检索结果-内蒙古大学图书馆

Scheduling of Low-Latency Medical Services in Healthcare Cloud with Deep Reinforcement Learning

Tsinghua science and technology 2025年第1期30卷 100-111页

作者： Hongfei Du Ming Liu Nianbo Liu Deying Li Wenzhong Li Lifeng Xu School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu 610054China School of Information Renmin University of ChinaBeijing 100872China Department of Computer Science and Technology Nanjing UniversityNanjing 210093China People’s Hospital of Quzhou City Wenzhou Medical UniversityQuzhou 324000China

In the current landscape of online data services,data transmission and cloud computing are often controlled separately by Internet Service Providers(ISPs)and cloud providers,resulting in significant cooperation challenges and suboptimal global data service *** this study,we propose an end-to-end scheduling method aimed at supporting low-latency and computation-intensive medical services within local wireless networks and healthcare *** approach serves as a practical paradigm for achieving low-latency data services in local private cloud *** meet the low-latency requirement while minimizing communication and computation resource usage,we leverage Deep Reinforcement Learning(DRL)algorithms to learn a policy for automatically regulating the transmission rate of medical services and the computation speed of cloud ***,we utilize a two-stage tandem queue to address this problem *** experiments are conducted to validate the effectiveness for our proposed method under various arrival rates of medical services.

关键词： medical service tandem queue cloud computing Deep Reinforcement Learning(DRL) resource allocation

来源：评论

学校读者我要写书评

暂无评论

Small Language Model Makes an Effective Long Text Extractor 39

Small Language Model Makes an Effective Long Text Extractor

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Chen, Yelin Zhang, Fanjin Tang, Jie School of Computer Science and Technology Xinjiang University Urumqi830049 China Department of Computer Science and Technology Tsinghua University Beijing100084 China

ISBN: (纸本)157735897X

Named Entity Recognition (NER) is a fundamental problem in natural language processing (NLP). However, the task of extracting longer entity spans (e.g., awards) from extended texts (e.g., homepages) is barely explored. Current NER methods predominantly fall into two categories: span-based methods and generation-based methods. Span-based methods require the enumeration of all possible token-pair spans, followed by classification on each span, resulting in substantial redundant computations and excessive GPU memory usage. In contrast, generation-based methods involve prompting or fine-tuning large language models (LLMs) to adapt to downstream NER tasks. However, these methods struggle with the accurate generation of longer spans and often incur significant time costs for effective fine-tuning. To address these challenges, this paper introduces a lightweight span-based NER method called SeNER, which incorporates a bidirectional arrow attention mechanism coupled with LogN-Scaling on the [CLS] token to embed long texts effectively, and comprises a novel bidirectional sliding-window plus-shaped attention (BiSPA) mechanism to reduce redundant candidate token-pair spans significantly and model interactions between token-pair spans simultaneously. Extensive experiments demonstrate that our method achieves state-of-the-art extraction accuracy on three long NER datasets and is capable of extracting entities from long texts in a GPU-memory-friendly manner. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Text mining

来源：评论

学校读者我要写书评

暂无评论

Learning to Communicate Among Agents for Large-Scale Dynamic Path Planning With Genetic Programming Hyperheuristic

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第5期6卷 1269-1283页

作者： Liao, Xiao-Cheng Hu, Xiao-Min Chen, Xiang-Ling Mei, Yi Jia, Ya-Hui Chen, Wei-Neng Victoria University of Wellington Centre for Data Science and Artificial Intelligence School of Engineering and Computer Science Wellington6140 New Zealand Guangdong University of Technology School of Computer Science and Technology Guangzhou510006 China Hanyang University Department of Electrical and Electronic Engineering Ansan15588 Korea Republic of South China University of Technology School of Future Technology Guangzhou510006 China Pazhou Lab Guangzhou510005 China South China University of Technology School of Computer Science and Engineering State Key Laboratory of Subtropical Building and Urban Science Guangzhou510006 China

