检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Barrett, Noah Sadeghi, Zahra Matwin, Stan Facutly of Computer Science Dalhousie University Halifax Canada Institute for Big Data Analytics Halifax Canada

Self-supervised Learning (SSL) is a machine learning algorithm for pretraining Deep Neural Networks (DNNs) without requiring manually labeled data. The central idea of this learning technique is based on an auxiliary stage aka pretext task in which labeled data are created automatically through data augmentation and exploited for pretraining the DNN. However, the effect of each pretext task is not well studied or compared in the literature. In this paper, we study the contribution of augmentation operators on the performance of self supervised learning algorithms in a constrained settings. We propose an evolutionary search method for optimization of data augmentation pipeline in pretext tasks and measure the impact of augmentation operators in several SOTA SSL algorithms. By encoding different combination of augmentation operators in chromosomes we seek the optimal augmentation policies through an evolutionary optimization mechanism. We further introduce methods for analyzing and explaining the performance of optimized SSL algorithms. Our results indicate that our proposed method can find solutions that outperform the accuracy of classification of SSL algorithms which confirms the influence of augmentation policy choice on the overall performance of SSL algorithms. We also compare optimal SSL solutions found by our evolutionary search mechanism and show the effect of batch size in the pretext task on two visual datasets. Copyright © 2023, The Authors. All rights reserved.

关键词： Evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

A Topological Perspective of Port Networks from Three Years (2017-2019) of AIS data 30

A Topological Perspective of Port Networks from Three Years ...

引用

30th Italian Symposium on Advanced database Systems, SEBD 2022

作者： Carlini, Emanuele De Lira, Vinicius Monteiro Soares, Amilcar Etemad, Mohammad Brandoli, Bruno Matwin, Stan Pisa Italy Department of Computer Science Memorial University of Newfoundland St. John's Canada Institute for Big Data Analytics Dalhousie University Halifax Canada

Complex network analysis is a fundamental tool to understand non-trivial aspects of graphs and networks and is widely used in many fields. In this paper, we apply complex network techniques to study port networks, in which nodes are ports and edges are maritime lines between ports. In particular, we study the temporal evolution of several topological features of a network of ports, including connected components, shortest paths, and clustering coefficients. We built the network with three years of Automatic Identification System data from 2017 to 2019. We highlight several interesting trends and behaviors that differentiate long-range vessels from short-range vessels. © 2022 CEUR-WS. All rights reserved.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

Revisiting primordial magnetic fields through 21-cm physics: Bounds and forecasts

arXiv

引用

arXiv 2024年

作者： Bhaumik, Arko Paul, Debarun Pal, Supratik Physics and Applied Mathematics Unit Indian Statistical Institute 203 B.T. Road Kolkata700108 India Technology Innovation Hub on Data Science Big Data Analytics and Data Curation Indian Statistical Institute 203 B.T. Road Kolkata700108 India

Primordial magnetic fields (PMFs) may significantly influence 21-cm physics via two mechanisms: (i) magnetic heating of the intergalactic medium (IGM) through ambipolar diffusion (AD) and decaying magnetohydrodynamic turbulence (DT), (ii) impact on the star formation rate density (SFRD) through small-scale enhancement of the matter power spectrum. In this analysis, we integrate both of these effects within a unified analytical framework and use it to determine upper bounds on the parameter space of a nearly scale-invariant non-helical PMF in the light of the global 21-cm signal observed by EDGES. Our findings reveal that the joint consideration of both effects furnishes constraints of the order B0 O(10−2) nG on the present-day magnetic field strength, which are considerably tighter compared to earlier analyses. We subsequently explore the prospects of detecting such a magnetized 21-cm power spectrum at the upcoming SKA-Low mission. For the relevant parameters of the PMF (B0 and nB) and the excess radio background (ξ), SNR estimation and Fisher forecast analysis indicate that it may be possible to constrain these three parameters with relative 1σ uncertainties 10% and an associated SNR 10 at SKA-Low. This also leads to possible correlations among these three parameters, thus revealing intriguing trends of interplay among the various physical processes involved. Copyright © 2024, The Authors. All rights reserved.

