检索结果-内蒙古大学图书馆

Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), IEEE Conference on

作者： Ishdeep Singla Athulya Jayan V Nikhil Maurya Akash Kumar Akshita Sharma Harshit Mishra Computer Science Engineering Chandigarh University India

ISBN: (数字)9798331521691

ISBN: (纸本)9798331521707

This study addresses the shortcomings in diagnosing and managing HCM (Hypertrophic Cardiomyopathy), a genetic heart disease affecting 0.5-1.5% of the global population, by employing advanced ML (Machine Learning) approaches. The research integrates multimodal data, including cardiac MRI, genetic, clinical, and outcome information, to design and implement interpretable ML models capable of diagnosing HCM, assessing disease progression, and indicating individualized treatment courses. To enhance generalization and facilitate clinical application, a large multi-center dataset is utilized. Results demonstrate that the developed ML models achieve higher accuracy in HCM diagnosis and risk stratification compared to traditional methods, with particular strength in early detection and prognosis prediction. This work not only aims to improve care for HCM patients but also establishes a foundation for applying ML algorithms to other complex genetic cardiac diseases, potentially shifting the paradigm of cardiovascular medicine towards a more targeted approach. The findings suggest significant potential for improving patient outcomes through personalized treatment strategies and more accurate risk evaluation, setting the stage for further research into ML applications in complex cardiovascular disorders and potentially revolutionizing precision medicine in cardiology.

关键词： Technological innovation Accuracy Precision medicine Magnetic resonance imaging Machine learning Predictive models Genetics Prognostics and health management Medical diagnostic imaging Diseases

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Hybrid MPI and OpenMP on Smith-Waterman Algorithm 3

Performance Analysis of Hybrid MPI and OpenMP on Smith-Water...

引用

3rd International Conference on Intelligent Systems, Advanced Computing, and Communication, ISACC 2025

作者： Ninama, Keyur Patel, Jainamkumar Girish, K.K. Reddy, Manideep Raya V S R S Bhowmik, Biswajit National Institute of Technology Karnataka Maharshi Sushrut CAS Lab BRICS Laboratory Dept. of Computer Science and Engineering Mangalore India National Institute of Technology Karnataka Ishwarchandra Vidyasagar AIT Lab BRICS Laboratory Dept. of Computer Science and Engineering Mangalore India National Institute of Technology Karnataka Maharshi Kanad QC Lab BRICS Laboratory Dept. of Computer Science and Engineering Mangalore India National Institute of Technology Karnataka BRICS Laboratory Dept. of Computer Science and Engineering Mangalore India

ISBN: (纸本)9798331523893

In the rapidly advancing field of bioinformatics, sequence alignment is a pivotal task for elucidating genetic statistics and evolutionary relationships. As the volume and complexity of biological data continue to grow, it becomes imperative to employ effective computational techniques to manage this expansion. The Smith-Waterman algorithm is a key tool for sequence alignment;however, its performance can be constrained by the substantial size of contemporary datasets. To overcome this limitation, this paper explores a hybrid parallelization strategy that integrates message passing interface (MPI) with open multi-processing (OpenMP). This approach aims to significantly enhance the algorithm's efficiency by leveraging the strengths of both parallelization models. By optimizing the scalability and execution speed of the Smith-Waterman algorithm on advanced high-performance computing (HPC) systems, the hybrid technique not only improves performance but also enables more rapid and accurate biological data analysis. © 2025 IEEE.

关键词： High-Performance Computing Message Passing Interface Open Multi-Processing Parallelization Performance Scalability

来源：评论

学校读者我要写书评

暂无评论

Asynchronous I/O Persistence for In-Memory Database Servers: Leveraging io_uring to Optimize Redis Persistence 17th

Asynchronous I/O Persistence for In-Memory Database Servers:...

引用

17th International Conference on Cloud Computing, CLOUD 2024, Held as Part of the Services Conference Federation, SCF 2024

作者： Chen, Le-Gao Li, Yanzhi Laohakangvalvit, Tipporn Sugaya, Midori Innovative Global Program College of Engineering Shibaura Institute of Technology Tokyo Koto City Japan Department of Computer Science and Engineering College of Engineering Shibaura Institute of Technology Tokyo Koto City Japan

