Pattern recognition in industry
- "Find it hard to extract and utilise valuable knowledge from the ever-increasing data deluge?" If so, this book will help, as it explores pattern recognition technology and its concomitant role in extracting useful information to build technical and business models to gain competitive in...
Saved in:
| Main Author | |
|---|---|
| Format | eBook Book |
| Language | English |
| Published |
Oxford
Elsevier
2005
Elsevier Science & Technology Elsevier Science |
| Edition | 1 |
| Subjects | |
| Online Access | Get full text |
| ISBN | 9780080445380 0080445381 |
| DOI | 10.1016/B978-0-08-044538-0.X5054-X |
Cover
Table of Contents:
- Front Cover -- Pattern Recognition in Industry -- Copyright Page -- Contents -- Preface -- Acknowledgments -- About the Author -- Part I: Philosophy -- Chapter 1. Introduction -- 1.1. Distinguishing Knowledge and Information from Data -- 1.2. Whence Pattern Recognition Technology -- 1.3. Thermodynamic Concept of Order Leading to Information Theory -- 1.4. Modeling Informed by Observation -- 1.5. Pattern Recognition Technology Triad -- References -- Chapter 2. Patterns Within Data -- 2.1. Types of Data -- 2.2. Characterizing Data -- 2.3. Distance Between Data -- 2.4. Organizing Data-Clustering / Auto-Classification -- 2.5. Organizing Data-Data Series Resonance -- 2.6. Organizing Data-Correlative Modeling -- References -- Chapter 3. Adapting Biological Principles for Deployment in Computational Science -- 3.1. Learning Organisms-An Introduction to Neural Nets -- 3.2. Supervised Learning -- 3.3. Unsupervised Learning -- 3.4. Models that Self-Organize Data (Unsupervised Learning) as well as Correlate them with Dependent Outcomes (Supervised Learning) -- 3.5. Genetic Algorithms -- References -- Chapter 4. Issues in Predictive Empirical Modeling -- 4.1. Pre-Conditioning Data: Pre- and Post-Processing -- 4.2. Detecting Extrapolative Conditions -- 4.3. Embedding Mechanistic Understanding / Experiential Judgment to Enhance Extrapolative Robustness -- 4.4. Insight into Model Behavior -- Part II: Technology -- Chapter 5. Supervised Learning-Correlative Neural Nets -- 5.1. Supervised Learning with Back-Propagation Neural Nets -- 5.2. Feedforward-Exercising the BP Net in Predictive Mode-Neuron Transformation Function -- 5.3. BP Training-Connection Weights Adjusted by the "Delta Rule" to Minimize Learning Errors -- 5.4. Back-Propagation Equations for General Transformation Functions -- 5.5. Back-Propagation Equations for Sigmoidal Transformation Functions
- 5.6. Conjugate Gradient Methodology for Rapid and Robust Convergence -- 5.7. Separating Signal from Noise in Training -- 5.8. Pre-Conditioning Data for BP Nets -- 5.9. Supervised Learning with Radial Basis Function Neural Nets -- 5.10. Seeding the Input Data Space with RBF Cluster Centers -- 5.11. Assigning Spheres of Influence to each Cluster -- 5.12. Activating Clusters from a Point in the Data Space -- 5.13. Developing RBF Correlation Models-Assigning Weights to Map Outcome -- 5.14. Pre-Conditioning Data for RBF Nets -- 5.15. Neural Net Correlation Models -- References -- Chapter 6. Unsupervised Learning: Auto-Clustering and Self-Organizing Data -- 6.1. Unsupervised Learning-Value to Industry -- 6.2. Auto-Clustering Using Radial Basis Functions -- 6.3. RBF Cluster Radius -- 6.4. Competitive Learning -- 6.5. Data Pre-Conditioning for Competitive Learning -- References -- Chapter 7. Customizing for Industrial Strength Applications -- 7.1. Modeling: The Quest for Explaining and Predicting Processes -- 7.2. Combining Empiricism with Mechanistic Understanding -- 7.3. Embedding an Idealized (Partially Correct) Model -- 7.4. Embedding A Priori Understanding in the Form of Constraints -- 7.5. Incorporating Mixed Data Types -- 7.6. Confidence Measure for Characterizing Predictions -- 7.7. Interpreting Trained Neural Net Structures -- 7.8. Graphical Interpretation of Trained Neural Net Structures -- References -- Chapter 8. Characterizing and Classifying Textual Material -- 8.1. Capturing a Document's Essential Features through Fingerprinting -- 8.2. Similar Documents Auto-Classified into Distinct Clusters -- 8.3. Activity Profiles of Authors Provide Competitive Insight -- 8.4. Visualizing a Document's Contents -- 8.5. Identifying Keywords through Entropic Analysis of Text Documents
