Data Mining in Time Series Database(V57).

By: Last, MarkContributor(s): Bunke, Horst | Bunke, HorstSeries: Series in Machine Perception and Artificial Intelligence SerPublisher: Singapore : World Scientific Publishing Co Pte Ltd, 2004Copyright date: ©2004Edition: 57th edDescription: 1 online resource (205 pages)Content type: text Media type: computer Carrier type: online resourceISBN: 9789812565402Subject(s): Data mining.;Distributed databasesGenre/Form: Electronic books. Additional physical formats: Print version:: Data Mining in Time Series Database(V57)DDC classification: 006.332 LOC classification: QA76.9.D343 -- D3834 2004ebOnline resources: Click to View
Contents:
Intro -- Contents -- Preface -- Chapter 1 Segmenting Time Series: A Survey and Novel Approach E. Keogh, S. Chu, D. Hart and M. Pazzani -- 1. Introduction -- 2. Background and Related Work -- 2.1. The Sliding Window Algorithm -- 2.2. The Top-Down Algorithm -- 2.3. The Bottom-Up Algorithm -- 2.4. Feature Comparison of the Major Algorithms -- 3. Empirical Comparison of the Major Segmentation Algorithms -- 3.1. Experimental Methodology -- 3.2. Experimental Results -- 4. A New Approach -- 4.1. The SWAB Segmentation Algorithm -- 4.2. Experimental Validation -- 5. Conclusions and Future Directions -- References -- Chapter 2 A Survey of Recent Methods for Efficient Retrieval of Similar Time Sequences M. L. Hetland -- 1. Introduction -- 1.1. Terminology and Notation -- 2. The Problem -- 2.1. Robust Distance Measures -- 2.2. Good Indexing Methods -- 2.3. Spatial Indices and the Dimensionality Curse -- 3. Signature Based Similarity Search -- 3.1. A Simple Example -- 3.2. Spectral Signatures -- 3.3. Piecewise Constant Approximation -- 3.4. Landmark Methods -- 4. Other Approaches -- 4.1. Using Suffix Trees to Avoid Redundant Computation -- 4.2. Data Reduction through Piecewise Linear Approximation -- 4.3. Search Space Pruning through Subsequence Hashing -- 5. Conclusion -- Appendix Distance Measures -- References -- Chapter 3 Indexing of Compressed Time Series E. Fink and K. B. Pratt -- 1. Introduction -- 2. Previous Work -- 3. Important Points -- 4. Similarity Measures -- 5. Pattern Retrieval -- 6. Concluding Remarks -- Acknowledgements -- References -- Chapter 4 Indexing Time-Series under Conditions of Noise M. Vlachos, D. Gunopulos and G. Das -- 1. Introduction -- 2. Background -- 2.1. Time Series Similarity Measures -- 2.2. Indexing Time Series -- 2.3. Motivation for Non-Metric Distance Functions -- 3. Similarity Measures Based on LCSS.
3.1. Original Notion of LCSS -- Dynamic Programming Solution [11,42] -- 3.2. Extending the LCSS Model -- 3.3. Differences between DTW and LCSS -- 4. Efficient Algorithms to Compute the Similarity -- 4.1. Computing the Similarity Function S1 -- 4.2. Computing the Similarity Function S2 -- 4.3. An Efficient Approximate Algorithm -- 5. Indexing Trajectories for Similarity Retrieval -- 5.1. Indexing Structure -- 5.2. Searching the Index Tree for Nearest Trajectories -- 6. Experimental Evaluation -- 6.1. Time and Accuracy Experiments -- 6.2. Clustering using the Approximation Algorithm -- 6.2.1. Determining the Values for δ and ε -- 6.2.2. Experiment 1 - Video Tracking Data -- 6.2.3. Experiment 2 - Australian Sign Language Dataset (ASL)3 -- 6.2.4. Experiment 3 - ASL with Added Noise -- 6.3. Evaluating the Quality and Efficiency of the Indexing Technique -- 7. Conclusions -- References -- Chapter 5 Change Detection in Classification Models Induced from Time Series Data G. Zeira, O. Maimon, M. Last and L. Rokach -- 1. Introduction -- 2. Change Detection in Classification Models of Data Mining -- 2.1. Classification Model Characteristics -- 2.2. Variety of Changes -- 2.3. Statistical Hypothesis Testing -- 2.4. Methodology -- 2.5. Change Detection Procedure -- 3. Experimental Evaluation -- 3.1. Design of Experiments -- 3.2. Results - Part 1 (Hit Rate and False Alarm Rate) -- 3.3. Results - Part 2 (Time Series Data) -- 4. A Real-World Case Study -- 4.1. Dataset Description -- 5. Conclusions and Future Work -- References -- Chapter 6 Classification and Detection of Abnormal Events in Time Series of Graphs H. Bunke and M. Kraetzl -- 1. Introduction -- 2. Preliminaries -- 3. Analysis of Graph Spectra -- 4. Graph Edit Distance -- 5. Median Graphs -- 6. Median Graphs and Abnormal Change Detection in Sequences of Graphs.
6.1. Median vs. Single Graph, Adjacent in Time (msa) -- 6.2. Median vs. Median Graph, Adjacent in Time (mma) -- 6.3. Median vs. Single Graph, Distant in Time (msd) -- 6.4. Median vs. Median Graph, Distant in Time (mmd) -- 7. Application to Computer Network Monitoring -- 7.1. Problem Description -- 7.2. Experimental Results -- 8. Conclusion -- References -- Chapter 7 Boosting Interval-Based Literals: Variable Length and Early Classification C. J. Alonso Gonzalez and J. J. Rodriguez Diez -- 1. Introduction -- 2. Boosting -- 3. Interval Based Literals -- 3.1. Relative Predicates -- 3.2. Region Based Predicates -- 3.3. Classifier Example -- 4. Variable Length Series -- 5. Early Classification -- 6. Experimental Validation -- 6.1. CBF (Cylinder, Bell and Funnel) -- 6.2. Control Charts -- 6.3. Trace -- 6.4. Auslan -- 7. Conclusions -- References -- Chapter 8 Median Strings: A Review X. Jiang, H. Bunke and J. Csirik -- 1. Introduction -- 2. Median String Problem -- 3. Theoretical Results -- 4. Fast Computation of Set Median Strings -- 4.1. Exact Set Median Search in Metric Spaces -- 4.2. Approximate Set Median Search in Arbitrary Spaces -- 5. Computation of Generalized Median Strings -- 5.1. An Exact Algorithm and Its Variants -- 5.2. Approximate Algorithms -- 5.2.1. Greedy Algorithms -- 5.2.2. Genetic Search -- 5.2.3. Perturbation-Based Iterative Refinement -- 5.3. Dynamic Computation of Generalized Median Strings -- 6. Experimental Evaluation -- 7. Discussions and Conclusion -- Acknowledgments -- References.
Summary: Adding the time dimension to real-world databases produces Time SeriesDatabases (TSDB) and introduces new aspects and difficulties to datamining and knowledge discovery. This book covers the state-of-the-artmethodology for mining time series databases. The novel data miningmethods presented in the book include techniques for efficientsegmentation, indexing, and classification of noisy and dynamic timeseries. A graph-based method for anomaly detection in time series isdescribed and the book also studies the implications of a novel andpotentially useful representation of time series as strings. Theproblem of detecting changes in data mining models that are inducedfrom temporal databases is additionally discussed.
Holdings
Item type Current library Call number Status Date due Barcode Item holds
Ebrary Ebrary Afghanistan
Available EBKAF0008952
Ebrary Ebrary Algeria
Available
Ebrary Ebrary Cyprus
Available
Ebrary Ebrary Egypt
Available
Ebrary Ebrary Libya
Available
Ebrary Ebrary Morocco
Available
Ebrary Ebrary Nepal
Available EBKNP0008952
Ebrary Ebrary Sudan