ISBN: (纸本)9783031771521

With constant hardware improvements allowing for increasingly large memory sizes, in-memory database servers have become an attractive option for various cloud applications. Even though in-memory database servers store all data in memory, there is still a need to persist data onto an I/O device for maintenance, updates, or protection against server crashes. In high throughput environments, persistence happens often, significantly hindering server throughput performance. Although popular in-memory database servers such as Redis or Dragonfly DB have adopted asynchronous I/O for socket operations, none have extended this approach for persistence. This results in significantly longer time taken for persistence, leading to prolonged reduced throughput. Given the recent introduction of io_uring, which has demonstrated significant performance improvements over past asynchronous I/O APIs, we have decided to implement it into Redis to evaluate its effectiveness in reducing these performance overheads. Our findings indicate that io_uring can significantly enhance performance in use cases with medium to large payload sizes but may hinder performances in use cases involving small payload sizes due to higher complexity and higher operation cost per datapoint. This method can likely be applied to other servers, such as Dragonfly DB and Memcached, to achieve similar benefits in appropriate scenarios. Additionally, in-memory database server persistence can serve as valuable benchmarks for further io_uring development and optimization. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

Toward AI-Augmented Formal Verification: A Preliminary Investigation of ENGRU and Its Challenges

引用

IEEE Access 2025年 13卷 84357-84379页

作者： Chanon Dechsupa Teerapong Panboonyuen Wiwat Vatanawood Praisan Padungweang Chakchai So-In Department of Computer Science College of Computing Khon Kaen University Khon Kaen Thailand Department of Computer Engineering Faculty of Engineering Chulalongkorn University Bangkok Thailand

State-space graphs and automata serve as fundamental tools for modeling and analyzing the behavior of computational systems. Recurrent neural networks (RNNs) and language models are deeply intertwined, as RNNS provide the foundational architecture that enables language models to process sequential data, capture contextual dependencies, and improve natural language processing tasks. Both RNNs and state-space graphs help evaluate discrete-time systems within this formal framework. However, the basic question of their equivalence remains an open challenge regarding the models governing sentence structure in natural language and the formal model in automata theory. In this paper, we present ENGRU (Enhanced Gated Recurrent Units), a novel deep learning-based approach for formal verification. ENGRU integrates concepts of model checking techniques, Colored Petri Nets (CPNs), and sequential learning to analyze systems at a high level of abstraction. CPN models undergo state-space enumeration via a model-checking tool, generating a state-space graph and an automaton from inherent state transition patterns. These graphs are transformed into sequential representations as sub-paths, enabling ENGRU to learn the execution paths and predict system behaviors. ENGRU effectively predicts goal states within discrete-time models by the competency of gated recurrent mechanisms, encouraging early bug detection and allowing predictive state-space exploration. Experimental results demonstrate that ENGRU’s ability achieves high accuracy and efficiency in goal state predictions. The source code for ENGRU is available at ( https://***/kaopanboonyuen/ENGRU ).

关键词： Model checking Analytical models Computational modeling Formal verification Image color analysis Petri nets Logic Time complexity Software reliability Explosions

来源：评论

学校读者我要写书评

暂无评论

AutoFHE: automated adaption of CNNs for efficient evaluation over FHE 24

AutoFHE: automated adaption of CNNs for efficient evaluation...

引用

Proceedings of the 33rd USENIX Conference on Security Symposium

作者： Wei Ao Vishnu Naresh Boddeti Computer Science and Engineering Michigan State University

ISBN: (纸本)9781939133441

Secure inference of deep convolutional neural networks (CNNs) under RNS-CKKS involves polynomial approximation of unsupported non-linear activation functions. However, existing approaches have three main limitations: 1) Inflexibility: The polynomial approximation and associated homomorphic evaluation architecture are customized manually for each CNN architecture and do not generalize to other networks. 2) Suboptimal Approximation: Each activation function is approximated instead of the function represented by the CNN. 3) Restricted Design: Either high-degree or low-degree polynomial approximations are used. The former retains high accuracy but slows down inference due to bootstrapping operations, while the latter accelerates ciphertext inference but compromises accuracy. To address these limitations, we present AutoFHE, which automatically adapts standard CNNs for secure inference under RNS-CKKS. The key idea is to adopt lay-erwise mixed-degree polynomial activation functions, which are optimized jointly with the homomorphic evaluation architecture in terms of the placement of bootstrapping operations. The problem is modeled within a multi-objective optimization framework to maximize accuracy and minimize the number of bootstrapping operations. AutoFHE can be applied flexibly on any CNN architecture, and it provides diverse solutions that span the trade-off between accuracy and latency. Experimental evaluation over RNS-CKKS encrypted CIFAR datasets shows that AutoFHE accelerates secure inference by 1.32× to 1.8× compared to methods employing high-degree polynomials. It also improves accuracy by up to 2.56% compared to methods using low-degree polynomials. Lastly, AutoFHE accelerates inference and improves accuracy by 103× and 3.46%, respectively, compared to CNNs under TFHE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Machine learning assisted plasmonic metascreen for enhanced broadband absorption in ultra-thin silicon films