Genetic programming hyperheuristic (GPHH) has recently become a promising methodology for large-scale dynamic path planning (LDPP) since it can produce reusable heuristics rather than disposable solutions. However, in this methodology, the extracted local and decentralized heuristic for agents that lack a global systemic view sometimes may be problematic. Therefore, a new challenge is to strike a balance between conciseness to improve generalization ability and incorporation of more global information to obtain better performance. In this work, we target the LDPP problem and propose a communication learning mechanism (ComLGP) for GPHH to address the above difficulties. In ComLGP, a communication function is introduced to serve as a communication protocol and exist in the form of an extra terminal in GPHH. Compared to the classic terminals which are fixed in genetic programing, this communication function undergoes optimization along with the evolutionary process of GPHH. In this way, the communication function can be learned which enables agents to communicate without a predefined communication protocol. Then, a caching and lazy updating mechanism for ComLGP is presented to accelerate the calculation of communication content. Last, we verified our method on 22 scenarios including two real world road networks. The experimental results demonstrate that the proposed ComLGP can successfully learn to communicate. Although in the absence of any manually designed communication features, ComLGP is capable of achieving performance competitive to the state-of-the-art method that employs a predefined communication protocol and outperforms the remaining compared methods in most scenarios. © 2020 IEEE.

关键词： Genetic programming

来源：评论

学校读者我要写书评

暂无评论

Assessing Secure OpenID-Based EAAA Protocol to Prevent MITM and Phishing Attacks in Web Apps

引用

computers, Materials & Continua 2023年第6期75卷 4713-4733页

作者： Muhammad Bilal Sandile C.Showngwe Abid Bashir Yazeed Y.Ghadi College of Computer Science and Technology Zhejiang UniversityHangzhou310027China Department of Mathematical Statistics and Actuarial Science Faculty of Natural and Agricultural SciencesUniversity of the Free StateBloemfontein9301South Africa Department of Computer Science National Textile UniversityFaisalabad38000Pakistan Department of Computer Science/Software Engineering Al Ain UniversityAbu DhabiUAE

To secure web applications from Man-In-The-Middle(MITM)and phishing attacks is a challenging task *** this purpose,authen-tication protocol plays a vital role in web communication which securely transfers data from one party to *** authentication works via OpenID,Kerberos,password authentication protocols,***,there are still some limitations present in the reported security *** this paper,the presented anticipated strategy secures both Web-based attacks by leveraging encoded emails and a novel password form pattern *** proposed OpenID-based encrypted Email’s Authentication,Authorization,and Accounting(EAAA)protocol ensure security by relying on the email authenticity and a Special Secret Encrypted Alphanumeric String(SSEAS).This string is deployed on both the relying party and the email server,which is unique and *** first authentication,OpenID Uniform Resource Locator(URL)identity,is performed on the identity provider side.A second authentication is carried out by the hidden Email’s server side and receives a third authentication *** Email’s third SSEAS authentication link manages on the relying party(RP).Compared to existing cryptographic single sign-on protocols,the EAAA protocol ensures that an OpenID URL’s identity is secured from MITM and phishing *** study manages two attacks such as MITM and phishing attacks and gives 339 ms response time which is higher than the already reported methods,such as Single Sign-On(SSO)and *** experimental sites were examined by 72 information technology(IT)specialists,who found that 88.89%of respondents successfully validated the user authorization provided to them via *** proposed EAAA protocol minimizes the higher-level risk of MITM and phishing attacks in an OpenID-based atmosphere.

