Meta-analytics : consensus approaches and system patterns for data analysis / Steven Simske.

Format:

Book

Author/Creator:

Simske, Steven, author.

Language:

English

Subjects (All):

Data mining.

Physical Description:

1 online resource (342 pages)

Edition:

First edition.

Place of Publication:

Cambridge, Massachusetts : Morgan Kaufmann Publishers, [2019]

System Details:

text file

Summary:

Meta-Analytics: Consensus Approaches and System Patterns for Data Analysis presents an exhaustive set of patterns for data science to use on any machine learning based data analysis task. The book virtually ensures that at least one pattern will lead to better overall system behavior than the use of traditional analytics approaches. The book is ‘meta’ to analytics, covering general analytics in sufficient detail for readers to engage with, and understand, hybrid or meta- approaches. The book has relevance to machine translation, robotics, biological and social sciences, medical and healthcare informatics, economics, business and finance. Inn addition, the analytics within can be applied to predictive algorithms for everyone from police departments to sports analysts. Provides comprehensive and systematic coverage of machine learning-based data analysis tasks Enables rapid progress towards competency in data analysis techniques Gives exhaustive and widely applicable patterns for use by data scientists Covers hybrid or ‘meta’ approaches, along with general analytics Lays out information and practical guidance on data analysis for practitioners working across all sectors

Contents:

