检索结果-内蒙古大学图书馆

Semiconductor chip's quality analysis based on its high dimensional test data

ANNALS OF OPERATIONS RESEARCH 2022年第1期311卷 183-194页

作者： Kai, Sun Jin, Wu Chinese Acad Sci Microwave Device & Integrated Circuits Lab Inst Microelect Beijing 611731 Peoples R China

A semiconductor chip usually has thousands test parameters in order to guaranteed its quality. Hence, a batch of chips' test data set include thousands of float data. The primary goal of dealing with this test data is to obtain the fault parameter distribution and judge the chip's quality. It is a challenge due to the large scale and complex relationship of the test data set. This paper presents a novel method to analyze the test data set by meshing the quality theory and scientific data visualization. First, transfer the test data set to a quality classifier matrix Q: a series of quality region is defined based on quality theory, which is the baseline to classify the test data set into different group and mark them with various number. Second, form a quality-spectrum: define a color rule based on the RGB color model and color the quality classifier matrix Q. Hence chip's quality distribution could be observed through the quality-spectrum. Furthermore, by analyzing the quality-spectrum, the chip's quality could be quantitative and fault diagnose has a data basic. One case is included to illustrate appropriateness of the proposed method.

关键词： data processing Quality-spectrum Industrial electronics Quality control

来源：评论

学校读者我要写书评

暂无评论

Gander: the Preliminary Design and Evaluation of an AR+Tablet System for Geospatial analysis

Gander: the Preliminary Design and Evaluation of an AR+Table...

引用

ieee International symposium on Mixed and Augmented Reality Workshops (ISMARW)

作者： Sathaporn Hu Muhammad Raza Derek Reilly Dalhousie University Algoma University

ISBN: (数字)9798331506919

ISBN: (纸本)9798331506926

We explore the combination of headworn augmented reality (AR) displays and handheld tablet devices to support geospatial analysis. In this paper, we present the design of an AR+tablet prototype named Gander. Gander supports the selection of attributes when building a predictive model for some geospatial phenomenon, such as water pollution is correlated to the proximity of large cities, and the comparison of these models. We conducted a walkthrough evaluation with five experts in geospatial analysis and GIS tools. The experts identified strengths and weaknesses in Gander’s design. We propose new design changes such as the pancake plot, rank-based visualization, and reducing steps on the tablet interface.

关键词： visualization Urban areas Buildings Prototypes Predictive models Water pollution Geospatial analysis Augmented reality

来源：评论

学校读者我要写书评

暂无评论

Zoom Transformer for Skeleton-Based Group Activity Recognition

引用

ieee TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2022年第12期32卷 8646-8659页

作者： Zhang, Jiaxu Jia, Yifan Xie, Wei Tu, Zhigang Wuhan Univ State Key Lab Informat Engn Surveying Mapping & Re Wuhan 430079 Peoples R China Wuhan Univ Renmin Hosp Dept Pain Wuhan 430060 Peoples R China Cent China Normal Univ Sch Comp Wuhan 430079 Peoples R China

Skeleton-based human action recognition has attracted increasing attention and many methods have been proposed to boost the performance. However, these methods still confront three main limitations: 1) Focusing on single-person action recognition while neglecting the group activity of multiple people (more than 5 people). In practice, multi-person group activity recognition via skeleton data is also a meaningful problem. 2) Unable to mine high-level semantic information from the skeleton data, such as interactions among multiple people and their positional relationships. 3) Existing datasets used for multi-person group activity recognition are all RGB videos involved, which cannot be directly applied to skeleton-based group activity analysis. To address these issues, we propose a novel Zoom Transformer to exploit both the low-level single-person motion information and the high-level multi-person interaction information in a uniform model structure with carefully designed Relation-aware Maps. Besides, we estimate the multi-person skeletons from the existing real-world video datasets i.e. Kinetics and Volleyball-Activity, and release two new benchmarks to verify the effectiveness of our Zoom Transfromer. Extensive experiments demonstrate that our model can effectively cope with the skeleton-based multi-person group activity. Additionally, experiments on the large-scale NTU-RGB+D dataset validate that our model also achieves remarkable performance for single-person action recognition. The code and the skeleton data are publicly available at https://***/Kebii/Zoom-Transformer

关键词： Skeleton Transformers Feature extraction Activity recognition data mining visualization Task analysis skeleton-based action visual transformer attention mechanism

来源：评论

学校读者我要写书评

暂无评论

On the aggregation functions used in fuzzy signatures based medical image analysis 23

On the aggregation functions used in fuzzy signatures based ...

