检索结果-内蒙古大学图书馆

An Initial Study of Machine Learning Underspecification Using Feature Attribution Explainable AI Algorithms: A COVID-19 Virus Transmission Case Study 18th

An Initial Study of Machine Learning Underspecification Usin...

引用

18th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2021

作者： Hinns, James Fan, Xiuyi Liu, Siyuan Raghava Reddy Kovvuri, Veera Yalcin, Mehmet Orcun Roggenbach, Markus Computer Science Department Swansea University Swansea United Kingdom Department of Data Science and Knowledge Engineering Maastricht University Maastricht Netherlands

ISBN: (纸本)9783030891879

From a dataset, one can construct different machine learning (ML) models with different parameters and/or inductive biases. Although these models give similar prediction performances when tested on data that are currently available, they may not generalise equally well on unseen data. The existence of multiple equally performing models exhibits underspecification of the ML pipeline used for producing such models. In this work, we propose identifying underspecification using feature attribution algorithms developed in Explainable AI. Our hypothesis is: by studying the range of explanations produced by ML models, one can identify underspecification. We validate this by computing explanations using the Shapley additive explainer and then measuring statistical correlations between them. We experiment our approach on multiple datasets drawn from the literature, and in a COVID-19 virus transmission case study. © 2021, Springer Nature Switzerland AG.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

A framework to provide customized reuse of open corpus content for adaptive systems 15

A framework to provide customized reuse of open corpus conte...

引用

26th ACM Conference on Hypertext and Social Media, HT 2015

作者： Bayomi, Mostafa CNGL Centre for Global Intelligent Content Knowledge and Data Engineering Group School of Computer Science and Statistics Trinity College Dublin Ireland

ISBN: (纸本)9781450333955

One of the main services that Adaptive Systems offer to their users is the provision of content that is tailored to individual user's needs. Some Adaptive Systems use a closed corpus content that has been prepared for them a priori, hence, they accept only a narrow field of content. Furthermore, the content is tightly coupled with other parts of the system, which also hinders its reusability. To address these limitations, recent systems started to make use of open Web content to provide a wider variety of content. Previous approaches have attempted to harness the information available on the web by providing adaptive systems with customizable information objects. Since adaptive systems are evolving towards the Semantic Web and the use of ontologies, existing systems are limited by their ability to service these documents solely through keyword-based queries. In this research we propose a novel framework that extends existing content provision system, Slicepedia. Our framework uses the conceptual representation of content to segment it in a semantic manner. The framework removes unnecessary content from web pages, such as navigation bars, and then semantically reveals the structural representation of text to build a tree-like hierarchy. This tree can be traversed to obtain different levels of content granularity that facilitate content discoverability and adaptivity. © 2015 ACM.

关键词： Adaptive systems

来源：评论

学校读者我要写书评

暂无评论

Clustering moving objects in spatial networks

引用

12th International Conference on database Systems for Advanced Applications

作者： Chen, Jidong Lai, Caifeng Meng, Xiaofeng Xu, Jianliang Hu, Haibo School of Information Renmin University of China Key Laboratory of Data Engineering and Knowledge Engineering MOE Department of Computer Science Hong Kong Baptist University

ISBN: (纸本)9783540717027

Advances in wireless networks and positioning technologies (e.g., CPS) have enabled new data management applications that monitor moving objects. In such new applications, realtime data analysis such as clustering analysis is becoming one of the most important requirements. In this paper, we present the problem of clustering moving objects in spatial networks and propose a unified framework to address this problem. Due to the innate feature of continuously changing positions of moving objects, the clustering results dynamically change. By exploiting the unique features of road networks, our framework first introduces a notion of cluster block (CB) as the underlying clustering unit. We then divide the clustering process into the continuous maintenance of CBs and periodical construction of clusters with different criteria based on CBs. The algorithms for efficiently maintaining and organizing the CBs to construct clusters are proposed. Extensive experimental results show that our clustering framework achieves high efficiency for clustering moving objects in real road networks.

关键词： spatial-temporal databases moving objects clustering spatial networks

来源：评论

学校读者我要写书评

暂无评论

Distance-based feature selection from probabilistic data

Distance-based feature selection from probabilistic data

引用

14th International Conference on Web-Age Information Management, WAIM 2013

作者： Zhao, Tingting Pei, Bin Zhao, Suyun Chen, Hong Li, Cuiping Key Lab of Data Engineering and Knowledge Engineering Ministry of Education China Department of Computer Science Renmin University of China China

ISBN: (纸本)9783642385612

Feature selection is a powerful tool of dimension reduction from datasets. In the last decade, more and more researchers have paid attentions on feature selection. Further, some researchers begin to focus on feature selection from probabilistic datasets. However, in the existing method of feature selection from probabilistic data, the distance hidden in probabilistic data is neglected. In this paper, we design a new distance measure to select informative feature from probabilistic databases, in which both the distance and randomness in the data are considered. And then, we propose a feature selection algorithm based on the new distance and develop two accelerative algorithms to boost the computation. Furthermore, we introduce a parameter into the distance to reduce the sensitivity to noise. Finally, the experimental results verify the effectiveness of our algorithms. © 2013 Springer-Verlag Berlin Heidelberg.