Front Cover

Meta-Analytics: Consensus Approaches and System Patterns for Data Analysis

Copyright

Dedication

Contents

Acknowledgments

Chapter 1: Introduction, overview, and applications

1.1. Introduction

1.2. Why is this book important?

1.3. Organization of the book

1.4. Informatics

1.5. Statistics for analytics

1.5.1. Value and variance

1.5.2. Sample and population tests

1.5.3. Regression and estimation

1.6. Algorithms for analytics

1.6.1. k-Means and k-nearest neighbor clustering

1.6.2. Unclustering

1.6.3. Markov models

1.7. Machine learning

1.7.1. Entropy

1.7.2. SVM and kernels

1.7.3. Probability

1.7.4. Dimensionality reduction and information gain

1.7.5. Optimization and search

1.7.6. Data mining and knowledge discovery

1.7.7. Recognition

1.7.8. Ensemble learning

1.8. Artificial intelligence

1.8.1. Genetic algorithms

1.8.2. Neural networks

1.8.3. Immunological algorithms

1.9. A platform for building a classifier from the ground up (binary case)

1.10. A platform for building a classifier from the ground up (general case)

1.10.1. Training and validation

1.10.2. Testing and deployment

1.10.3. Comparing training and testing data set results

1.11. Summary

References

Further reading

Chapter 2: Ground truthing

2.1. Introduction

2.2. Pre-validation

2.3. Optimizing settings from training data

2.4. Learning how to Learn

2.5. Deep learning to deep unlearning

2.6. Summary

Chapter 3: Experimental design

3.1. Introduction

3.2. Data normalization

3.2.1. Simple (unambiguous) normalization

3.2.2. Bias normalization

3.2.3. Normalization and experimental design tables

3.3. Designs for the pruning of aging data

3.4. Systems of systems

3.4.1. Systems

3.4.2. Hybrid systems.

3.4.3. Dynamically-updated systems

3.4.4. Interfaces

3.4.5. Gain

3.4.6. Domain normalization

3.4.7. Sensitivity analysis

3.5. Summary

Chapter 4: Meta-analytic design patterns

4.1. Introduction

4.2. Cumulative response patterns

4.2.1. Identifying zones of interest

4.2.2. Zones of interest for sequence-dependent predictive selection

4.2.3. Traditional cumulative gain curves, or lift curves

4.3. Optimization of analytics

4.3.1. Decision trees

4.3.2. Putative-identity triggered patterns

4.3.3. Expectation-maximization and maximum-minimum patterns

4.4. Model agreement patterns

4.4.1. Hybrid regression

4.4.2. Modeling and model fitting

4.5. Co-occurrence and similarity patterns

4.6. Sensitivity analysis patterns

4.7. Confusion matrix patterns

4.8. Entropy patterns

4.9. Independence pattern

4.10. Functional NLP patterns (macro-feedback)

4.11. Summary

Chapter 5: Sensitivity analysis and big system engineering

5.1. Introduction

5.2. Sensitivity analysis of the data set itself

5.3. Sensitivity analysis of the solution model

5.4. Sensitivity analysis of the individual algorithms

5.5. Sensitivity analysis of the hybrid algorithmics

5.6. Sensitivity analysis of the path to the current state

5.7. Summary

Chapter 6: Multipatch predictive selection

6.1. Introduction

6.2. Predictive selection

6.3. Means of predicting

6.4. Means of selecting

6.5. Multi-path approach

6.6. Applications

6.7. Sensitivity analysis

6.8. Summary

Reference

Chapter 7: Modeling and model fitting

7.1. Introduction

7.2. Chemistry analogues for analytics

7.3. Organic chemistry analogues for analytics

7.4. Immunological and biological analogues for analytics.

7.5. Anonymization analogues for model design and fitting

7.6. LSE, error variance, and entropy: Goodness of fit

7.7. Make mine multiple models!

7.8. Summary

Chapter 8: Synonym-antonym and reinforce-void patterns

8.1. Introduction

8.2. Synonym-antonym patterns

8.3. Reinforce-void patterns

8.4. Broader applicability of these patterns

8.5. Summary

Chapter 9: Analytics around analytics

9.1. Introduction

9.2. Analytics around analytics

9.2.1. Entropy and occurrence vectors

9.2.2. Functional metrics

9.2.3. E-M (expectation-maximization) approaches

9.2.4. System design concerns

9.3. Optimizing settings from training data

9.4. Hybrid methods

9.5. Other areas for investigation around the analytics

9.6. Summary

Chapter 10: System design optimization

10.1. Introduction

10.1.1. System considerations-Revisiting the system gains

10.1.2. System gains-Revisiting and expanding the system biases

10.1.3. Nothing ventured, nothing gained

10.2. Module optimization

10.3. Clustering and regularization

10.3.1. Sum of squares regularization

10.3.2. Variance regularization

10.3.3. Cluster size regularization

10.3.4. Small cluster regularization

10.3.5. Number of clusters regularization

10.3.6. Discussion of regularization methods

10.4. Analytic system optimization

10.5. Summary

Chapter 11: Aleatory and expert system techniques

11.1. Introduction

11.2. Revisiting two earlier aleatory patterns

11.2.1. Sequential removal of features aleatory pattern

11.2.2. Sequential variation of feature output aleatory pattern

11.3. Adding random elements for testing

11.4. Hyperspectral aleatory approaches.

11.5. Other aleatory applications in machine and statistical learning

11.6. Expert system techniques

11.7. Summary

Chapter 12: Application I: Topics and challenges in machine translation, robotics, and biological sciences

12.1. Introduction

12.2. Machine translation

12.3. Robotics

12.4. Biological sciences

12.5. Summary

Chapter 13: Application II: Medical and health-care informatics, economics, business, and finance

13.1. Introduction

13.2. Healthcare

13.3. Economics

13.4. Business and finance

13.5. Summary

13.6. Postscript: Psychology

Chapter 14: Discussion, conclusions, and the future of data

14.1. Chapter 1

14.2. Chapter 2

14.3. Chapter 3

14.4. Chapter 4

14.5. Chapter 5

14.6. Chapter 6

14.7. Chapter 7

14.8. Chapter 8

14.9. Chapter 9

14.10. Chapter 10

14.11. Chapter 11

14.12. Chapter 12

14.13. Chapter 13

14.14. The future of meta-analytics

Index

Back Cover.

Notes:

Includes bibliographical references and index.

Description based on print version record.

ISBN:

9780128146248

0128146249

OCLC:

1104211997