引用

23rd ieee International symposium on Computational Intelligence and Informatics, CINTI 2023

作者： Nagy, Szilvia Kovacs, Melinda Lilik, Ferenc Koczy, Laszlo T. Széchenyi István University Department of Telecommunications Győr Hungary Széchenyi István University Multidisciplinary PhD School of Engineerineg Győr Hungary Széchenyi István University Győr Hungary Budapest University of Technology and Economics Budapest Hungary

ISBN: (纸本)9798350342949

The paper proposes the use of fuzzy signatures for modeling and analysis of pre-processed medical images, as an example, CT images of the liver are analyzed. Fuzzy signatures are used for the case of distinguishing larger and smaller malignant lesions from each other and from other (benign) nodular diseases in liver computed tomography images. As computed tomography phases are sometimes missing, the treatment of missing data is also briefly addressed. As the size of the malignant lesion influences its manifestation on the images, separate sub-signatures are developed for large and small lesions with the size being a separate layer of the signature. From the medical experts' point of view besides the tree structure of the signature it is crutial to determine the aggregations themselves, which model the ways experts fuse and combine the available information. For the subtrees for small and large lesions in the sub-roots algebraic multiplication seems to be the best fitting t-norm, while in the subtree weighted means. © 2023 ieee.

关键词： Trees (mathematics)

来源：评论

学校读者我要写书评

暂无评论

CUDA data analysis of Small and Medium-Sized Management Platform based on the MultiDimensional Perspective of Information visualization

CUDA Data Analysis of Small and Medium-Sized Management Plat...

引用

2022 International Conference on Edge Computing and Applications, ICECAA 2022

作者： You, Maoqun School of Economics and Management Yuxi Agriculture Vocation-Tecanical College YunNan YuXi653106 China

ISBN: (数字)9781665482325

ISBN: (纸本)9781665482325

The paper focuses on the discussion and analysis of the application of ERP human resource management system in enterprises, clarifying the concept and operation mechanism of human resource management, and then analyzes the problems and application directions of the application of the system in enterprises. Design and implement a set based on CUDA A parallel analysis framework for large-scale HR web log data. The efficiency of a single node in processing log data analysis is compared with that of traditional CPU serial analysis. The research trend is the research on human resource management in the context of big data;the research frontier is the intensity of human resource management, strategic human resource management, andorganizational identity. © 2022 ieee.

关键词： Human resource management

来源：评论

学校读者我要写书评

暂无评论

Multi-Level GNN Preconditioner for Solving large Scale Problems

Multi-Level GNN Preconditioner for Solving Large Scale Probl...

引用

1st International Conference on Smart Energy Systems and Artificial Intelligence (SESAI)

作者： Nastorg, Matthieu Gratien, Jean-Marc Faney, Thibault Bucci, Michele Alessandro Charpiat, Guillaume Schoenauer, Marc Univ Paris Saclay CNRS INRIA LISN F-91405 Orsay France IFP Energies Nouvelles Digital Sci & Technol Direct F-92852 Rueil Malmaison France Safran Tech Digital Sci & Technol Dept F-78114 Chateaufort France

ISBN: (纸本)9798350364613;9798350364606

large-scale numerical simulations often come at the expense of daunting computations. High-Performance Computing has enhanced the process, but adapting legacy codes to leverage parallel GPU computations remains challenging. Meanwhile, Machine Learning models can harness GPU computations effectively but often struggle with generalization and accuracy. Graph Neural Networks (GNNs), in particular, are great for learning from unstructured data like meshes but are often limited to small-scale problems. Moreover, the capabilities of the trained model usually restrict the accuracy of the data -driven solution. To benefit from both worlds, this paper introduces a novel preconditioner integrating a GNN model within a multi-level Domain Decomposition framework. The proposed GNN-based preconditioner is used to enhance the efficiency of a Krylov method, resulting in a hybrid solver that can converge with any desired level of accuracy. The efficiency of the Krylov method greatly benefits from the GNN preconditioner, which is adaptable to meshes of any size and shape, is executed on GPUs, and features a multi-level approach to enforce the scalability of the entire process. Several experiments are conducted to validate the numerical behavior of the hybrid solver, and an in-depth analysis of its performance is proposed to assess its competitiveness against a C++ legacy solver.