引用

光：科学与应用（英文版） 2025年第2期14卷 411-421页

作者： Waqas W.Ahmed Haicheng Cao Changqing Xu Mohamed Farhat Muhammad Amin Xiaohang Li Xiangliang Zhang Ying Wu Division of Computer Electrical and Mathematical Sciences and EngineeringKing Abdullah University of Science and Technology(KAUST)Thuwal 23955-6900Saudi Arabia College of Engineering Taibah UniversityMadinah 42353Saudi Arabia Division of Computer Electrical and Mathematical Sciences and EngineeringKing Abdullah University of Science and Technology(KAUST)Thuwal 23955-6900Saudi Arabia Division of Physical Science and Engineering King Abdullah University of Science and Technology(KAUST)Thuwal 23955-6900Saudi Arabia Division of Computer Electrical and Mathematical Sciences and EngineeringKing Abdullah University of Science and Technology(KAUST)Thuwal 23955-6900Saudi Arabia Department of Computer Science and Engineering University of Notre DameNotre DameIN 46556USA

We propose and demonstrate a data-driven plasmonic metascreen that efficiently absorbs incident light over a wide spectral range in an ultra-thin silicon *** embedding a double-nanoring silver array within a 20 nm ultrathin amorphous silicon(a-Si)layer,we achieve a significant enhancement of light *** enhancement arises from the interaction between the resonant cavity modes and localized plasmonic modes,requiring precise tuning of plasmon resonances to match the absorption region of the silicon active *** facilitate the device design and improve light absorption without increasing the thickness of the active layer,we develop a deep learning framework,which learns to map from the absorption spectra to the design *** inverse design strategy helps to tune the absorption for selective spectral *** optimized design surpasses the bare silicon planar device,exhibiting a remarkable enhancement of over 100％.Experimental validation confirms the broadband enhancement of light absorption in the proposed *** proposed metascreen absorber holds great potential for light harvesting applications and may be leveraged to improve the light conversion efficiency of ultra-thin silicon solar cells,photodetectors,and optical filters.

关键词： silicon learning absorption assisted broadband enhanced films machine metascreen plasmonic

来源：评论

学校读者我要写书评

暂无评论

NeuroSpectrum Deep Learning Architecture for Automated Spectrum Analysis Resource Optimization 3

NeuroSpectrum Deep Learning Architecture for Automated Spect...

引用

3rd International Conference on Integrated Circuits and Communication Systems, ICICACS 2025

作者： Tamilselvan, K. Senthil, T. Buvaneswari, B. Santhiya, S. Soni, Mukesh Avs Engineering College Department of Information Technology Salem India Sreenivasa Institute of Technology and Management Studies Department of Computer Science and Engineering chittoor India Panimalar Engineering College Department of Information Technology Chennai India Excel Engineering College Department of Computer Science and Engineering Namakkal India Lovely Professional University Department of Division of Research and Development Phagwara India

ISBN: (纸本)9798331508456

Reducing the availability of spectrum, as well as the nature of allocating channels in the present wireless telecommunication networks, is one of the critical issues facing modern wireless technology. Such spectrum sensing methods may not be highly flexible or efficient in environments where the radio spectrum undergoes frequent changes. This work proposes an improved deep learning spectrum sensing model based on CNN with reinforcement learning. The proposed system design calls for a pioneering dual mechanism, incorporating both spectrum analysis and a resource allocation learning model. The framework achieved remarkable performance metrics: 96 The spectrum sensing method achieved a 3% occurrence rate, a 94.8% precision rate for resource allocation, and a 92.7% efficiency in dynamic spectrum use. In the present investigation, we have employed it for different topologies and got 45% improvement in spectrum efficiency and 87% less interference than the conventional techniques. The work challenges deep learning-based spectrum sensing as well as management and indicates that they are prepared for the following generation of the wireless communication network. © 2025 IEEE.

