Anomaly-detection and health-analysis techniques for core router systems. (2020)

• Record Type: Book
• Further Information: Note: Shi Jin, Zhaobo Zhang, Krishnendu Chakrabarty, Xinli Gu.
• Other Names: Jin, Shi; Zhang, Zhaobo; Chakrabarty, Krishnendu; Gu, Xinli
• Contents: Intro -- Preface -- Acknowledgments -- Contents -- 1 Introduction -- 1.1 Network Hierarchy -- 1.2 Overview of Core Router Systems -- 1.3 Prior Work on Anomaly Detection and Health Assessment -- 1.3.1 Techniques Based on Statistical Models -- 1.3.1.1 Parametric Techniques -- 1.3.1.2 Non-parametric Techniques -- 1.3.1.3 Advantages and Disadvantages of Statistical Techniques -- 1.3.2 Methods Based on Clustering -- 1.3.2.1 Advantages and Disadvantages of Clustering Methods -- 1.3.3 Methods Based on Classification -- 1.3.3.1 Advantages and Disadvantages of Classification Methods 1.4 Research Challenges and Motivation -- 1.5 Outline of the Book -- References -- 2 Accurate Anomaly Detection Using Correlation-Based Time-Series Analysis -- 2.1 Motivation -- 2.2 Time-Series-Based Anomaly Detection -- 2.2.1 Distance-Based Anomaly Detection -- 2.2.2 Window-Based Anomaly Detection -- 2.2.3 Prediction-Based Anomaly Detection -- 2.2.3.1 Autoregressive Model -- 2.2.3.2 Support Vector Regression -- 2.2.3.3 Decision Tree -- 2.2.3.4 Artificial Neural Networks -- 2.2.3.5 Recurrent Neural Networks -- 2.2.4 Feature-Categorization-Based Hybrid Anomaly Detection 2.3 Correlation-Based Feature Selection -- 2.3.1 Linear Correlation Analysis Between Features -- 2.3.2 Feature Selection -- 2.3.3 Non-linear Correlation Analysis Between Features -- 2.4 Experimental Results -- 2.4.1 Anomaly Insertion -- 2.4.2 Feature Selection and Categorization -- 2.4.3 Anomaly Detection -- 2.5 Conclusion -- References -- 3 Changepoint-Based Anomaly Detection -- 3.1 Motivation -- 3.2 Framework of Changepoint-Based Anomaly Detection -- 3.3 Changepoint Detection -- 3.3.1 Density-Estimation-Based Method -- 3.3.2 Density-Ratio-Estimation-Based Method -- 3.3.3 Clustering-Based Method 3.3.4 Hybrid Method -- 3.4 Changepoint Window Learning -- 3.5 Anomaly Detection -- 3.6 Experimental Setup -- 3.7 Results of Changepoint Detection -- 3.8 Results of Anomaly Detection -- 3.9 Conclusion -- References -- 4 Hierarchical Symbol-Based Health-Status Analysis -- 4.1 Motivation -- 4.2 Framework of Time-Series-Based Health Analysis -- 4.3 Time Series Symbolization -- 4.3.1 Symbolic Aggregate Approximation -- 4.3.2 1d Symbolic Aggregate Approximation -- 4.3.3 Moving-Average-Based Trend Approximation -- 4.3.4 Non-parametric Shape Approximation -- 4.4 Symbol-Based Pattern Learning 4.4.1 Hierarchical Clustering -- 4.4.2 Rule Discovery -- 4.5 Symbol-Based Classification and Prediction -- 4.5.1 Distance-Based Method -- 4.5.2 Vector-Space-Model-Based Method -- 4.6 Experiments and Results -- 4.6.1 Metrics for Health Analysis -- 4.6.2 Results on Health Analysis -- 4.7 Conclusion -- References -- 5 Self-learning and Efficient Health-Status Analysis -- 5.1 Motivation -- 5.2 Framework of Feature-Based Self-learning Health Analysis -- 5.3 Feature Extraction and Selection -- 5.3.1 Autoencoder-Based Feature Learning -- 5.4 Self-learning for Health Analysis -- 5.5 Experimental Results
• Publisher Details: Cham : Springer
• Publication Date: 2020
• Extent: 1 online resource, illustrations
• Subjects: 004.6/65; Routing (Computer network management); Time-series analysis; Routing (Computer network management); Time-series analysis; Electronic books
• Languages: English
• ISBNs: 9783030336646; 3030336646
• Related ISBNs: 3030336638; 9783030336639
• Notes: Note: Includes bibliographical references and index.
• Access Rights: Legal Deposit; Only available on premises controlled by the deposit library and to one user at any one time; The Legal Deposit Libraries (Non-Print Works) Regulations (UK).
• Access Usage: Restricted: Printing from this resource is governed by The Legal Deposit Libraries (Non-Print Works) Regulations (UK) and UK copyright law currently in force.