检索结果-内蒙古大学图书馆

One-dimensional convolutional neural network with data characterization measurement for cotton yarn quality prediction

引用

CELLULOSE 2023年第6期30卷 4025-4039页

作者： Wang, Menglei Wang, Jingan Gao, Weidong Guo, Mingrui Jiangnan Univ Minist Educ Key Lab Ecotext Wuxi 214122 Peoples R China

Yarn quality prediction is vital in cotton spinning mills for stable production, quality assurance, and cost control. This paper built a machine-learning-based yarn quality prediction method, which contains two core parts: a novel data characterization measurement and a one-dimensional convolutional neural network (YQP-1D-CNN). Firstly, apply a statistical measurement to extract the variable-size feature indicators from the spinning technical parameters, including raw cotton blending scheme, yarn structure parameters, and production process parameters. Then a 1D-CNN prediction model is established based on the prior structural relationships between the feature indicators. Compared with the traditional ANNs, YQP-1D-CNN possesses fewer parameters, requires fewer training samples, and represents better generalization performance. In the experiment, 200 samples are collected from a spinning mill. The prediction performance of the proposed method is compared with six models in existing research. The YQP-1D-CNN shows the highest prediction R-2 and the lowest MSE on the overall performance. The better performance of YQP-1D-CNN on poor sample quality is verified better as well. The ablation experiment proves the advantages of the data characterization measurement and the 1D-CNN structure on yarn quality prediction. Finally, the YQP-1D-CNN has been tested in the spinning mill and met practical production requirements.

关键词： Yarn quality prediction data characterization One-dimensional convolutional neural network Spinning technical parameters Machine learning

来源：评论

学校读者我要写书评

暂无评论

Addressing big data variety using an automated approach for data characterization

引用

JOURNAL OF BIG data 2022年第1期9卷 1-28页

作者： Vranopoulos, Georgios Clarke, Nathan Atkinson, Shirley Univ Plymouth Sch Engn Comp & Math Plymouth Devon England

The creation of new knowledge from manipulating and analysing existing knowledge is one of the primary objectives of any cognitive system. Most of the effort on Big data research has been focussed upon Volume and Velocity, while Variety, "the ugly duckling" of Big data, is often neglected and difficult to solve. A principal challenge with Variety is being able to understand and comprehend the data. This paper proposes and evaluates an automated approach for metadata identification and enrichment in describing Big data. The paper focuses on the use of self-learning systems that will enable automatic compliance of data against regulatory requirements along with the capability of generating valuable and readily usable metadata towards data classification. Two experiments towards data confidentiality and data identification were conducted in evaluating the feasibility of the approach. The focus of the experiments was to confirm that repetitive manual tasks can be automated, thus reducing the focus of a data Scientist on data identification and thereby providing more focus towards the extraction and analysis of the data itself. The origin of the datasets used were Private/Business and Public/Governmental and exhibited diverse characteristics in relation to the number of files and size of the files. The experimental work confirmed that: (a) the use of algorithmic techniques attributed to the substantial decrease in false positives regarding the identification of confidential information;(b) evidence that the use of a fraction of a data set along with statistical analysis and supervised learning is sufficient in identifying the structure of information within it. With this approach, the issues of understanding the nature of data can be mitigated, enabling a greater focus on meaningful interpretation of the heterogeneous data.