关键词： Random processes

来源：评论

学校读者我要写书评

暂无评论

A Stacking-Based Deep Neural Network Approach for Effective Network Anomaly Detection

引用

computers, Materials & Continua 2021年第2期66卷 2217-2227页

作者： Lewis Nkenyereye Bayu Adhi Tama Sunghoon Lim Department of Computer and Information Security Sejong UniversitySeoul05006Korea Data Science Group Institute for Basic Science(IBS)Daejeon34126Korea Department of Industrial Engineering Ulsan National Institute of Science and TechnologyUlsan44919Korea

An anomaly-based intrusion detection system(A-IDS)provides a critical aspect in a modern computing infrastructure since new types of attacks can be *** prevalently utilizes several machine learning algorithms(ML)for detecting and classifying network *** date,lots of algorithms have been proposed to improve the detection performance of A-IDS,either using individual or ensemble *** particular,ensemble learners have shown remarkable performance over individual learners in many applications,including in cybersecurity ***,most existing works still suffer from unsatisfactory results due to improper ensemble *** aim of this study is to emphasize the effectiveness of stacking ensemble-based model for A-IDS,where deep learning(e.g.,deep neural network[DNN])is used as base learner *** effectiveness of the proposed model and base DNN model are benchmarked empirically in terms of several performance metrics,i.e.,Matthew’s correlation coefficient,accuracy,and false alarm *** results indicate that the proposed model is superior to the base DNN model as well as other existing ML algorithms found in the literature.

关键词： Anomaly detection deep neural network intrusion detection system stacking ensemble

来源：评论

学校读者我要写书评

暂无评论

Estimating summary quality with pairwise preferences

Estimating summary quality with pairwise preferences

引用

2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, NAACL HLT 2018

作者： Zopf, Markus Research Training Group AIPHES Knowledge Engineering Group Department of Computer Science Hochschulstraße 10 Darmstadt64289 Germany

ISBN: (纸本)9781948087278

Automatic evaluation systems in the field of automatic summarization have been relying on the availability of gold standard summaries for over ten years. Gold standard summaries are expensive to obtain and often require the availability of domain experts to achieve high quality. In this paper, we propose an alternative evaluation approach based on pairwise preferences of sentences. In comparison to gold standard summaries, they are simpler and cheaper to obtain. In our experiments, we show that humans are able to provide useful feedback in the form of pairwise preferences. The new framework performs better than the three most popular versions of ROUGE with less expensive human input. We also show that our framework can reuse already available evaluation data and achieve even better results. © 2018 The Association for Computational Linguistics.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Open framework middleware for intelligent WSN topology adaption in smart buildings

Open framework middleware for intelligent WSN topology adapt...

引用

2009 International Conference on Ultra Modern Telecommunications and Workshops

作者： Brennan, Rob Tai, Wei O'Sullivan, Declan Aslam, Muhammad S. Rea, Susan Pesch, Dirk Knowledge and Data Engineering Group School of Computer Science and Statistics Trinity College Dublin Ireland Centre for Adaptive Wireless Systems Department of Electronic Engineering Cork Institute of Technology Ireland

ISBN: (纸本)9781424439416

Modern smart buildings utilize sensor networks for facilities management applications such as energy monitoring. However as buildings become progressively more embedded with sensor networks, the challenge of managing and maintaining the sensor networks themselves becomes ever more significant. As a cost-sensitive activity, facilities management deployments are less likely to deploy node redundancy and specialized technical staff to maintain these networks. Hence there are strong requirements for the network to efficiently self-diagnose, self-heal and integrate with standard buildings management systems. This paper introduces a solution for WSN management in smart buildings that addresses these issues. It is based on the deployment of the Open Framework Middleware for sensor networks coupled with a structured knowledge and rule-based fault analysis engine to perform network event correlation and root cause analysis. The system also explicitly interfaces with a Building Management System (BMS) or the scheduling of network maintenance activities such as sensor battery replacement. ©2009 IEEE.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Learning analogy-preserving sentence embeddings for answer selection 23

Learning analogy-preserving sentence embeddings for answer s...