关键词： Cognitive radio

来源：评论

学校读者我要写书评

暂无评论

A Comparative Performance Assessment for Prediction of Loan Approval in Financial Sector

引用

Procedia computer science 2025年 258卷 298-307页

作者： Lopamudra Hota Puneet Kumar Jain Arun Kumar Department of Computer Science and Engineering National Institute of Technology Rourkela 769008 India

Loans are the bank’s assets since they generate income in terms of interest to banks. Lending a loan to a customer creates credit and liability for the bank and the customer. The profit and loss of a bank depend on the customer’s ability to pay back the loan or not, i.e., defaulter or not. Therefore, predicting the probability of loan repayment becomes a crucial task. For this purpose, ensemble learning methods have been incorporated extensively, and studies have reported the superiority of these methods over conventional classification methods. This paper provides a comprehensive comparative performance assessment of various ensemble methods for predicting Loan approval in the banking sector. Ensemble algorithms, including bagging, boosting and stacking, are considered with the Neural Network (NN), Decision Tree (DT), Logistic Regression (LR), Naive Bayes (NB), and Support Vector Machine (SVM) as baseline classifiers, which are regarded as benchmarks. The quantitative analysis has been presented in terms of accu- racy (ACC), Receiver Operating characteristic Curve (ROC), Area Under the Curve (AUC), Kolmogorov-Smirnov Statistic (KS), Cohen’s Kappa Score (CKS), and Brier Score (BS). The experimental results affirm that ensemble learning performs better than individual learning. LR outperforms other baseline classifiers, whereas the RF (bagging DT) performs the best among the ensemble approaches, followed by XGB and LightGBM, respectively.

关键词： Loan Approval Machine Learning Logistic Regression XGBoost SVM KNN Decision Tree Random Forest

来源：评论

学校读者我要写书评

暂无评论

Sensing Metal Coil Vibration of Headsets for Eavesdropping on Online Conversations With Out-of-Vocabulary Words Using RFID

引用

IEEE Transactions on Mobile Computing 2025年

作者： Chen, Yunzhong Yu, Jiadi Chen, Yingying Kong, Linghe Zhu, Yanmin Chen, Yichao Shanghai Jiao Tong University Department of Computer Science and Engineering Shanghai China Rutgers University Department of Electrical and Computer Engineering United States

As one of the most essential accessories, headsets have been widely used in common online conversations. The metal coil vibration patterns of headset speakers/microphones have been proven to be highly correlated with the speaker-produced/microphone-received sound. This paper presents an online conversation eavesdropping system, RFSpy, which uses only one RFID tag attached on a headset to alternately sense metal coil vibrations of headset speaker and microphone for eavesdropping on speaker-produced and microphone-received sound. In some accessible scenarios, assuming attackers secretly attach a small, battery-free RFID tag under one ear cushion of an eavesdropped user's headset without being noticed. Meanwhile, RFID readers are camouflaged as decorations placed in/out of rooms to transmit and receive RF signals. When the eavesdropped user talks with other users online through the headset, RFSpy first activates the RFID tag to capture the metal coil vibration patterns of headset speaker and microphone upon RF signals. Then, RFSpy reconstructs sound spectrograms from the RF signal-based vibration patterns for not only trained words but also untrained (i.e., out-of-vocabulary) words utilizing designed SSR network. Finally, RFSpy converts the sound spectrograms to conversation content through sound recognition API. Extensive experiments demonstrate that RFSpy can eavesdrop on online conversations with out-of-vocabulary words effectively. © 2002-2012 IEEE.

关键词： Microphones

来源：评论

学校读者我要写书评

暂无评论

Enhanced Electric Vehicle charging strategy through Graph Convolutional Networks integrated with deep reinforcement learning

引用

International Journal of Information Technology (Singapore) 2025年 1-13页

作者： Ghode, Shilpa Digalwar, Mayuri Computer Science and Engineering Department Indian Institute of Information Technology Maharashtra Nagpur 441108 India Information Technology Department G H Raisoni College of Engineering Hingna Road Maharashtra Nagpur 440016 India

The increasing adoption of Electric Vehicles (EVs) necessitates advanced charging infrastructure. This manuscript presents a novel approach that integrates Graph Convolutional Networks (GCNs) and the Deep Dyna Reinforcement Learning (D2RL) algorithm to optimize EV charging. GCNs are utilized to model charging stations as interconnected nodes, incorporating factors such as connectivity and distance. The D2RL algorithm effectively schedules charging by considering vehicle location, battery status, station availability, and electricity costs. Our approach demonstrates enhanced efficiency over traditional methods, achieving a high infrastructure utilization rate and substantial energy savings across various sample sizes. The proposed model maintains its effectiveness even at larger scales, showcasing its scalability and adaptability. Additionally, it improves critical performance metrics related to energy impact and demand response, contributing to more efficient EV charging while reducing energy costs and minimizing environmental impact. © Bharati Vidyapeeth's Institute of computer Applications and Management 2025.

关键词： Battery life enhancement Charging infrastructure optimization Deep dyna reinforcement learning Electric Vehicles Graph Convolutional 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：