关键词： Big data Variety data characterization data origination data Format data confidentiality Delimiter determination Metadata Contextual integrity

来源：评论

学校读者我要写书评

暂无评论

Anomaly detection and optimization using scalable and flexible network data characterization

Global Transitions Proceedings

引用

Global Transitions Proceedings 2021年第1期2卷 67-72页

作者： Purna Chandra Sethi Prafulla Kumar Behera Rama Devi Women's University Bhubaneswar Odisha India Utkal University Vani Vihar Bhubaneswar Odisha India

In the context of information security, cyber-attacks such as malware attacksviruses, etc. caused substantial damage to many organizations. The most common solution to cyber-attack is cryptography algorithm implementation and Intrusion Detection System (IDS). Different cryptography algorithms with varying key length are used to achieve different level of information security. IDS is used to control and alert the cyber-attacks which are much redundant and generates many false alerts in real-life application. The filtering and classification of false alerts involve substantial investment and security analysis. We propose a self-Learning data management approach in terms of data characterization using SeLINA for minimizing false alerts. Two different queues namely: Normal Queue and Suspicious Queues are used for storing information without and with any false alert respectively. The Suspicious Queue then undergoes filtering using feature detection technique. Subsequently, the filtered information with no false alert signals is transferred to the Normal Queue using a supervised hierarchical structure towards minimization of false alerts. Consequently, this technique is helpful to identify malicious attacks easily with no extra time for false alert detection.

关键词： data characterization False intrusion alert Filtering and data management Self learning insightful network analyzer (SeLINA) Supervised analytical hierarchy process

来源：评论

学校读者我要写书评

暂无评论

A new data characterization for selecting clustering algorithms using meta-learning

引用

INFORMATION SCIENCES 2019年 477卷 203-219页

作者： Pimentel, Bruno Almeida de Carvalho, Andre C. P. L. F. Univ Sao Paulo Inst Ciencias Matemat & Comp Sao Carlos SP Brazil

Meta-learning has been successfully used for algorithm recommendation tasks. It uses machine learning to induce meta-models able to predict the best algorithms for a new dataset. In this paper, meta-models are applied to a set of meta-features, describing a dataset, to predict the performance of clustering algorithms applied to this dataset. The paper also proposes a new set of meta-features, based on correlation and dissimilarity measures. Experimental results show that these meta-features improve the recommendation. Additionally, this paper evaluates the importance of each meta-feature for the recommendation. (C) 2018 Elsevier Inc. All rights reserved.

关键词： data characterization Clustering Meta-learning

来源：评论

学校读者我要写书评

暂无评论

Userspace Hypervisor data characterization in Virtualized Environment 24

Userspace Hypervisor Data Characterization in Virtualized En...

引用

24th IEEE International Conference on Parallel and Distributed Systems (ICPADS)

作者： Wang, Bin Vandierendonck, Hans Karakonstantis, Georgios Nikolopoulos, Dimitrios S. Queens Univ Belfast Ctr Data Sci & Scalable Comp Belfast Antrim North Ireland

ISBN: (纸本)9781538673089

Memory-intensive applications have grown rapidly in order to adapt to the changing needs of businesses. Memory errors become even more concerning, since configuration at extended operating points makes hardware more susceptible to system failures. Some data structures may be more sensitive to errors and may cause system crashes more easily than others. A failure in a critical data structure can cause a complete system crash. In this paper, we propose a sensitivity characterization to the hypervisor structures. First, we implement an error injection framework using Syscall ptrace. We profile both static and dynamic data structures through our framework. Then we provide the detailed analysis and data characterization to QEMU hypervisor. Finally, we discuss the checkpointing and selective mechanism in case the QEMU hypervisor crash.

关键词： Reliability data characterization virtualization error injection

来源：评论

学校读者我要写书评

暂无评论

data and Reliability characterization Strategy for Automatic Detection of Valve Stiction in Control Loops

引用

IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY 2017年第3期25卷 769-780页

作者： Garcia, Octavio Pozo Zakharov, Alexey Jamsa-Jounela, Sirkka-Liisa AF Consult Proc Control EE-13516 Tallinn Estonia Aalto Univ Dept Chem Technol FI-00076 Espoo Finland

Automatic detection of process faults requires process expertise to determine the fault symptoms and mathematical algorithms that can classify these symptoms correctly. Thus, the effectiveness of any automatic detection system can degrade because of inconsistencies in the process data and variations in the assumed symptoms of the fault. This paper presents a novel detection strategy based on the data characterization and reliability analysis. In more detail, at first process data are profiled to automatically select and apply the most suitable detection algorithms. Then, the reliability of the detection decisions made by the selected detection algorithms is evaluated, considering a decision unreliable when the process data fit neither the healthy case nor the faulty case assumed by the method. The strategy is applied to a valve stiction detection system. In addition, an exponential fitting method, recognizing three oscillation patterns associated with stiction, is proposed and incorporated into the system. The resulting system was tested on a benchmark data, and the results are discussed.

