TORUS 1 -- toward an open resource using services : cloud computing for environmental data /

Bibliographic Details
Other Authors: Laffly, Dominique
Format: eBook
Language:English
Published: Hoboken : Wiley, 2020.
Subjects:
Online Access:Connect to the full text of this electronic book
Table of Contents:
  • Cover
  • Half-Title Page
  • Title Page
  • Copyright Page
  • Contents
  • Preface: Why TORUS? Toward an Open Resource Using Services, or How to Bring Environmental Science Closer to Cloud Computing
  • Structure of the book
  • PART 1: Integrated Analysis in Geography: The Way to Cloud Computing
  • Introduction to Part 1
  • Introduction: the landscape as a system
  • 1. Geographical Information and Landscape, Elements of Formalization
  • 2. Sampling Strategies
  • 2.1. References
  • 3. Characterization of the Spatial Structure
  • 4. Thematic Information Structures
  • 5. From the Point to the Surface, How to Link Endogenous and Exogenous Data
  • 5.1. References
  • 6. Big Data in Geography
  • Conclusion to Part 1: Why Here But Not There?
  • PART 2: Basic Mathematical, Statistical and Computational Tools
  • 7. An Introduction to Machine Learning
  • 7.1. Predictive modeling: introduction
  • 7.2. Bayesian modeling
  • 7.2.1. Basic probability theory
  • 7.2.2. Bayes rule
  • 7.2.3. Parameter estimation
  • 7.2.4. Learning Gaussians
  • 7.3. Generative versus discriminative models
  • 7.4. Classification
  • 7.4.1. Naïve Bayes
  • 7.4.2. Support vector machines
  • 7.5. Evaluation metrics for classification evaluation
  • 7.5.1. Confusion matrix-based measures
  • 7.5.2. Area under the ROC curve (AUC)
  • 7.6. Cross-validation and over-fitting
  • 7.7. References
  • 8. Multivariate Data Analysis
  • 8.1. Introduction
  • 8.2. Principal component analysis
  • 8.2.1. How to measure the information
  • 8.2.2. Scalar product and orthogonal variables
  • 8.2.3. Construction of the principal axes
  • 8.2.4. Analysis of the principal axes
  • 8.2.5. Analysis of the data points
  • 8.3. Multiple correspondence analysis
  • 8.3.1. Indicator matrix
  • 8.3.2. Cloud of data points
  • 8.3.3. Cloud of levels
  • 8.3.4. MCA or PCA?
  • 8.4. Clustering
  • 8.4.1. Distance between data points
  • 8.4.2. Dissimilarity criteria between clusters
  • 8.4.3. Variance (inertia) decomposition
  • 8.4.4. k-means method
  • 8.4.5. Agglomerative hierarchical clustering
  • 8.5. References
  • 9. Sensitivity Analysis
  • 9.1. Generalities
  • 9.2. Methods based on linear regression
  • 9.2.1. Presentation
  • 9.2.2. R practice
  • 9.3. Morris' method
  • 9.3.1. Elementary effects method (Morris' method)
  • 9.3.2. R practice
  • 9.4. Methods based on variance analysis
  • 9.4.1. Sobol' indices
  • 9.4.2. Estimation of the Sobol' indices
  • 9.4.3. R practice
  • 9.5. Conclusion
  • 9.6. References
  • 10. Using R for Multivariate Analysis
  • 10.1. Introduction
  • 10.1.1. The dataset
  • 10.1.2. The variables
  • 10.2. Principal component analysis
  • 10.2.1. Eigenvalues
  • 10.2.2. Data points (Individuals)
  • 10.2.3. Supplementary variables
  • 10.2.4. Other representations
  • 10.3. Multiple correspondence analysis
  • 10.4. Clustering
  • 10.4.1. k-means algorithm
  • 10.5. References
  • PART 3: Computer Science
  • 11. High Performance and Distributed Computing
  • 11.1. High performance computing