检索结果-内蒙古大学图书馆

4th International Conference on Computer Communication and Artificial Intelligence (CCAI)

作者： Chennareddy, Venkateswarlu Patibandla, Pavan Kumar Koppula, Rama C. Amplitude Inc Director Engn San Francisco CA 94103 USA

ISBN: (纸本)9798350362770;9798350362763

As data volumes continue to surge, the resources required to accelerate data processing for model training remain a challenge, leading to increased costs and extended processing times. This paper presents a novel method that combines Adaptive Sampling with an automated, self-adaptive comprehensive test suite module to address these challenges. This approach maintains model accuracy and ensures coverage of essential data required for business use cases. Experiments conducted on diverse datasets, some as large as several hundred terabytes, demonstrate that this method can reduce processing times by up to 75%. This achievement is realized by efficiently identifying and processing only representative samples. These results demonstrate the promising potential for improving data processing efficiency in model training for Artificial Intelligence (AI) applications across diverse sectors.

关键词： efficient data processing adaptive sampling self-adaptive comprehensive test suite artificial intelligence data efficiency

来源：评论

学校读者我要写书评

暂无评论

efficient and Privacy-Preserving Massive data processing for Smart Grids

引用

IEEE ACCESS 2021年 9卷 70616-70627页

作者： Shen, Hua Zhang, Mingwu Wang, Hao Guo, Fuchun Susilo, Willy Hubei Univ Technol Sch Comp Sci Wuhan 430068 Peoples R China Guilin Univ Elect Technol Guangxi Key Lab Trusted Software Guilin 541004 Peoples R China Shandong Normal Univ Sch Informat Sci & Engn Jinan 250014 Peoples R China Univ Wollongong Inst Cybersecur & Cryptol Sch Comp & Informat Technol Wollongong NSW 2522 Australia

Analysis and utilization of massive meter data can help decision-makers provide reasonable decisions. Therefore, multi-functional meter data processing has received considerable attention in recent years. Nevertheless, it might compromise users' privacy, such as releasing users' lifestyles and habits. In this paper, we propose an efficient and privacy-preserving massive data process for smart grids. The presented protocol utilizes the Paillier homomorphic encryption and Horner's Rule to achieve a privacy-preserving two-level random permutation method, making large-scale meter data permuted randomly and sufficiently in a privacy-preserving way. As a result, the analysis center can simultaneously implement various data processing functions (such as variance, comparing, linear regression analysis), and it does not know the source of data. The security analysis shows that our protocol can realize data confidentiality and data source anonymity. The detailed analyses demonstrate that our protocol is efficient in terms of computational and communication costs. Furthermore, it can support fault tolerance of entity failures and has flexible system scalability.

关键词： Meters Cryptography Encryption data processing Protocols Smart grids data models Privacy preservation efficient data processing random permutation massive meter data smart grids

来源：评论

学校读者我要写书评

暂无评论

BDPS: An efficient Spark-Based Big data processing Scheme for Cloud Fog-IoT Orchestration

引用

INFORMATION 2021年第12期12卷 517页

作者： Hossen, Rakib Whaiduzzaman, Md Uddin, Mohammed Nasir Islam, Md. Jahidul Faruqui, Nuruzzaman Barros, Alistair Sookhak, Mehdi Mahi, Md. Julkar Nayeen Jagannath Univ Dept Comp Sci & Engn Dhaka 1100 Bangladesh Jahangirnagar Univ Inst Informat Technol Dhaka 1342 Bangladesh Queensland Univ Technol Sch Informat Syst Brisbane Qld 4000 Australia Texas A&M Univ Dept Comp Sci Corpus Christ TX 78412 USA

The Internet of Things (IoT) has seen a surge in mobile devices with the market and technical expansion. IoT networks provide end-to-end connectivity while keeping minimal latency. To reduce delays, efficient data delivery schemes are required for dispersed fog-IoT network orchestrations. We use a Spark-based big data processing scheme (BDPS) to accelerate the distributed database (RDD) delay efficient technique in the fogs for a decentralized heterogeneous network architecture to reinforce suitable data allocations via IoTs. We propose BDPS based on Spark-RDD in fog-IoT overlay architecture to address the performance issues across the network orchestration. We evaluate data processing delays from fog-IoT integrated parts using a depth-first-search-based shortest path node finding configuration, which outperforms the existing shortest path algorithms in terms of algorithmic (i.e., depth-first search) efficiency, including the Bellman-Ford (BF) algorithm, Floyd-Warshall (FW) algorithm, Dijkstra algorithm (DA), and Apache Hadoop (AH) algorithm. The BDPS exhibits low latency in packet deliveries as well as low network overhead uplink activity through a map-reduced resilient data distribution mechanism, better than in BF, DA, FW, and AH. The overall BDPS scheme supports efficient data delivery across the fog-IoT orchestration, outperforming faster node execution while proving effective results, compared to DA, BF, FW and AH, respectively.