关键词： Magnetoplasma

来源：评论

学校读者我要写书评

暂无评论

Chronic Kidney Disease Detection: A Comparative study

Chronic Kidney Disease Detection: A Comparative study

引用

data analytics for Business and Industry: Way Towards a Sustainable Economy (ICDABI), International Conference on

作者： Zyad Allam Reem Jamal Fathan Shajrah Sawsan Hilal MSc. in Big Data Science and Analytics College of Science University of Bahrain Department of Mathematics College of Science University of Bahrain

ISBN: (数字)9798350369786

ISBN: (纸本)9798350369793

Chronic Kidney Disease (CKD) is a life-threatening disease that impairs the kidney making it unable to filter blood properly. Early detection and appropriate treatment of CKD are crucial to increase the survival rate of patients. This study provides a comparative analysis for the identification of CKD. Toward this end, the performance of Logistic Regression, Random Forest, K-Nearest Neighbor (KNN), and Support Vector Machines (SVM) classifiers have been compared. The CKD dataset of the UCI machine learning repository was used to train the models. The dataset consists of 400 cases of which 246 (62%) have CKD and 149 (38%) are healthy cases. The dataset has 25 features distributed into 11 numerical features and 14 categorical features, including the response. The data were loaded into R for exploratory data analysis and pre-processing. To solve the issue of data imbalance, the researchers performed oversampling on the dataset. This study revealed that Random Forest outperforms others with an accuracy of 0.992 and 0.993 F1Score after oversampling. Additionally, the study found that oversampling increased the performance of Logistic Regression, Random Forest and SVM models. However, KNN experienced a decrease in performance after resampling, from 0.988 to 0.967 accuracy after oversampling.

关键词： Support vector machines Logistic regression data analysis Accuracy Nearest neighbor methods Chronic kidney disease Vectors

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Community Discovery Algorithm via Reconstructed Graph Neural NEtwork

Unsupervised Community Discovery Algorithm via Reconstructed...

引用

2023 International Conference on Machine Learning and Cybernetics, ICMLC 2023

作者： Wang, Jinghong Wang, Hui Yang, Jiateng Wang, Xizhao College of Computer and Cyber Security Hebei Normal University Shijiazhuang050024 China Hebei Normal University Hebei Key Laboratory of Network and Information Security Shijiazhuang050024 China Hebei Provincial Engineering Research Center for Supply Chain Big Data Analytics & Data Security Hebei Normal University Shijiazhuang050024 China College of Artificial Intelligence Big Data Hebei Polytechnic Institute Shijiazhuang050020 China College of Computer Science and Software Engineering ShenZhen University Shenzhen518060 China

ISBN: (纸本)9798350303780

Community discovery can help discover potential community structures in the network, which is a fundamental and important issue in network science. Graph neural network-based algorithms are receiving increasing attention. However, there is relatively little work on community discovery directly using graph neural networks in reconstruction design. This paper presents an Unsupervised Community Discovery Algorithm based on reconstructed graph neural network (UCDA) to resolve the above issues. To be specific, this paper proposes a framework that learns the representation based on network structure and attribute information. By designing input network features, encoders, decoders, this algorithm uses an unsupervised loss function to train neural networks to extract communities in an integrated way. These designs integrate higher-order modularity information with network features and use variational graph auto-encoders to train data end-to-end under the graph neural network framework. Based on a series of experiments with a wide range of datasets and advanced algorithms, UCDA has achieved superior performance and shows strong competitiveness. Finally, the last section reports the results analysis, t-SNE visualization and hyperparameter analysis, which depicts stable performance and powerful network modularity ability. © 2023 IEEE.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Interactive Visualization for Statistical Modelling through a Shiny App in R

Interactive Visualization for Statistical Modelling through ...