关键词： Control loops data characterization fault detection oscillation static friction (stiction) valves

来源：评论

学校读者我要写书评

暂无评论

data Heterogeneity, characterization, and Integration in the Context of Autonomous Vehicles 43

Data Heterogeneity, Characterization, and Integration in the...

引用

43rd Annual Conference of the IEEE-Industrial-Electronics-Society (IECON)

作者： Pimentel, Juan R. Kettering Univ Dept Elect & Comp Engn Flint MI 48504 USA

ISBN: (纸本)9781538611272

data is pervasive in autonomous vehicles and in this paper we provide a context to understand its nature, its characterization, and how it can be integrated into a complete autonomous vehicle system. The context is used to characterize data generated from the perception system and to discuss critical issues that arise in the development and implementation of complex autonomous systems such as those featuring the SAE levels 4 and 5 of vehicle driving automation.

关键词： data heterogeneity data characterization data integration perception system perception sensors autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Improving the accuracy and interpretability of multi-scenario building energy consumption prediction considering characteristics of training dataset

引用

ENERGY AND BUILDINGS 2024年 324卷

作者： Fang, Haizhou Tan, Hongwei Yuan, Xiaolei Lin, Xiaojie Zhao, Dafang Kosonen, Risto Tongji Univ Sch Mech Engn 4800 Caoan Rd Shanghai 201804 Peoples R China Tongji Univ Res Ctr Green Bldg & New Energy Shanghai 200092 Peoples R China Aalto Univ Sch Mech Engn Espoo 02150 Finland Zhejiang Univ Coll Energy Engn Hangzhou Peoples R China Osaka Univ Dept Informat Sci & Technol Osaka Japan

This paper proposed an innovative method, which adopts training dataset characteristics, to improve the accuracy and interpretability of multi-scenario building energy consumption prediction. It adopts the pattern recognition, time-spectrograms, and quantitative training data properties (discrete level, daily cycle intensity, and feature similarity), and investigates and analyzes the data from both time and frequency domain perspectives. The prediction predictability and interpretability are enhanced by combining correlation analysis with the input features. The results show that the energy consumptions for both air conditioning (AC) system and the whole building are significantly impacted by the opening rate of AC terminal equipment. The proposed method can achieve a reduction of 68% for the coefficient-of-variation of root mean square error of the total energy consumption and a decrease of 67% for the AC energy consumption. Training data for AC and total energy consumptions show that the discrete level and feature similarity affect prediction performance, with correlation values ranging from 0.4 to 0.6. Moreover, there is a negative correlation (-0.37) between the daily cycle's intensity and the AC energy use, indicating that prediction performance improves with increasing daily cycle intensity. For building energy consumption prediction, the three indicators, namely discreteness, daily cycle intensity, and feature similarity of training data, can be used to effectively analyze the prediction effect of the target task.