Access a wide range of magazines and books using Pressreader and Ebook central.

Enjoy your reading, British Council Sudan.

Available
Ebrary Ebrary Tunisia
Available
Total holds: 0

Intro -- Contents -- Preface -- Chapter 1 Segmenting Time Series: A Survey and Novel Approach E. Keogh, S. Chu, D. Hart and M. Pazzani -- 1. Introduction -- 2. Background and Related Work -- 2.1. The Sliding Window Algorithm -- 2.2. The Top-Down Algorithm -- 2.3. The Bottom-Up Algorithm -- 2.4. Feature Comparison of the Major Algorithms -- 3. Empirical Comparison of the Major Segmentation Algorithms -- 3.1. Experimental Methodology -- 3.2. Experimental Results -- 4. A New Approach -- 4.1. The SWAB Segmentation Algorithm -- 4.2. Experimental Validation -- 5. Conclusions and Future Directions -- References -- Chapter 2 A Survey of Recent Methods for Efficient Retrieval of Similar Time Sequences M. L. Hetland -- 1. Introduction -- 1.1. Terminology and Notation -- 2. The Problem -- 2.1. Robust Distance Measures -- 2.2. Good Indexing Methods -- 2.3. Spatial Indices and the Dimensionality Curse -- 3. Signature Based Similarity Search -- 3.1. A Simple Example -- 3.2. Spectral Signatures -- 3.3. Piecewise Constant Approximation -- 3.4. Landmark Methods -- 4. Other Approaches -- 4.1. Using Suffix Trees to Avoid Redundant Computation -- 4.2. Data Reduction through Piecewise Linear Approximation -- 4.3. Search Space Pruning through Subsequence Hashing -- 5. Conclusion -- Appendix Distance Measures -- References -- Chapter 3 Indexing of Compressed Time Series E. Fink and K. B. Pratt -- 1. Introduction -- 2. Previous Work -- 3. Important Points -- 4. Similarity Measures -- 5. Pattern Retrieval -- 6. Concluding Remarks -- Acknowledgements -- References -- Chapter 4 Indexing Time-Series under Conditions of Noise M. Vlachos, D. Gunopulos and G. Das -- 1. Introduction -- 2. Background -- 2.1. Time Series Similarity Measures -- 2.2. Indexing Time Series -- 2.3. Motivation for Non-Metric Distance Functions -- 3. Similarity Measures Based on LCSS.