引用

23rd Conference on Computational Natural Language Learning, CoNLL 2019

作者： Diallo, Aïssatou Zopf, Markus Fürnkranz, Johannes Research Training Group AIPHES Department of Computer Science Technische Universität Darmstadt Germany Knowledge Engineering Group Department of Computer Science Technische Universität Darmstadt Germany

ISBN: (纸本)9781950737727

Answer selection aims at identifying the correct answer for a given question from a set of potentially correct answers. Contrary to previous works, which typically focus on the semantic similarity between a question and its answer, our hypothesis is that question-answer pairs are often in analogical relation to each other. Using analogical inference as our use case, we propose a framework and a neural network architecture for learning dedicated sentence embeddings that preserve analogical properties in the semantic space. We evaluate the proposed method on benchmark datasets for answer selection and demonstrate that our sentence embeddings indeed capture analogical properties better than conventional embeddings, and that analogy-based question answering outperforms a comparable similarity-based technique. © 2019 Association for Computational Linguistics.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Managing user-centric adaptive services for pervasive computing

Managing user-centric adaptive services for pervasive comput...

引用

International Conference on Autonomic Computing (ICAC)

作者： D. Lewis T. O'Donnell K. Feeney A. Brady V. Wade Computer Science Department Knowledge and Data Engineering Group Dublin Ireland Knowledge and Data Engineering Group Computer Science Department Trinity College Dublin Ireland

Pervasive computing environments need to exhibit highly adaptive behavior to meet the changing task requirements and operational context of visiting mobile users. However this must be balanced with the need of resource owners to meet their goals in administering how users use their resources. This presents challenges of how to manage adaptive systems and how such management should be exercised by people, both average pervasive computing users and administrators of pervasive computing resources. This paper presents some of the issues involved in reconciling dynamic user-centric adaptation with the management of autonomic systems to meet high-level management policies. It discusses our architectural approach and presents some initial research results in addressing these issues.

关键词： Pervasive computing Resource management Programmable control Adaptive control Personal digital assistants data engineering computer science Educational institutions Mobile computing Adaptive systems

来源：评论

学校读者我要写书评

暂无评论

Classification of Awareness in Disorders of Consciousness Using Common Spatial Pattern

Classification of Awareness in Disorders of Consciousness Us...

引用

International Winter Workshop on Brain-computer Interface (BCI)

作者： Chaewon Lee Sunho Lee Jitka Annen Ji-Hoon Jeong Marcello Massimini Silvia Casarotto Melanie Boly Olivier Bodart Aurore Thibaut Mario Rosanova Steven Laureys Olivia Gosseries Minji Lee Dept. Data Science The Catholic University of Korea Bucheon Republic of Korea Dept. Biomedical Software Engineering The Catholic University of Korea Bucheon Republic of Korea Dept. GIGA-Research & Neurology University and University Hospital of Liege Liege Belgium Dept. Data Analysis University of Ghent Ghent Belgium Dept. Computer Science Chungbuk National University Cheongju Republic of Korea Dept. Biomedical and Clinical Sciences “L. Sacco” University of Milan Milan Italy Dept. Psychiatry Wisconsin Institute for Sleep and Consciousness University of Wisconsin Madison USA Coma Science Group GIGA-Consciousness GIGA Research Center University of Liège Liège Belgium Centre du Cerveau Neurology department University Hospital of Liège Liège Belgium

ISBN: (数字)9798331521929

ISBN: (纸本)9798331521936

Advances in intensive care have improved the survival rate of patients with severe acute brain injury, but diagnostic errors for patients with disorders of consciousness are still high. Accurate diagnosis of these patients is very important because effective treatment can vary depending on the diagnosis. In this study, we propose a framework for classifying unresponsive wakefulness syndrome and minimally conscious state, focusing on awareness. In particular, power spectral density and common spatial patterns were used together, considering that spatial information is a key feature in consciousness. The 16 patients with unresponsive wakefulness syndrome and 14 with minimally conscious state underwent resting-state electroencephalography measurements. In addition, we compared the performance by utilizing each frequency (delta, theta, alpha, beta, gamma bands) related to consciousness. As a result, the highest accuracy of 95.06% was achieved by the EEGNet classifier, especially in the beta frequency band. We demonstrated that spatial information is very important in consciousness, as we observed that classification performance improved when common spatial patterns were used. These results provide insight into various frameworks for diagnosing patients with disorders of consciousness and may help patients survive by increasing the diagnosis rate in the future.

关键词： Deep learning Accuracy Biological system modeling Pattern classification Focusing Signal processing Brain modeling Particle measurements Electroencephalography Robustness

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：