关键词： Building energy consumption prediction data characterization Occupancy information Accuracy improvement Interpretability

来源：评论

学校读者我要写书评

暂无评论

Towards consistent data characterization in material flow analysis

引用

MATERIAUX & TECHNIQUES 2014年第5期102卷

作者： Schwab, O. Rechberger, H. Vienna Univ Technol Vienna Austria

The material consumption of societies can be balanced in national resource budgets (NRB). A NRB is the application of Material Flow Analysis (MFA) to defined regions, such as nations (see, for example, the paper by Buchner et al. [ 1]). NRB can support decision-making in resource management in national policies and businesses, i. e. regarding security of resource supply, security of waste disposal, and design of recycling strategies. To a vast degree, the precision of NRB is dependent on the available information base. Most often, this information base consists of cross-disciplinary data, that is, unstructured data from various disciplines that come in different forms and from heterogeneous sources such as official trade statistics, scientific findings and consumer behavior studies. Until now, there is no common understanding about data and information in MFA. In this study we regard data and information from the perspective of a material flow analyst and propose (i) a terminology for communication about MFA input data;and (ii) a database analysis concept to describe and structure MFA databases. This MFA data concept is illustrated in a case study of a national Aluminium budget. The findings indicate that meta-information can help to structure MFA databases, and can contribute to improved system understanding. This study intends to promote the transparent documentation and precise communication of MFA input data and can be the foundation for better data interpretation and comprehensive MFA data quality assessment.

关键词： Material flow analysis national resource budget MFA data database analysis data characterization data quality

来源：评论

学校读者我要写书评

暂无评论

CT-Norm: A Toolkit To Characterize and Harmonize Variability in CT

CT-Norm: A Toolkit To Characterize and Harmonize Variability...

引用

2025 Conference on Medical Imaging

作者： Kulkarni, Kimaya Yadav, Anil Welland, Spencer Harrison Zhang, Tengyue Nael, Kambiz Aberle, Denise Prosper, Ashley Goldin, Jonathan Hoffman, John M. Kim, Grace Hyun J. Brown, Matthew S. McNitt-Gray, Michael F. Hsu, William UCLA David Geffen Sch Med Med & Imaging Informat Grp Dept Radiol Sci Los Angeles CA 90095 USA Univ Calif Los Angeles Samueli Sch Engn Dept Bioengn Los Angeles CA 90095 USA UCLA David Geffen Sch Med Dept Radiol Sci Ctr Comp Vis & Imaging Biomarkers Los Angeles CA 90095 USA UCLA David Geffen Sch Med Dept Radiol Sci Los Angeles CA 90095 USA

ISBN: (纸本)9781510686007;9781510686014

Consistent and reliable data in computed tomography (CT) is vital for developing generalizable AI algorithms for downstream tasks. However, differences in CT acquisition and reconstruction parameters lead to inconsistencies in image quality and characteristics, impacting AI model performance. We introduce CT-Norm, an open-source toolkit addressing CT variability through three primary modules: data characterization, data Harmonization, and Robustness Analysis. Each module targets a distinct phase of the data understanding pipeline. The data characterization module identifies voxel-level and DICOM header-level variability for a given dataset and generates a summary report. The data harmonization module provides a solution to mitigate the identified variability. It offers a library of image-level harmonization methods, ranging from traditional image processing to convolutional neural networks (CNNs) and generative adversarial networks (GANs). This module is validated on an external test set from the TCIA low-dose CT (LDCT) dataset using peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and learned perceptual image patch similarity (LPIPS). Without the harmonization module, the observed PSNR, SSIM, and LPIPS were 19.77 +/- 0.368, 0.33 +/- 0.010, and 0.43 +/- 0.012, respectively, which increased to 26.51 +/- 0.281 and 0.57 +/- 0.011 for PSNR and SSIM, while LPIPS decreased to 0.30 +/- 0.004 after applying CNN-based harmonization. The robustness analysis module allows developers to evaluate the sensitivity of an existing deep-learning model across the identified variability within a dataset. This allows developers to identify the imaging conditions under which a model performs optimally and those where it may struggle, enabling them to optimize the model further and enhance its reliability. CT-Norm enables model developers and end users to understand data variability, harmonize inconsistencies, and optimize AI models for robust and generalizab

关键词： Computed Tomography CT Variability Harmonization data characterization Convolutional Neural Network Generative Adversarial Network

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：