检索结果-内蒙古大学图书馆

3rd International Conference on Smart Generation Computing, Communication and Networking, SMART GENCON 2023

作者： Rampal, Sourav Shreenidhi, H.S. Shrivastava, Manish Singh, Akhilendra Pratap Reena, R. Kumar, Santosh D. Centre of Interdisciplinary Research in Business and Technology Chitkara University Institute of Engineering and Technology Chitkara University Punjab India Computer Science and Engineering Faculty of Engineering and Technology Karnataka Bangalore India Vivekananda Global University Department of Computer Science & Engineering Jaipur India Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh India Prince Shri Venkateshwara Padmavathy Engineering College Department of Computer Science and Engineering Chennai127 India Vishwakarma Institute of Information Technology Department of Artificial Intelligence & Data Science Pune India

ISBN: (纸本)9798350319125

Actual-time statistics to date has updated, increasingly vital as the quantity of virtual information maintains updated growth. Effective choice-making depends on the capability to update access and analyze records. Device up-to-date techniques provide the capacity up-to-date enhance actual-time statistics up-to-date by using offering insights from disparate pieces of information. This paper addresses the potential benefits of using a system with date knowledge and updated, up-to-date strategies to broaden extra clever statistics up to date updated systems. It summarizes the contemporary nation of the art and highlights studies in machine learning and up-to-date imaginative and prescient real-time information up-to-date. It additionally discusses present-day challenges and shows viable instructions for future studies. The studies network up-to-date recall how gadgets are up to date know that updated strategies may be used to facilitate real-time get entry updated the maximum precious pieces of facts. © 2023 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

FED: Fast and Efficient dataset Deduplication Framework with GPU Acceleration

arXiv

引用

arXiv 2025年

作者： Son, Youngjun Kim, Chaewon Lee, Jaejin Graduate school of Data Science Seoul National University Korea Republic of Department of Computer Science Seoul National University Korea Republic of

dataset deduplication plays a crucial role in enhancing data quality, ultimately improving the training performance and efficiency of large language models. A commonly used method for data deduplication is the MinHash LSH algorithm. Recently, NVIDIA introduced a GPU-based MinHash LSH deduplication method, but it remains suboptimal, leaving room for further improvement in processing efficiency. This paper proposes a GPU-accelerated deduplication framework, FED, that optimizes MinHash LSH for GPU clusters and leverages computationally efficient, partially reusable non-cryptographic hash functions. FED significantly outperforms the CPU-based deduplication tool in SlimPajama (using 64 logical CPU cores) by up to 107.2 times and the GPU-based tool in NVIDIA NeMo Curator by up to 6.3 times when processing 30 million documents on a node with four GPUs. Notably, our method dramatically accelerates the previously time-consuming MinHash signature generation phase, achieving speedups of up to 260 compared to the CPU baseline. Despite these gains in efficiency, FED maintains high deduplication quality, with the duplicate document sets reaching a Jaccard similarity of over 0.96 compared to those identified by the standard MinHash algorithm. In large-scale experiments, the deduplication of 1.2 trillion tokens is completed in just 6 hours in a four-node, 16-GPU environment. The related code is publicly available on GitHub (https://***/mcrl/FED). © 2025, CC BY.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

Research challenges of Blockchain in 6G Network

Research challenges of Blockchain in 6G Network

引用

2022 IEEE Delhi Section Conference, DELCON 2022

作者： Sharma, Tripti Prasad, Sanjeev Kumar Sharma, Vedangi Inderprastha Engineering College Department of Computer Science-Data Science Ghaziabad India Galgotias University School of Computer Science and Engineering Greater Noida India Galgotias University School of Computing Science and Engineering U.P. Greater Noida India

ISBN: (纸本)9781665458832

Nowadays, video calling is very much on demand, and it is working well in 4G and 5G networks. It is a time for 3D calling, virtual reality live streaming and holographic communication;Accessing this type of data requires ultrahigh data rates and traffic volumes, so there is a need to speed up the network and increase bandwidth. On the other hand, a system has to maintain reliability, security, availability and flexibility in a wireless network. The network should be extended to the needs of human life such as entertainment, healthcare, smart cities, and transportation expected to enhance service quality with a high-end user experience. To establish this network, the telecommunications infrastructure must upgrade its unmatched service level requirements. The two major disruptive emerging components in this modern Internet-enabled era of technology are both the blockchain and the Internet of Things (IoT). In research and industry, IoT has experienced exponential growth, although it suffers from several limitations, such as poor interoperability, decentralization, privacy, and security vulnerabilities. 6G applications can be based on telecommunications networks and IoT, therefore, the network must ensure strong connectivity between nodes and also ensure the privacy and security of user data. Blockchain is the appropriate solution for these issues. Blockchain is a distributed ledger technology and it enables many industries to exchange data over the Internet between trusted parties without using a central server. The blockchain facilitates verifiable, transparent and secure digital asset transactions with proof of right and ownership. There is a need to identify the role of blockchain in 6G networks. This paper explores the challenges and opportunities related to blockchain for 6G networks, future application possibilities and upcoming research directions. © 2022 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

