)9798331508845

Personal safety, especially for women during nighttime, remains a pressing concern. The "Guardian Pi"is an innovative solution that leveragesthe Raspberry Pi 4 Model B+ to provide real-time surveillance and emergency response. Equipped with a 360-degree HD camera and a high-frequency sound sensor, the system captures and analyzes environmental data, including audio and video, to detect potential threats. Upon identifying an emergency, the system instantly alerts pre-registered contacts and nearby police stations via email using Simple Mail Transfer Protocol (SMTP), providing continuous updates every second. The software, developed in Python, integrates key libraries such as GPIO for hardware interfacing, ensuring a reliable and user-friendly security solution. Guardian PiU aims to enhance nighttime safety by offering a proactive and efficient surveillance system for emergency situations. © 2024 IEEE.

关键词： Acoustic devices

来源：评论

学校读者我要写书评

暂无评论

Experience Replay: A value-based deep reinforcement approach for Malicious URL Detection in the Tor Browser 7

Experience Replay: A value-based deep reinforcement approach...

引用

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

作者： Sakthivel, K. Srinithi, V. Swetha, S. Birundha, S. Sri Ramakrishna Engineering College Department of Computer Science and Engineering Coimbatore India

ISBN: (数字)9798350372816

ISBN: (纸本)9798350372816

Finding malicious URLs in Tor is vital for preserving network security and privacy. Subsequently, Tor affords anonymity by routing traffic over a sequence of relays, it's frequently used for both legitimate and illegitimate reasons. Malicious URLs can lead to various threats, including phishing, malware distribution and illegal activities. Malicious URLs can lead users to websites hosting malware, which can compromise their devices and steal sensitive information. This article explores the discovery of malicious URLs through the ML and DL concepts. By utilizing the techniques such as Correlation-based Feature Selection, Chi-Square Test, Principal Component Analysis and Linear Discriminant Analysis, key features which are contributing to URL detection can be identified. Implementation of KNN, K Means, Random Forest, DQN and DDQN algorithms offers intuitions into how malicious URLs are recognized and alleviates possible cyber threats efficiently. The objective of this work includes, the idea of applying the deep reinforcement learning methods in which can pave the way to surge the accuracy of detection of the URL. It also stretches an idea for employing the deep reinforcement learning techniques for the chosen dataset. © 2024 IEEE.

关键词： Phishing

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Security System Using Blockchain Technology for Strong FMC Relationship

引用

Intelligent Automation & Soft Computing 2023年第1期35卷 111-128页

作者： K.Meenakshi K.Sashi Rekha Department of Computer Science and Engineering Saveetha School of EngineeringSaveetha Institute of Medical and Technical SciencesChennaiIndia

Blockchain technology is a shared database of logs of all consumer transactions which are registered on all machines on a *** transactions in the system are carried out by consensus processes and to preserve confidentiality all thefiles contained cannot be *** technology is the fundamental software behind digital currencies like Bitcoin,which is common in the *** computing is a method of using a network of external machines to store,monitor,and process information,rather than using the local computer or a local personal *** software is currently facing multiple problems including lack of data protection,data instability,and *** paper aims to give the highest security for multiple user environments in cloud for storing and accessing the data in *** users who are legitimate are only allowed for storing and accessing the data as like a secured block chain *** like the Blockchain which does not require a centralized system for transactions,the proposed system is also independent on centralized network *** decentralized system is developed in such a way to avoid *** system enables the fabricator to spend less or null resources to perform the validations for its direct operated *** ensures the product fabricator to avoid the circulation of its duplicate *** customer as an end-user is also ensured to have only the genuine products from the *** Fabricator(F),Merchant(M)and consumer(C)forms an interconnected triangular structure without allowing any counterfeiting agents in their secured cloud *** pro-posed approach provides the stability in the security system of the cloud using the chaining mechanism within different blocks at each *** takes roughly 4.7,6.2,and 7.5 ms,respectively,to register each node in the proposed system for 5,10,and 15 *** overall registration time for each scenario is 11.9,26.2,and 53.1 ms,despite the fact th

关键词： Blockchain bitcoin customer merchant data protection digital currencies transaction

来源：评论

学校读者我要写书评

暂无评论

Leveraging Attention Mechanisms to Enhance EfficientNet for Precise Analysis of Chest CT Images 2nd

Leveraging Attention Mechanisms to Enhance EfficientNet for...

引用

2nd International Conference on Big Data, IoT and Machine Learning, BIM 2023

作者： Rakibul Haque, Md. Al Mamun, Md. Department of Computer Science & Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh

ISBN: (纸本)9789819989362

Accurate and timely analysis of chest CT images is crucial for effectively diagnosing and treating a wide range of respiratory, cardiovascular, and infectious diseases, making it a vital component of modern medicine. Manual interpretation of chest CT can be time-consuming, prone to subjective variability, and potentially error-prone, highlighting the need for automation to improve efficiency and accuracy in analyzing large volumes of images. Researchers are increasingly using convolutional neural networks (CNNs) for chest CT image analysis due to their ability to learn complex features and patterns from large datasets. Despite significant advancements in analytical techniques, the challenge of determining which regions of interest to focus on and how to assign appropriate levels of importance to various features during chest CT image analysis remains an ongoing concern in the field. Our paper introduces a novel approach, utilizing soft and channel attention mechanisms in conjunction with an improved version of EfficientNetB0 to efficiently extract and prioritize critical features essential for accurately detecting and diagnosing lung-related ailments from chest CT scans. We have evaluated the proposed approach on a large dataset of CT scans, with the objective of accurately identifying COVID-19, non-COVID-19, and Community-Acquired Pneumonia (CAP) cases. Experimental results demonstrated exceptional performance, with achieved accuracy of 99.41%, surpassing other state-of-the-art methods in the field. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

A Review of the Repercussions of Mobile Phones and the Internet on Education

A Review of the Repercussions of Mobile Phones and the Inter...