关键词： Secure user authentication SSO OpenID phishing attack MITM attack

来源：评论

学校读者我要写书评

暂无评论

Scalable and Foldable Origami-Inspired Supernumerary Robotic Limbs for Daily Tasks

引用

IEEE Access 2024年 12卷 53436-53447页

作者： Kusunoki, Mikiya Nguyen, Linh Viet Tsai, Hsin-Ruey Ho, Van Anh Xie, Haoran Japan Advanced Institute of Science and Technology Ishikawa Nomi923-1211 Japan National Chengchi University Department of Computer Science Taipei11605 Taiwan

This paper proposes scalable and foldable origami-inspired supernumerary robotic limbs that incorporate a continuous robotic mechanism to support activities of daily living and presents an analysis of the fabrication process, evaluation experiments, and potential applications of the proposed devices. These origami-inspired supernumerary robotic limbs can augment the wearer's capabilities by adjusting the scale of the devices to various applications and attaching them to parts of the body, such as the hands, wrists, arms, and torso. The device's foldable structure can be unfolded for use and folded when not in use to prevent physical contact with the wearer and the surrounding environment so that the user's movements are not restricted. In addition, the robotic limb is made of lightweight and soft material, so it remains safe to use and fatigue-free even after prolonged use. To verify the durability of the origami module, we conducted an experiment in which the module underwent 1,000 compressions using different materials and layer counts based on the folding pattern outlined in the manufacturing diagram. The experimental results showed that polypropylene was a suitable material for reducing fatigue due to folding, compared to polypropylene synthetic paper or plain paper. We also observed decreased stiffness with an increase in the number of layers. We believe that the proposed origami-inspired supernumerary robotic limbs can enlighten novel approaches to combining soft robotic and wearable devices. © 2013 IEEE.

关键词： Fabrication

来源：评论

学校读者我要写书评

暂无评论

Sports Events Recognition Using Multi Features and Deep Belief Network

引用

computers, Materials & Continua 2024年第10期81卷 309-326页

作者： Bayan Alabdullah Muhammad Tayyab Yahay AlQahtani Naif Al Mudawi Asaad Algarni Ahmad Jalal Jeongmin Park Department of Information Systems College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Department of Computer Science Air UniversityIslamabad44000Pakistan Department of Computer Science Applied CollegeKing Khalid UniversityAbha61421Saudi Arabia Department of Computer Science College of Computer Science and Information SystemNajran UniversityNajran55461Saudi Arabia Department of Computer Sciences Faculty of Computing and Information TechnologyNorthern Border UniversityRafha91911Saudi Arabia Department of Computer Engineering Korea Polytechnic UniversitySiheung-siGyeonggi-do237Republic of Korea

In the modern era of a growing population,it is arduous for humans to monitor every aspect of sports,events occurring around us,and scenarios or *** recognition of different types of sports and events has increasingly incorporated the use of machine learning and artificial *** research focuses on detecting and recognizing events in sequential photos characterized by several factors,including the size,location,and position of people’s body parts in those pictures,and the influence around those *** approaches utilized,here are feature descriptors such as MSER(Maximally Stable Extremal Regions),SIFT(Scale-Invariant Feature Transform),and DOF(degree of freedom)between the joint points are applied to the skeleton ***,for the same purposes,other features such as BRISK(Binary Robust Invariant Scalable Keypoints),ORB(Oriented FAST and Rotated BRIEF),and HOG(Histogram of Oriented Gradients)are applied on full body or *** integration of these techniques increases the discriminative nature of characteristics retrieved in the identification process of the event,hence improving the efficiency and reliability of the entire *** extracted features are passed to the early fusion and DBscan for feature fusion and *** deep belief,network is employed for *** results demonstrate a separate experiment’s detection average recognition rate of 87%in the HMDB51 video database and 89%in the YouTube database,showing a better perspective than the current methods in sports and event identification.

关键词： Machine learning silhouettes extremal regions joint points scalable keypoints

来源：评论

学校读者我要写书评

暂无评论

Transfer Learning and Explainable AI for Brain Tumor Classification: A Study Using MRI Data from Bangladesh 6

Transfer Learning and Explainable AI for Brain Tumor Classif...