-Critical Nonuniqueness for the 2D Navier-Stokes Equations

引用

Annals of PDE 2023年第2期9卷 1-56页

作者： Cheskidov, Alexey Luo, Xiaoyutao Department of Mathematics Statistics and Computer Science University of Illinois at Chicago Chicago USA Department of Mathematics Duke University Durham USA

In this paper, we consider the 2D incompressible Navier-Stokes equations on the torus. It is well known that for any $$L^2$$ divergence-free initial data, there exists a global smooth solution that is unique in the class of $$C_t L^2$$ weak solutions. We show that such uniqueness would fail in the class $$C_t L^p$$ if $$ p<2$$ . The non-unique solutions we constructed are almost $$L^2$$ -critical in the sense that (i) they are uniformly continuous in $$L^p$$ for every $$p<2$$ ; (ii) the kinetic energy agrees with any given smooth positive profile except on a set of arbitrarily small measure in time.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Proposed Architecture for Quality based Decision Making Approach for Software Re-engineering 4

A Proposed Architecture for Quality based Decision Making Ap...

引用

4th International Conference on Electronics and Sustainable Communication Systems, ICESC 2023

作者： Govil, Nikhil Sharma, Nidhi Thakral, Deepti Garg, Shalini Dixit, Rinku Chaoudhary, Shailee Lohmor Gla University Department of Cea Mathura India Galgotia College of Engineering and Technology Department of It Greater Noida India Manav Rachna University Department of Computer Science & Technology Faridabad India Mit Art Design and Technology University MH Pune India New Delhi Institute of Management Department of Aiml - Data Science New Delhi India

ISBN: (纸本)9798350300093

Software re-engineering is the modification of a software system after it has been reverse engineered, typically to add new functionality or fix errors. Software re-engineering involves a set of activities aimed at reorganizing a legacy system into a new target system that conforms to quality constraints. This paper explores the reasons behind the failure of software re-engineering. This study has analyzed that some of the most common factors for failure of software re-engineering are misestimated costs and quality inefficiencies. As far as the quality factor is concerned, first the basic tool architecture is analyzed which is used by most of the tools for reverse engineering, reorganization and re-engineering. In this research study, a quality evaluation factor as well as parameters for quality evaluation are added to that particular architecture, so that customers and maintainers can analyze whether they achieve the desired goals for re-engineering. Ifthe target is achieved then the finished product is obtained again with quality and within budget, otherwise more attention should be given to the quality evaluation process. © 2023 IEEE.

关键词： Reverse engineering

来源：评论

学校读者我要写书评

暂无评论

Research on Image Captioning Based on Vision-language Pre-trained Models 9

Research on Image Captioning Based on Vision-language Pre-tr...

引用

9th International Conference on Big data and Information Analytics, BigDIA 2023

作者： Zou, Shiwei Xie, Yuxiang Yan, Jie Wei, Yingmei Luan, Xidao National University of Defense Technology Department of System Engineering Hunan China National University of Defense Technology Key Laboratory of Big Data Department of System Engineering Changsha Hunan China Changsha University Department of Computer Science Hunan China

ISBN: (纸本)9798350330076

Image captioning is a vision-language task that targets at describing an image by generating a coherent sentence automatically. This technology allows computers to understand and describe images like humans, enabling further processing of images. It combines computer vision and natural language processing, which has broad applications. This paper investigates the use of vision-language pre-trained models for image captioning. In this article, an encoder-decoder hybrid model based on PVT and BERT is pre-trained on a large number of image-text pairs, possessing both understanding and generation capabilities. It consists of three components: 1) an image encoder and a text encoder to extract image and text features respectively;2) a multi-modal encoder to align the two modalities;and 3) a text decoder to generate text descriptions. This paper focuses on enhancing the image encoder of the model, which is crucial for image captioning models as it extracts high-resolution image features. Reducing training time and computational costs can also improve the applicability of pre-trained models. Improvements are made to the original model by modifying the ViT model architecture of the image encoding module. A feature pyramid structure and a spatial reduction attention mechanism is introduced to extract high-resolution image features while reducing computational complexity and memory usage, thereby enhancing the universality of the model. The model adopts a pure Transformer architecture, discarding the convolutional structure, and achieves excellent performance. © 2023 IEEE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Deep Knowledge Tracing with Concept Trees 9th

Deep Knowledge Tracing with Concept Trees

引用

19th International Conference on Advanced data Mining and Applications, ADMA 2023

作者： Zhang, Yupei An, Rui Zhang, Wenxin Liu, Shuhui Shang, Xuequn School of Computer Science Northwestern Polytechnical University Xi’an China MIIT Big Data Storage and Management Lab Xi’an China Department of Automation Tsinghua University Beijing China

ISBN: (纸本)9783031466632

Knowledge tracing aims to diagnose the student’s knowledge status and predict the responses to the next questions, which is a critical task in personalized learning. The existing studies consider more academic features, while this paper introduces DKCT, a deep knowledge tracing model with concept trees, to integrate the hierarchical concept tree that describes the structure of concepts in a question. DKCT casts the knowledge concept tree (KCT) in a question from the views of feature, breadth, and difficulty into a KCT representation at first. Then, DKCT is composed of an encoder network with multi-head attention on the question representations and a decoder network with multi-head attention on the interaction embeddings. Finally, DKCT integrates the student embeddings by using fully connected networks to predict the responses to the next questions. Extensive experiments conducted on two real-world educational datasets show that DKCT has a higher prediction accuracy than the currently popular KT models. This work paves the way to consider KCT for knowledge tracing. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Statistical Language Model for Information Retrieval System Implementation in Digital Library

Statistical Language Model for Information Retrieval System ...