关键词： Graph Neural Networks Multi-Level Domain Decomposition Partial Differential Equations Hybrid Solvers

来源：评论

学校读者我要写书评

暂无评论

Design and application of data-driven LESN-DHP controller

Design and application of data-driven LESN-DHP controller

引用

2023 ieee International symposium on Product Compliance Engineering - Asia 2023, ISPCE-AS 2023

作者： Wang, Hongyu Yang, Cuili Yu, Jiamin Wu, Yubao Beijing University of Technology Faculty of Information Technology Beijing Laboratory for Intelligent Environmental Protection Beijing Institute of Artificial Intelligence Beijing100124 China Zhongke High End Chemical Industry Technology Research Institute Co. Ltd Shandong Industry Research Institute China

ISBN: (纸本)9798350361209

Wastewater treatment processes (WWTP) involve a large number of physical, chemical and biological reactions. The concentration of Dissolved oxygen (DO) is an important water quality parameter in WWTP, so it is necessary to control it. To solve this difficult problem, a dual heuristic programming (DHP) controller based on leakage Echo State Network (LESN) is proposed, which is called LESN-DHP controller for short. Firstly, a neural network (NN) identifier based on LESN (LESN-Identifier) is designed to obtain the dynamic characteristics of the system. Then the stability analysis and convergence analysis are carried out. Finally, the feasibility of LESN-DHP controller is verified by simulation experiment. The simulation experiment proves that LESN-DHP controller has better control effect and better performance index than other controllers in WWTP. © 2023 ieee.

关键词： Controllers

来源：评论

学校读者我要写书评

暂无评论

DocEng 2023 - Proceedings of the 2023 ACM symposium on Document Engineering

DocEng 2023 - Proceedings of the 2023 ACM Symposium on Docum...

引用

2023 ACM symposium on Document Engineering, DocEng 2023

ISBN: (纸本)9798400700279

The proceedings contain 28 papers. The topics discussed include: making a difference through data visualization;the evolution and growth of engineering documents for consumer engagement;quality space and time competition on binarizing document images;dynamic topic modelling with tensor decomposition as a tool to explore the legal precedent relevance over time;static pruning for multi-representation dense retrieval;genetic generative information retrieval;improving zero-shot text matching for financial auditing with large language models;automatically inferring the document class of a scientific article;label dependency learning for multilabel text classification;character relationship mapping in major fictional works using text analysis methods;muti-task CTC for joint handwriting recognition & character bounding box prediction;using yolo network for automatic processing of finite automata images with application to bit-strings recognition;layout analysis of historic architectural program documents;algorithm parallelism for improved extractive summarization;and YinYang, a fast and robust adaptive document image binarization for optical character recognition.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Parallel Processing Techniques for Analyzing large Video Files: a Deep Learning Based Approach 20

Parallel Processing Techniques for Analyzing Large Video Fil...

引用

20th ieee Int symposium on Parallel and Distributed Processing with Applicat / 15th ieee Int Conf on Social Comp and Networking / 12th ieee Int Conf on Big data and Cloud Comp / 12th ieee Int Conf on Sustainable Comp and Commun

作者： Syed, Azhar Talha Majtundar, Shikharesh Carleton Univ Dept Syst & Comp Eng Ottawa ON Canada

ISBN: (纸本)9781665464970

Videos are a popular type of media that require analysis to extract the information underlying the data in a timely manner. Often due to the very large size of such data and the involvement of computationally expensive operations, performing the analysis can take a significant amount of time. This paper presents techniques to speed up deep learning-based analysis to perform tasks like tracking objects and filtering video data by applying parallel processing techniques. The proposed approach and techniques leverage parallel processing on two levels: by using GPUs for analyzing individual frames and by distributing the processing load over a fleet of Executor nodes. Experiments with Apache Spark and TensorFlow-based prototypes built for handling various video analysis use cases were conducted on an Amazon EC2 cloud for various combinations of system and workload parameters. Insights into system performance including the reduction in processing time that accrues from applying the proposed parallel processing technique in each scenario are reported in the paper.

关键词： video analysis parallel processing deep learning performance Apache-Spark GPU processing

来源：评论

学校读者我要写书评

暂无评论

ChatGrid: Power Grid visualization Empowered by a large Language Model

ChatGrid: Power Grid Visualization Empowered by a Large Lang...

引用

Energy data visualization (EnergyVis), Workshop on

作者： Sichen Jin Shrirang Abhyankar School of Interactive Computing Georgia Institute of Technology Electricity Infrastructure and Buildings Division Pacific Northwest National Laboratory

ISBN: (数字)9798350379235

ISBN: (纸本)9798350379242

This paper presents a novel open system, ChatGrid, for easy, intuitive, and interactive geospatial visualization of large-scale transmission networks. ChatGrid uses state-of-the-art techniques for geospatial visualization of large networks, including 2.5D map views, animated flows, hierarchical and level-based filtering and aggregation to provide visual information in an easy, cognitive manner. The highlight of ChatGrid is a natural language query based interface powered by a large language model (ChatGPT) that offers a natural and flexible interactive experience whereby users can ask questions and ChatGrid provides responses both in text and visually. This paper discusses the architecture, implementation, design decisions, and usage of large language models for ChatGrid.

关键词： visualization Filtering large language models Conferences Natural languages data visualization Open systems Chatbots Power grids Geospatial analysis

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：