关键词： efficient data processing depth-first search map reduction in-memory accelerator spark

来源：评论

学校读者我要写书评

暂无评论

Exploring the ferromagnetic semiconducting and transport properties of Na2ReX6 (X = Cl and Br) for spintronic and thermoelectric applications

引用

CHEMICAL PHYSICS LETTERS 2025年 858卷

作者： Mustafa, Ghulam M. Waseem, Muhammad Yasir, M. Ammar Noor, N. A. Mumtaz, Sohail Atiq, Shahid Elhindi, Khalid M. Univ Educ Dept Phys Div Sci & Technol Lahore 54770 Punjab Pakistan RIPHAH Int Univ Dept Phys Campus Lahore Lahore 53700 Pakistan Univ Sargodha Dept Phys Sargodha 40100 Pakistan Kwangwoon Univ Elect & Biol Phys Seoul 01897 South Korea Univ Punjab Ctr Excellence Solid State Phys Lahore Pakistan King Saud Univ Coll Food & Agr Sci Plant Prod Dept POB 2460 Riyadh 11451 Saudi Arabia

This study explores the ferromagnetic semiconducting nature of Na2ReX6 (X = Cl and Br) double perovskites utilizing WIEN2k code. The lattice constant increases from 9.88, and 10.52 & Aring;, and computed elastic constants, Poisson, and Pugh's ratio validate their mechanical stability. In addition, the electronic band structure demonstrates a ferromagnetic semiconducting nature with a large band gap of 3.8 and 3.0 eV, and the value of the magnetic moment is witnessed as 3.0 mu B which is suitable for their spintronic applications. The ZT value is noticed as 0.75 and 0.76 illustrating their potential for thermoelectric applications.

关键词： Spintronics efficient data processing Memory devices Double perovskites Transport properties

来源：评论

学校读者我要写书评

暂无评论

Abnormal Driving Behavior Detection: A Machine and Deep Learning Based Hybrid Model

引用

INTERNATIONAL JOURNAL OF INTELLIGENT TRANSPORTATION SYSTEMS RESEARCH 2025年第1期23卷 568-591页

作者： Uddin, Md. Ashraf Hossain, Nibir Ahamed, Asif Islam, Md Manowarul Khraisat, Ansam Alazab, Ammar Ahamed, Md. Khabir Uddin Talukder, Md. Alamin Crown Inst Higher Educ 116 Pacific Highway Sydney NSW 2062 Australia Jagannath Univ Dept Comp Sci & Engn Dhaka Bangladesh Deakin Univ Sch Informat Technol Waurn Ponds Campus Geelong Australia Torrens Univ Ctr Artificial Intelligence & Optimizat Adelaide Australia Bangamata Sheikh Fojilatunnesa Mujib Sci & Technol Dept Comp Sci & Engn Jamalpur Bangladesh Int Univ Business Agr & Technol Dept Comp Sci & Engn Dhaka Bangladesh

Car accidents remain a leading cause of unintentional fatalities, with many incidents stemming from driver behaviors that impact vehicle control, such as steering, braking, accelerating, and gear shifting. Activities like searching for items, using mobile devices, or listening to the radio can distract drivers visually, audibly, and physically, posing significant risks to road safety. While various methods have been developed to detect such distractions, their effectiveness often falls short in real-world applications. This paper introduces a novel approach that combines machine learning (ML) and deep learning (DL) techniques to identify both safe and risky driving behaviors. Six ML classifiers were evaluated on real-world data to distinguish between driving behaviors such as aggressive, fatigued, and normal driving, with the Random Forest classifier demonstrating superior performance. Additionally, a specialized deep-learning baseline model was developed using ResNet50 and efficientNetB6 to classify driving-related images into distinct categories. The hybrid model integrates ML for analyzing tabular data and DL for image recognition, achieving a classification accuracy of 99.3% on the UAH-Drive dataset. Deep learning experiments further revealed that the Base Model outperformed other models, achieving accuracies of 99.32% on the UAH-Drive dataset and 99.87% on the SFD3 dataset. This research presents a robust hybrid ML-DL framework for detecting abnormal driving behaviors, addressing shortcomings of existing techniques in real-world conditions, and offering valuable insights for improving road safety and reducing accidents.