引用

International Conference on computer Research and Development (ICCRD)

作者： Muhammad Fikri Hasani Septian Eka Ady Buananta Reynaldi Adidarma Ryu Orlando Tamin Hafizh Ash Shiddiqi Vanessa Santonso Amanda Sugito Computer Science Departement School of Computer Science Bina Nusantara University Jakarta Indonesia Information Systems Department School of Information Systems Bina Nusantara University Jakarta Indonesia Statistics Department School of Computer Science Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798331531881

ISBN: (纸本)9798331531898

A digital library is an application of technology to traditional libraries, providing access to various resources like books, magazines, and newspapers in digital formats. This allows users to access information anytime and anywhere with an internet connection. However, current digital library systems still face challenges, particularly in performing contextual searches for documents. Additionally, the order of search results does not always indicate how relevant the documents are to the user's query. This research explores the use of Statistical Language Models (SLM) to enhance the information retrieval system (IRS) process in digital library applications. SLM uses statistical techniques to analyze the semantic relationships between documents and users, increasing the efficiency and relevance of data collection. This research aims to identify the strengths and weaknesses of SLM, apply it to the IRS application, and evaluate its effectiveness for the application to be developed. These findings contribute to the development of an effective IRS, increasing the efficiency and accessibility of digital library applications and increasing the integration of SLM in IRS in the future.

关键词： Statistical analysis Semantics Information retrieval Transformers Libraries User experience Usability Research and development System implementation Testing

来源：评论

学校读者我要写书评

暂无评论

A Framework based on Deep Neural Network for Ranking-oriented Software Defect Prediction 23

A Framework based on Deep Neural Network for Ranking-oriente...

引用

23rd IEEE International Conference on Software Quality, Reliability, and Security, QRS 2023

作者： Dai, Jiapeng Yang, Xiaoxing Huang, Bingding Lu, Xiaofen Shenzhen Technology University College of Big Data and Internet Shenzhen518118 China Southern University of Science and Technology Guangdong Provincial Key Laboratory of Brain-Inspired Intelligent Computation Department of Computer Science and Engineering Shenzhen518055 China

ISBN: (纸本)9798350319583

Software systems are getting larger and more complex than ever before. In order to improve software reliability, software defect prediction is applied to assist developers in bug discovery. The ranking-oriented software defect prediction aims to rank software modules according to the predicted defect counts. However, existing ranking-oriented defect prediction models are constructed based on traditional hand-crafted features, which might overlook the rich syntactic information buried inside the source codes. In this paper, we propose a universal deep learning-based framework called United Deep Network for ranking-oriented software defect prediction. This framework utilizes deep neural networks to automatically generate features from source code with the syntactic and structural information preserved, and it can combine extracted features with traditional hand-crafted features in order to take advantage of both kinds of features to construct prediction models. Experimental results over 29 sets of data show the good performance of the proposed framework for building ranking-oriented defect prediction models. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Dimensionless machine learning: imposing exact units equivariance

The Journal of Machine Learning Research

引用

The Journal of Machine Learning Research 2023年第1期24卷 4923-4954页

作者： Soledad Villar Weichi Yao David W. Hogg Ben Blum-Smith Bianca Dumitrascu Department of Applied Mathematics and Statistics Mathematical Institute for Data Science Johns Hopkins University Department of Technology Operations and Statistics New York University Center for Cosmology and Particle Physics Department of Physics New York University Max-Planck-Institut für Astronomie Flatiron Institute a Division of the Simons Foundation Department of Applied Mathematics and Statistics Johns Hopkins University Department of Computer Science and Technology Cambridge University

Units equivariance (or units covariance) is the exact symmetry that follows from the requirement that relationships among measured quantities of physics relevance must obey self-consistent dimensional scalings. Here, we express this symmetry in terms of a (non-compact) group action, and we employ dimensional analysis and ideas from equivariant machine learning to provide a methodology for exactly units-equivariant machine learning: For any given learning task, we first construct a dimensionless version of its inputs using classic results from dimensional analysis and then perform inference in the dimensionless space. Our approach can be used to impose units equivariance across a broad range of machine learning methods that are equivariant to rotations and other groups. We discuss the in-sample and out-of-sample prediction accuracy gains one can obtain in contexts like symbolic regression and emulation, where symmetry is important. We illustrate our approach with simple numerical examples involving dynamical systems in physics and ecology.

关键词：

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：