- 16.3. Genetic Algorithm-Simulation Model Coupling -- 16.4. Results and Conclusion -- Chapter 17. Predicting Distillation Tower Temperatures: Mining Data for Capturing Distinct Operational Variability -- 17.1. Background -- 17.2. Issue -- 17.3. Model Configuration -- 17.4. Identifying Distinctly Different Operating Conditions -- 17.5. Results -- Chapter 18. Enabling New Process Design Based on Laboratory Data -- 18.1. Background -- 18.2. Model Configuration-Bi-Level Focus for "Spot-Lighting" Region of Interest -- 18.3. Model Results -- 18.4. Conclusion -- Chapter 19. Forecasting Price Changes of a Composite Basket of Commodities -- 19.1. Background -- 19.2. Approach and Model Configuration -- 19.3. Model Results -- 19.4. Conclusions -- Chapter 20. Corporate Demographic Trend Analysis -- 20.1. Background -- 20.2. Issues -- 20.3. Approach and Model Configuration -- 20.4. Model Results and Conclusions -- Epilogue -- Appendices -- Appendix A1. Thermodynamics and Information Theory -- A1.1. Thermodynamic Concepts Set the Stage for Quantifying Information -- A1.2. Equilibrium as a State of Disorder-Organization as a Value-Adding Process -- A1.3. Entropy, Disorder, and Uncertainty -- A1.4. Opportunities Found in Imbalances -- A1.5. Appreciation through Quantification -- A1.6. Quantifying Information Transfer -- A1.7. Information Content in a System -- References -- Appendix A2. Modeling -- A2.1. What Are Models -- A2.2. Mechanistic Modeling-General Laws -- A2.3. Particular Laws and Constitutive Relations -- A2.4. Combining General Laws and Constitutive Relations -- A2.5. Modeling Directly from Data -- Reference -- Index
- 8.6. Automation Shrinks Time and Resources Required to Keep up with the World -- References -- Chapter 9. Pattern Recognition in Time Series Analysis -- 9.1. Leading Indices as Drivers -- 9.2. Concept of Resonance in Quantifying Similarities between Time Series -- 9.3. Identifying Leading Indicators -- 9.4. Forecasting -- Reference -- Chapter 10. Genetic Algorithms -- 10.1. Background -- 10.2. Definitions -- 10.3. Setting the Stage -- 10.4. Selection -- 10.5. Mating -- 10.6. Mutation -- 10.7. "Breeding" Fit Solutions -- 10.8. Discovering Profitable Operating Strategies -- 10.9. Product Formulation -- References -- Part III: Case Studies -- Chapter 11. Harnessing the Technology for Profitability -- 11.1. Process Industry Application Modes -- 11.2. Business Applications -- 11.3. Case Studies that Follow -- Chapter 12. Reactor Modeling Through in Situ Adaptive Learning -- 12.1. Background -- 12.2. Reactor Catalyst Deactivation -- 12.3. Model Configuration -- 12.4. In Situ Modeling Scheme -- 12.5. Validation Procedure -- 12.6. Validation Results -- 12.7. Roles Played by Modeling and Plant Operational Teams -- 12.8. Competitive Advantage Derived through this Approach -- Reference -- Chapter 13. Predicting Plant Stack Emissions to Meet Environmental Limits -- 13.1. Background -- 13.2. Reactor Flow and Model Configuration -- 13.3. Model Training and Results -- 13.4. Identifying Optimal Operating Windows for Enhancing Profits -- Chapter 14. Predicting Fouling/Coking in Fired Heaters -- 14.1. Background -- 14.2. Model Configuration -- 14.3. Model Results -- 14.4. Conclusions -- Chapter 15. Predicting Operational Credits -- 15.1. Background -- 15.2. Issues -- 15.3. Model Configuration -- 15.4. Model Results -- 15.5. Plant Follow-Up -- Chapter 16. Pilot Plant Scale-up by Interpreting Tracer Diagnostics -- 16.1. Background -- 16.2. Issue