关键词： Driver behavior efficient data processing Machine learning Classification Deep learning

来源：评论

学校读者我要写书评

暂无评论

Enabling High-Throughput Parallel I/O in Particle-in-Cell Monte Carlo Simulations with openPMD and Darshan I/O Monitoring

Enabling High-Throughput Parallel I/O in Particle-in-Cell Mo...

引用

2024 International Conference on Cluster Computing

作者： Williams, Jeremy J. Medeiros, Daniel Costea, Stefan Tskhakaya, David Poeschel, Franz Widera, Rene Huebl, Axel Klasky, Scott Podhorszki, Norbert Kos, Leon Podolnik, Ales Hromadka, Jakub Narwal, Tapish Steiniger, Klaus Bussmann, Michael Laurel, Erwin Markidis, Stefano KTH Royal Inst Technol Stockholm Sweden Univ Ljubljana LeCAD Ljubljana Slovenia CAS Inst Plasma Phys Prague Czech Republic Helmholtz Zentrum Dresden Rossendorf Dresden Germany Lawrence Berkeley Natl Lab Berkeley CA USA Oak Ridge Natl Lab Oak Ridge TN USA Max Planck Comp & Data Facil Garching Germany

ISBN: (纸本)9798350383461;9798350383454

Large-scale HPC simulations of plasma dynamics in fusion devices require efficient parallel I/O to avoid slowing down the simulation and to enable the post-processing of critical information. Such complex simulations lacking parallel I/O capabilities may encounter performance bottlenecks, hindering their effectiveness in data-intensive computing tasks. In this work, we focus on introducing and enhancing the efficiency of parallel I/O operations in Particle-in-Cell Monte Carlo simulations. We first evaluate the scalability of BIT1, a massivelyparallel electrostatic PIC MC code, determining its initial write throughput capabilities and performance bottlenecks using an HPC I/O performance monitoring tool, Darshan. We design and develop an adaptor to the openPMD I/O interface that allows us to stream PIC particle and field information to I/O using the BP4 backend, aggressively optimized for I/O efficiency, including the highly efficient ADIOS2 interface. Next, we explore advanced optimization techniques such as data compression, aggregation, and Lustre file striping, achieving write throughput improvements while enhancing data storage efficiency. Finally, we analyze the enhanced high-throughput parallel I/O and storage capabilities achieved through the integration of openPMD with rapid metadata extraction in BP4 format. Our study demonstrates that the integration of openPMD and advanced I/O optimizations significantly enhances BIT1's I/O performance and storage capabilities, successfully introducing high throughput parallel I/O and surpassing the capabilities of traditional file I/O.

关键词： openPMD Darshan ADIOS2 Parallel I/O efficient data processing Distributed Storage Large-Scale PIC Simulations

来源：评论

学校读者我要写书评

暂无评论

Improving Detection Efficiency: Optimizing Block Size in the Local Outlier Factor (LOF) Algorithm

Improving Detection Efficiency: Optimizing Block Size in the...

引用

International Joint Conference on Rough Sets (IJCRS)

作者： Horyn, Czeslaw Nowak-Brzezinska, Agnieszka Univ Silesia Inst Comp Sci Bankowa 12 Katowice Poland

ISBN: (纸本)9783031509582;9783031509599

Detecting outliers in data is essential in various fields, such as finance, healthcare, and many other domains with anomalies. Among well-known outlier detection algorithms, Local Outlier Factor (LOF) is widely used for identifying unusual data points. However, the computational time of LOF significantly increases when dealing with large datasets containing numerical and categorical features. We propose an innovative approach using block size optimisation to speed up the outlier detection process while maintaining high accuracy. By optimizing the block size, we achieve a significant improvement in LOF's performance without compromising its effectiveness. Experiment results on diverse datasets containing mixed categorical and numerical features demonstrate the effectiveness of our method in accelerating outlier detection while retaining high detection accuracy. This advancement in outlier detection has the potential to improve decision-making processes. It empowers the timely identification of anomalous events, which is significant in critical applications, including cybersecurity.

关键词： Outlier detection Local Outlier Factor efficient data processing Cybersecurity

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：