引用

2021 International Conference on data analytics for Business and Industry, ICDABI 2021

作者： Khedr, Ahmed Hilal, Sawsan University of Bahrain Big Data Science and Analytics Sakhir Bahrain University of Bahrain College of Science Sakhir Bahrain

ISBN: (纸本)9781665416566

The importance of analytics and visualization tools has been growing over the last decades to handle big data which steamed from all aspects of life. The focus of this paper was on visualization as a crucial tool in presenting complex raw data and modelling results to provide easy-to-understand actionable information that facilitate decision-making. However, limited research distinguished between 'data visualization' and 'model visualization', which has been clearly made in this paper. Furthermore, this paper aimed to shed light on the importance of interactive visualizations to compliment statistical data modelling using R and Shiny for its advanced capabilities. Specifically, a methodology has been proposed based on a hybrid development lifecycle that adopts the Agile Software Development Lifecycle and the data analytics Lifecycle. Finally, by presenting a case study to model the dynamics of COVID-19, it was found that R and Shiny alongside the proposed hybrid development lifecycle significantly reduced the amount of time required to build visually interactive applications. The reported results highlighted the effectiveness of the adopted approach in assisting and guiding researchers and developers in building interactive applications that leverage big data analytics. © 2021 IEEE.

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

Early Fault Detection in Rolling Element Bearings Using Signal Processing Techniques 21

Early Fault Detection in Rolling Element Bearings Using Sign...

引用

21st IEEE International Conference on Smart Communities: Improving Quality of Life using AI, Robotics and IoT, HONET 2024

作者： Saeed, Ali Akram, M. Usman Khan, Muazzam A. Afsar, M. Belal Quaid-i-Azam University ICESCO Big Data Analytics and Edge Computing Department of Computer Science Islamabad Pakistan National University of Sciences and Technology Department of Computer and Software Engineering Islamabad Pakistan Otto-von-Guericke University Faculty of Electrical Engineering and Information Technology Maddeburg Germany

ISBN: (纸本)9798350378078

Many studies show that bearings are the most vulnerable components in low-voltage motors. While advanced bearing diagnostic systems exist, their cost can be a barrier for non-critical machinery due to the potential wait time to see a return on investment. This research explores the feasibility of using basic signal processing techniques on vibration data for bearing condition monitoring. The primary goal is to analyze benchmark data from the Case Western Reserve University (CWRU) dataset and establish a baseline performance for this data-driven approach. The results suggest that, with appropriate signal processing, it's possible to achieve early detection of bearing faults, leading to more efficient time-To-fault identification with higher detection accuracies ranging from 83 % to 100 % for each method. © 2024 IEEE.

关键词： Signal modulation

来源：评论

学校读者我要写书评

暂无评论

NC-ALG: Graph-Based Active Learning under Noisy Crowd 40

NC-ALG: Graph-Based Active Learning under Noisy Crowd

引用

40th IEEE International Conference on data Engineering, ICDE 2024

作者： Zhang, Wentao Wang, Yexin You, Zhenbang Li, Yang Cao, Gang Yang, Zhi Cui, Bin Center for Machine Learning Research Peking University China Key Lab of High Confidence Software Technologies Peking University China Institute of Advanced Algorithms Research Shanghai China Institute of Computational Social Science Peking University Qingdao China National Engineering Labratory for Big Data Analytics and Applications China TEG Tencent Inc. Department of Data Platform China Beijing Academy of Artificial Intelligence China

ISBN: (纸本)9798350317152

Graph Neural Networks (GNNs) have achieved great success in various data mining tasks but they heavily rely on a large number of annotated nodes, requiring considerable human efforts. Despite the effectiveness of existing GNN-based Active Learning (AL) methods, they assume that the annotated labels are always correct, which is contradictory to the error-prone labeling process in a practical crowdsourcing environment. Besides, due to this impractical assumption, existing works only focus on optimizing the node selection in AL but neglect optimizing the labeling process. Therefore, we present NC-ALG, the first GNN-based AL framework that optimizes both the node selection and node labeling process under a noisy crowd. For node selection, NC-ALG introduces a new measurement to model influence reliability and an effective influence maximization objective to select nodes. For node labeling, NC-ALG significantly reduces the labeling cost by considering the model-predicted labels and the labels of mirror nodes. To the best of our knowledge, this is the first attempt to consider GNN-based AL under the practical noisy crowd. Empirical studies on public datasets demonstrate that NC-ALG significantly outperforms existing methods in terms labeling efficiency. Notably, it only takes NC-ALG one-third of the labeling budget that the competitive baseline GRAIN needs to achieve an accuracy of 70.7 % on PubMed. © 2024 IEEE.