引用

6th International Conference on Sustainable Technologies for Industry 5.0, STI 2024

作者： Sarker, Shuvashis Ahsanullah University of Science and Technology Department of Computer Science and Engineering Dhaka Bangladesh

ISBN: (数字)9798331531973

ISBN: (纸本)9798331531973

Brain tumors, regardless of being benign or ma-lignant, pose considerable health risks, with malignant tumors being more perilous due to their swift and uncontrolled prolifer-ation, resulting in malignancy. Timely identification is crucial for enhancing patient outcomes, particularly in nations such as Bangladesh, where healthcare infrastructure is constrained. Manual MRI analysis is arduous and susceptible to inaccuracies, rendering it inefficient for prompt diagnosis. This research sought to tackle these problems by creating an automated brain tumor classification system utilizing MRI data obtained from many hospitals in Bangladesh. Advanced deep learning models, including VGGI6, VGGI9, and ResNet50, were utilized to classify glioma, meningioma, and various brain cancers. Explainable AI (XAI) methodologies, such as Grad-CAM and Grad-CAM++, were employed to improve model interpretability by emphasizing the critical areas in MRI scans that influenced the categorization. VGG16 achieved the most accuracy, attaining 99.17%. The integration of XAI enhanced the system's transparency and stability, rendering it more appropriate for clinical application in resource-limited environments such as Bangladesh. This study highlights the capability of deep learning models, in conjunction with explainable artificial intelligence (XAI), to enhance brain tumor detection and identification in areas with restricted access to advanced medical technologies. © 2024 IEEE.

关键词： Clinical research

来源：评论

学校读者我要写书评

暂无评论

Classification of Malicious Financial Applications Using Naive Bayes and K-Nearest Neighbor Algorithms Based on Android Permission 3

Classification of Malicious Financial Applications Using Nai...

引用

3rd International Conference on Advancement in Electrical and Electronic Engineering, ICAEEE 2024

作者： Akter, Mst.Liya Moon, Iffath Tanjim Akter, Md Akash Shafa, Zhakia Tabessum Pundra University of Science & Technology Department of Computer Science & Engineering Bogura Bangladesh

ISBN: (纸本)9798350388282

In this research paper, we have worked on some of the vulnerabilities of multiple Android apps and we have explored the vulnerability of financial apps and how these apps are acting as threats in our daily lives. Nowadays in this digital technology, mobile banking is a common way for people to make transactions, through which many people have to face financial loss. we have identified the vulnerability of Financial Apps by performing permission analysis to protect the privacy, security, and financial security of these customers. Our main contribution is to find out the vulnerabilities of Android apps by permission analysis to create high-quality datasets using static analysis and learning the dataset through the K-Nearest Neighbor algorithm and Naive Bayes algorithm to show whether it is malicious or not by using the 'Permission Dataset' and the 'Appdroid' dataset. To test the developed model, we have extracted 1438 combined attributes from these two datasets and took 12050 applications samples. According to the test findings, the Naive Bayes model can improve detection by roughly 82.07% For further testing it was 88.25% that the KNN model has the highest accuracy and efficiency. Our main objective is to identify multiple Financials app vulnerabilities and create a comprehensive dataset by permission analysis and by using a machine learning model, as a result, we have been able to determine the Financial Apps vulnerability and safeguard these financial security, privacy, and safety. © 2024 IEEE.