引用

2024 International Conference on Recent Innovation in Smart and Sustainable Technology, ICRISST 2024

作者： Komarasamy, Dinesh Madhuradharani, Madhuradharani Krishnakumar, B. Parthasarathy, Karthikeyan Chinnusamy, Koushalya. K Kongu Engineering College Department of Computer Science and Engineering TamilNadu Perundurai India

ISBN: (纸本)9798350309140

Mobile phones and the Internet are the most popular modern technologies for life organization, especially in the field of education. Rapid advancements in mobile technology have made it possible for Internet-based mobile learning and its impact on education has become a studied concern. Numerous studies have been developed to discover the impacts of mobile, social media and Internet technology on education. The discoveries aided the studies in determining whether these technologies emphatically or contrarily influence the life of students regarding their academic performance. Motivated by this novel, this review paper synthesizes existing articles regarding three perspectives: benefits of technology in education;how mobile phones and the Internet have favored learning and teaching modes;the repercussions of mobile phones and the internet on education. The majority of works accentuate how these technologies unearth various dimensions of learning for students and highlight the obstacles in education. In contrast, the researchers also analyzed the awful academic situations among students created by mobile technology. But the inquiries are limited to the repercussions and impacts of these innovations and do not include any definite suggestions to mitigate the effects. This paper provides various ways for responsible and effective use of these technologies among students. © 2024 IEEE.

关键词： Education Internet Mobile phone and Social media Technology

来源：评论

学校读者我要写书评

暂无评论

CryptoGuard Education: Unleashing Cipher-Text Magic for Enhanced Security 1

CryptoGuard Education: Unleashing Cipher-Text Magic for Enha...

引用

1st International Conference on Cognitive, Green and Ubiquitous Computing, IC-CGU 2024

作者： Abinaya, P. Senthil Kumar, J. Gowtham Babu, E. Jeyanth Balaji, K. Mepco Schlenk Engineering College Department of Computer Science and Engineering Sivakasi India

ISBN: (数字)9798350361780

ISBN: (纸本)9798350361780

As educational systems increasingly rely on secure and efficient data access control, This paper introduces a comprehensive approach to fortifying data access security in educational systems by integrating 'Ciphertext Policy Attribute-Based Encryption' (CP-ABE). Leveraging the foundational modules of Attribute Identification and Access Policy, our proposed work aims to establish a dynamic and granular access control system. Using the implementation of CP-ABE, user attributes are meticulously mapped and identified, forming the basis for context-aware access policies that consider roles, academic levels, and department affiliations. The system's adaptability is emphasized, featuring real-time adjustments based on changing attributes to ensure relevance and responsiveness. Security and privacy are paramount, with CP-ABE encryption techniques employed to safeguard data confidentiality. © 2024 IEEE.

关键词： Access control

来源：评论

学校读者我要写书评

暂无评论

***: Learning-based Governance for Decentralized Finance (DeFi)

引用

IEEE Transactions on Services Computing 2025年第03期18卷 1278-1292页

作者： Xu, Jiahua Feng, Yebo Perez, Daniel Livshits, Benjamin University College London United Kingdom Exponential Science Foundation United Kingdom Nanyang Technological University School of Computer Science and Engineering Singapore Imperial College London Computer Science Department United Kingdom Imperial College London United Kingdom

Decentralized finance (DeFi) is an integral component of the blockchain ecosystem, enabling a range of financial activities through smart-contract-based protocols. Traditional DeFi governance typically involves manual parameter adjustments by protocol teams or token holder votes, and is thus prone to human bias and financial risks, undermining the system's integrity and security. While existing efforts aim to establish more adaptive parameter adjustment schemes, there remains a need for a governance model that is both more efficient and resilient to significant market manipulations. In this paper, we introduce "***", a learning-based governance framework that employs a deep Qnetwork (DQN) reinforcement learning (RL) strategy to perform semi-automated, data-driven parameter adjustments. We create a DeFi environment with an encoded action-state space akin to the Aave lending protocol for simulation and testing purposes, where *** has demonstrated the capability to retain funds that would have otherwise been lost to price oracle attacks. In tests with real-world data, *** outperforms the benchmark approaches by at least 14% and the static baseline model by tenfold, in terms of the preset performance metric–protocol profitability. Overall, the comprehensive evaluations confirm that *** is more efficient and effective than traditional governance methods, thereby enhancing the security, profitability, and ultimately, the sustainability of DeFi protocols. © 2025 IEEE. All rights reserved.

关键词： Decentralized finance

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：