关键词： Crowdsourcing

来源：评论

学校读者我要写书评

暂无评论

JH-Ranker: Enhancing Keigo Recognition in Japanese Sentences through Multi-Task Learning

JH-Ranker: Enhancing Keigo Recognition in Japanese Sentences...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Shih-Wei Guo Ya-Chun Wen Jui-Ling Yu Yao-Chung Fan Department of Computer Science and Engineering National Chung Hsing University Taichang Taiwan Department of Japanese Language and Literature Providence University Taichang Taiwan Department of Data Science and Big Data Analytics Providence University Taichang Taiwan

ISBN: (数字)9798350359312

ISBN: (纸本)9798350359329

In the realm of Japanese language processing, the correct use and understanding of Keigo (honorific language) presents unique challenges due to its complex nature and context-dependent usage. This study introduces JH-Ranker, a novel multitask learning framework designed to enhance the recognition and understanding of Keigo in Japanese sentences. We are employing a multi-task learning approach by integrating Keigo Level, Contextual Field, and the Classification of Keigo Forms. This integration aims to foster a comprehensive understanding of Keigo usage from small data, leveraging the synergy between multiple tasks to achieve a deeper understanding with limited resources. Utilizing the BERT-base-japanese pre-trained model, JH-Ranker takes into account various Keigo Forms and contextual scenarios, going beyond mere syntactic analysis. Our methodology also emphasizes the importance of considering both the form and context of Keigo, addressing challenges such as label imbalance and overfitting in multi-label classification tasks. The experimental results demonstrate a significant improvement in Keigo recognition capabilities over existing baseline models, particularly in discerning the subtle nuances of Keigo Levels and Forms. This study contributes to the realm of natural language processing by presenting a refined and contextually sensitive method for Japanese Keigo. It demonstrates the language model’s adaptability and understanding in dealing with intricate honorific language scenarios.

关键词： Adaptation models Analytical models Neural networks Syntactics Multitasking Natural language processing Context modeling

来源：评论

学校读者我要写书评

暂无评论

Hybrid Machine Learning Algorithms for Polycystic Ovary Syndrome Detection

Hybrid Machine Learning Algorithms for Polycystic Ovary Synd...

引用

2022 International Conference on data analytics for Business and Industry, ICDABI 2022

作者： Alshakrani, Sara Hilal, Sawsan Zeki, Ahmed M. University of Bahrain MSc. in Big Data Science and Analytics College of Science Bahrain University of Bahrain College of Science Department of Mathematics Bahrain University of Bahrain College of Information Technology Department of Information Systems Bahrain

ISBN: (纸本)9781665490580

This study focuses on identifying Polycystic Ovary Syndrome (PCOS), which is a serious medical hormonal disorder condition that affects a woman's ability in childbearing age and causes infertility, diabetes, heart problems, endometrial cancer, and other related health issues as it raises the possibility of long-term problems. Therefore, early detection is of great importance. Taking into account Machine Learning (ML) and ensemble learning algorithms with their superior detection capabilities, especially in the medical field. The Hybrid Random Forest Logistic Regression (HRFLR), combined Extreme Boosting with Random Forest, Linear Support Vector Machine, Light Gradient Boosting Model, and CatBoost model are among the hybrid ML models used in this study. To support the unique techniques, the classification models have been investigated using the PCOS dataset containing physical and clinical parameters of women collected from 10 different hospitals across Kerala, India, and downloaded from the Kaggle repository. The top 14 physical and clinical features were selected to detect PCOS by using the univariate feature selection approach. Follicle No. (R) and Follicle No. (L) were found to be the most significant among all features. The evaluation metrics used to test all the models are Accuracy, Precision, Recall, F1-Score, ROC curve plot, Area Under the Curve Score, and K-fold Cross Validation (CV). Finally, results were discussed and compared to indicate that CatBoost outperforms other models, with an accuracy score of 92% also HRFLR obtained the best CV accuracy with 92% when K = 30, both applied on the top 14 features only using 80:20 data split ratio. Consequently, CatBoost is effective at detecting PCOS patients. © 2022 IEEE.

关键词： Feature Selection

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：