关键词： Static analysis

来源：评论

学校读者我要写书评

暂无评论

Energy efficient quantum-informed ant colony optimization algorithms for industrial internet of things

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第3期5卷 1077-1086页

作者： Jannu, Srikanth Dara, Suresh Thuppari, Chaitanya Vidyarthi, Ankit Ghosh, Debjani Tiwari, Prayag Muhammad, Ghulam Department of Computer Science and Engineering Vaagdevi Engineering College Singaram506005 India School of Technology Woxsen University HyderabadTS 502 345 India Department of Computer Science and Engineering and Information Technology Jaypee Institute of Information Technology Noida201309 India Department of Computer Science and Engineering Bennett University Greater Noida201309 India School of Information Technology Halmstad University Halmstad301 18 Sweden Department of Computer Engineering College of Computer and Information Sciences King Saud University Riyadh11451 Saudi Arabia

One of themost prominent research areas in information technology is the Internet of Things (IoT) as its applications are widely used, such as structural monitoring, health care management systems, agriculture and battlefield management, and so on. Due to its self-organizing network and simple installation of the network, the researchers have been attracted to pursue research in the various fields of IoTs. However, a huge amount of work has been addressed on various problems confronted by IoT. The nodes densely deploy over critical environments and those are operated on tiny batteries. Moreover, the replacement of dead batteries in the nodes is almost impractical. Therefore, the problem of energy preservation and maximization of IoT networks has become the most prominent research area. However, numerous state-of-The-Art algorithms have addressed this issue. Thus, it has become necessary to gather the information and send it to the base station in an optimized method to maximize the network. Therefore, in this article, we propose a novel quantum-informed ant colony optimization (ACO) routing algorithm with the efficient encoding scheme of cluster head selection and derivation of information heuristic factors. The algorithm has been tested by simulation for various network scenarios. The simulation results of the proposed algorithm show its efficacy over a few existing evolutionary algorithms using various performance metrics, such as residual energy of the network, network lifetime, and the number of live IoT nodes. Impact Statement-Toward IoT-based applications, here we presented the Quantum-inspired ACO clustering algorithm for network lifetime. IoT nodes in the clustering phase choose theirCH through the distance between cluster member IoT nodes and the residual energy. Thus, CH selection reduces the energy consumption of member IoT nodes. Therefore, our significant contributions are summarized as follows. i. Developing Quantum-informed ACO clustered routing algor

关键词： Energy efficiency

来源：评论

学校读者我要写书评

暂无评论

NeuroMindscape: Deep Learning Unraveling Alzheimer's Stages Through MRI-Based Classification 2

NeuroMindscape: Deep Learning Unraveling Alzheimer's Stages ...

引用

2nd International Conference on Power, Electrical, Electronics and Industrial Applications, PEEIACON 2024

作者： Barua, Sumitro Dipu, Dipon Deb Broti, Tanjila Ahsanullah University of Science & Technology Department of Computer Science & Engineering Dhaka Bangladesh

ISBN: (纸本)9798331517984

A degenerative condition, Alzheimer's disease can result in a gradual deterioration of the capacity to carry out basic tasks, often beginning with mild memory loss. Alzheimer's disease is the predominant contributor to the onset of dementia, affecting approximately 1 in 9 adults aged 65 and older (10.7%) [14]. Lately, there has been a lot of promise in employing deep learning, especially the Convolutional Neural Networks (CNN), in various medical imaging research works, including Alzheimer's Disease (AD). This research focuses on improving Convolutional Neural Network (CNN) model for accurate classification of 4 stages of dementia - mild, moderate, non-mild and very mild. This research introduces a model comprising seven layers, showcasing less complexity compared to its predecessors. Despite its simplicity, it demonstrates remarkable proficiency in multi-class classification tasks, achieving high-performance levels. With this model, the publicly available Kaggle Alzheimer's Disease (AD) dataset [16] was accessed, which contains 6400 MRI images with the total size of 36 Megabytes, and another Kaggle dataset [17] which adopted GAN (Generative adversarial network) based approach on the former dataset to produce synthetic images to address the imbalance issue. Using the original Kaggle Alzheimer's dataset and through methods of traditional augmentation to balance out of the classes, The proposed model successfully distinguished between the four stages of the disease with 97.05% accuracy which outperformed most of its other predeceasing CNN models, and on the Kaggle GAN-based dataset [17] the model achieved 97% accuracy. © 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

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：