3.1. Original Notion of LCSS -- Dynamic Programming Solution [11,42] -- 3.2. Extending the LCSS Model -- 3.3. Differences between DTW and LCSS -- 4. Efficient Algorithms to Compute the Similarity -- 4.1. Computing the Similarity Function S1 -- 4.2. Computing the Similarity Function S2 -- 4.3. An Efficient Approximate Algorithm -- 5. Indexing Trajectories for Similarity Retrieval -- 5.1. Indexing Structure -- 5.2. Searching the Index Tree for Nearest Trajectories -- 6. Experimental Evaluation -- 6.1. Time and Accuracy Experiments -- 6.2. Clustering using the Approximation Algorithm -- 6.2.1. Determining the Values for δ and ε -- 6.2.2. Experiment 1 - Video Tracking Data -- 6.2.3. Experiment 2 - Australian Sign Language Dataset (ASL)3 -- 6.2.4. Experiment 3 - ASL with Added Noise -- 6.3. Evaluating the Quality and Efficiency of the Indexing Technique -- 7. Conclusions -- References -- Chapter 5 Change Detection in Classification Models Induced from Time Series Data G. Zeira, O. Maimon, M. Last and L. Rokach -- 1. Introduction -- 2. Change Detection in Classification Models of Data Mining -- 2.1. Classification Model Characteristics -- 2.2. Variety of Changes -- 2.3. Statistical Hypothesis Testing -- 2.4. Methodology -- 2.5. Change Detection Procedure -- 3. Experimental Evaluation -- 3.1. Design of Experiments -- 3.2. Results - Part 1 (Hit Rate and False Alarm Rate) -- 3.3. Results - Part 2 (Time Series Data) -- 4. A Real-World Case Study -- 4.1. Dataset Description -- 5. Conclusions and Future Work -- References -- Chapter 6 Classification and Detection of Abnormal Events in Time Series of Graphs H. Bunke and M. Kraetzl -- 1. Introduction -- 2. Preliminaries -- 3. Analysis of Graph Spectra -- 4. Graph Edit Distance -- 5. Median Graphs -- 6. Median Graphs and Abnormal Change Detection in Sequences of Graphs.

6.1. Median vs. Single Graph, Adjacent in Time (msa) -- 6.2. Median vs. Median Graph, Adjacent in Time (mma) -- 6.3. Median vs. Single Graph, Distant in Time (msd) -- 6.4. Median vs. Median Graph, Distant in Time (mmd) -- 7. Application to Computer Network Monitoring -- 7.1. Problem Description -- 7.2. Experimental Results -- 8. Conclusion -- References -- Chapter 7 Boosting Interval-Based Literals: Variable Length and Early Classification C. J. Alonso Gonzalez and J. J. Rodriguez Diez -- 1. Introduction -- 2. Boosting -- 3. Interval Based Literals -- 3.1. Relative Predicates -- 3.2. Region Based Predicates -- 3.3. Classifier Example -- 4. Variable Length Series -- 5. Early Classification -- 6. Experimental Validation -- 6.1. CBF (Cylinder, Bell and Funnel) -- 6.2. Control Charts -- 6.3. Trace -- 6.4. Auslan -- 7. Conclusions -- References -- Chapter 8 Median Strings: A Review X. Jiang, H. Bunke and J. Csirik -- 1. Introduction -- 2. Median String Problem -- 3. Theoretical Results -- 4. Fast Computation of Set Median Strings -- 4.1. Exact Set Median Search in Metric Spaces -- 4.2. Approximate Set Median Search in Arbitrary Spaces -- 5. Computation of Generalized Median Strings -- 5.1. An Exact Algorithm and Its Variants -- 5.2. Approximate Algorithms -- 5.2.1. Greedy Algorithms -- 5.2.2. Genetic Search -- 5.2.3. Perturbation-Based Iterative Refinement -- 5.3. Dynamic Computation of Generalized Median Strings -- 6. Experimental Evaluation -- 7. Discussions and Conclusion -- Acknowledgments -- References.

Adding the time dimension to real-world databases produces Time SeriesDatabases (TSDB) and introduces new aspects and difficulties to datamining and knowledge discovery. This book covers the state-of-the-artmethodology for mining time series databases. The novel data miningmethods presented in the book include techniques for efficientsegmentation, indexing, and classification of noisy and dynamic timeseries. A graph-based method for anomaly detection in time series isdescribed and the book also studies the implications of a novel andpotentially useful representation of time series as strings. Theproblem of detecting changes in data mining models that are inducedfrom temporal databases is additionally discussed.

Description based on publisher supplied metadata and other sources.

Electronic reproduction. Ann Arbor, Michigan : ProQuest Ebook Central, 2019. Available via World Wide Web. Access may be limited to ProQuest Ebook Central affiliated libraries.

There are no comments on this title.

to post a comment.